Citation
Export Demand and Promotions of U.S. Beef

Material Information

Title:
Export Demand and Promotions of U.S. Beef Impact of Food Safety and Protectionism
Creator:
Ferrara, Oscar
Place of Publication:
[Gainesville, Fla.]
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (183 p.)

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Food and Resource Economics
Committee Chair:
Ward, Ronald W.
Committee Members:
Adams, Charles M.
Burkhardt, Robert J.
Morris, Jon D.
Graduation Date:
8/9/2008

Subjects

Subjects / Keywords:
Beef ( jstor )
Elasticity of demand ( jstor )
Exports ( jstor )
Imports ( jstor )
International trade ( jstor )
Market demand ( jstor )
Market prices ( jstor )
Market share ( jstor )
Price elasticity ( jstor )
Simulations ( jstor )
Food and Resource Economics -- Dissertations, Academic -- UF
armington, beef, bse, exports, product
Genre:
Electronic Thesis or Dissertation
born-digital ( sobekcm )
Food and Resource Economics thesis, Ph.D.

Notes

Abstract:
Trade disruptive policies have been used for a long time to address a large number of politic, economic, and food safety issues. Policy shocks (i.e., embargoes or export bans) have significant short-and long-term effects on trade relations and production of agricultural commodities, shaping the economies of producing countries as well as the direction and magnitude of trade flows. The consequences of the first case of BSE in North America in December 2003 exemplify how food safety issues significantly affect consumer preferences and producer welfare. After the announcement of the first case of BSE in North America in December 2003, almost immediately 53 countries banned imports from the United States to prevent the disease from entering their countries and to safeguard human health. The resulting negative shift in aggregate market demand for U.S. beef products and the loss of export markets were the immediate consequences of either government-imposed trade restrictions or changes in import demands. Today, U.S. beef exports show an increasing trend thanks in part to trade liberalization policies and information campaigns (i.e., promotions) aimed to expand the presence of U.S. products in major beef importing countries. The objective of this research is to analyze import demands for beef products in selected Asia?s countries using a model of product differentiation by country of origin and to estimate empirically the impact of U.S. beef promotions on re-capturing pre-ban market levels. Based on Armington?s specification (1969), the model for this research assumes a two-stage budgeting allocation and a product's imperfect substitutability between export sources. Hence, this study presumes that beef produced in the United States presents unique attributes (i.e., superior quality and taste) and that foreign consumers can perceive this difference from competing beef sources. Departing from Armington's Constant Elasticity of Substitution (CES) specification, this work proposes the use of a Constant Ratio Elasticity of Substitution (CRES) that allows elaticities to vary proportionally while the substitutability is not necessarly identical. This research includes eight countries divided into four exporting and four importing countries respectively. Based on countries? import/export yearly data, the entire dataset includes almost 2000 observations with information recorded over time across selected countries during an 11-year period (January 1995 to December 2006). Resulting product demand, market share, and elasticity estimates determine the degree of substitution among competing beef products and provide the coefficients to measure the economic impact of trade barriers and effectiveness of market access and promotion policies in four major markets for U.S. beef: Japan, Republic of Korea, Hong Kong, and Taiwan. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (Ph.D.)--University of Florida, 2008.
Local:
Adviser: Ward, Ronald W.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-08-31
Statement of Responsibility:
by Oscar Ferrara.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Ferrara, Oscar. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
8/31/2010
Classification:
LD1780 2008 ( lcc )

Downloads

This item has the following downloads:

ferrara_o.pdf

ferrara_o_Page_111.txt

ferrara_o_Page_162.txt

ferrara_o_Page_113.txt

ferrara_o_Page_039.txt

ferrara_o_Page_176.txt

ferrara_o_Page_077.txt

ferrara_o_Page_121.txt

ferrara_o_Page_129.txt

ferrara_o_Page_166.txt

ferrara_o_Page_092.txt

ferrara_o_Page_164.txt

ferrara_o_Page_056.txt

ferrara_o_Page_007.txt

ferrara_o_Page_157.txt

ferrara_o_Page_009.txt

ferrara_o_Page_050.txt

ferrara_o_Page_069.txt

ferrara_o_Page_163.txt

ferrara_o_Page_169.txt

ferrara_o_Page_132.txt

ferrara_o_Page_088.txt

ferrara_o_Page_183.txt

ferrara_o_Page_041.txt

ferrara_o_Page_083.txt

ferrara_o_Page_017.txt

ferrara_o_Page_091.txt

ferrara_o_Page_084.txt

ferrara_o_Page_019.txt

ferrara_o_Page_004.txt

ferrara_o_Page_122.txt

ferrara_o_Page_065.txt

ferrara_o_Page_138.txt

ferrara_o_Page_124.txt

ferrara_o_Page_022.txt

ferrara_o_Page_082.txt

ferrara_o_Page_034.txt

ferrara_o_Page_105.txt

ferrara_o_Page_174.txt

ferrara_o_Page_016.txt

ferrara_o_Page_151.txt

ferrara_o_Page_135.txt

ferrara_o_Page_058.txt

ferrara_o_Page_102.txt

ferrara_o_Page_159.txt

ferrara_o_Page_052.txt

ferrara_o_Page_140.txt

ferrara_o_Page_110.txt

EJGUG4W85_348XAN_xml.txt

ferrara_o_Page_093.txt

ferrara_o_Page_165.txt

ferrara_o_Page_090.txt

ferrara_o_Page_180.txt

ferrara_o_Page_109.txt

ferrara_o_Page_112.txt

ferrara_o_Page_072.txt

ferrara_o_Page_018.txt

ferrara_o_Page_101.txt

ferrara_o_Page_128.txt

ferrara_o_Page_094.txt

ferrara_o_Page_005.txt

ferrara_o_Page_043.txt

ferrara_o_Page_013.txt

ferrara_o_Page_026.txt

ferrara_o_Page_032.txt

ferrara_o_Page_171.txt

ferrara_o_Page_175.txt

ferrara_o_Page_146.txt

ferrara_o_Page_049.txt

ferrara_o_Page_054.txt

ferrara_o_Page_167.txt

ferrara_o_Page_172.txt

ferrara_o_Page_015.txt

ferrara_o_Page_036.txt

ferrara_o_Page_033.txt

ferrara_o_Page_037.txt

ferrara_o_Page_156.txt

ferrara_o_Page_064.txt

ferrara_o_Page_154.txt

ferrara_o_Page_108.txt

ferrara_o_Page_060.txt

ferrara_o_Page_038.txt

ferrara_o_Page_173.txt

ferrara_o_Page_107.txt

ferrara_o_Page_001.txt

ferrara_o_Page_148.txt

ferrara_o_Page_106.txt

ferrara_o_Page_177.txt

ferrara_o_Page_063.txt

ferrara_o_Page_153.txt

ferrara_o_Page_061.txt

ferrara_o_Page_143.txt

ferrara_o_Page_031.txt

ferrara_o_Page_008.txt

ferrara_o_Page_053.txt

ferrara_o_Page_125.txt

ferrara_o_Page_029.txt

ferrara_o_Page_020.txt

ferrara_o_Page_014.txt

ferrara_o_Page_117.txt

ferrara_o_Page_096.txt

ferrara_o_Page_040.txt

ferrara_o_Page_137.txt

ferrara_o_Page_130.txt

ferrara_o_Page_114.txt

ferrara_o_Page_003.txt

ferrara_o_Page_134.txt

ferrara_o_Page_116.txt

ferrara_o_Page_086.txt

ferrara_o_Page_100.txt

ferrara_o_Page_051.txt

ferrara_o_Page_035.txt

ferrara_o_Page_028.txt

ferrara_o_Page_046.txt

ferrara_o_Page_012.txt

ferrara_o_Page_030.txt

ferrara_o_Page_182.txt

ferrara_o_Page_076.txt

ferrara_o_Page_098.txt

ferrara_o_Page_071.txt

ferrara_o_Page_066.txt

ferrara_o_Page_133.txt

ferrara_o_Page_147.txt

ferrara_o_Page_139.txt

ferrara_o_Page_155.txt

ferrara_o_Page_087.txt

ferrara_o_Page_131.txt

ferrara_o_Page_048.txt

ferrara_o_Page_179.txt

ferrara_o_Page_097.txt

ferrara_o_Page_115.txt

ferrara_o_Page_150.txt

ferrara_o_Page_178.txt

ferrara_o_Page_021.txt

ferrara_o_Page_011.txt

ferrara_o_Page_181.txt

ferrara_o_Page_024.txt

ferrara_o_Page_080.txt

ferrara_o_Page_059.txt

ferrara_o_Page_047.txt

ferrara_o_Page_075.txt

ferrara_o_Page_168.txt

ferrara_o_Page_103.txt

ferrara_o_Page_006.txt

ferrara_o_Page_068.txt

ferrara_o_Page_118.txt

ferrara_o_Page_081.txt

ferrara_o_Page_152.txt

ferrara_o_Page_120.txt

ferrara_o_Page_149.txt

ferrara_o_Page_095.txt

ferrara_o_Page_142.txt

ferrara_o_Page_127.txt

ferrara_o_Page_145.txt

ferrara_o_Page_073.txt

ferrara_o_Page_062.txt

ferrara_o_Page_074.txt

ferrara_o_Page_104.txt

ferrara_o_Page_045.txt

ferrara_o_Page_067.txt

ferrara_o_Page_161.txt

ferrara_o_Page_027.txt

ferrara_o_Page_085.txt

ferrara_o_Page_123.txt

ferrara_o_Page_158.txt

ferrara_o_Page_002.txt

ferrara_o_Page_025.txt

ferrara_o_Page_078.txt

ferrara_o_Page_023.txt

ferrara_o_Page_136.txt

ferrara_o_Page_055.txt

ferrara_o_Page_079.txt

ferrara_o_Page_170.txt

ferrara_o_Page_042.txt

ferrara_o_Page_010.txt

ferrara_o_Page_070.txt

ferrara_o_Page_099.txt

ferrara_o_Page_126.txt

ferrara_o_Page_141.txt

ferrara_o_pdf.txt

ferrara_o_Page_160.txt

ferrara_o_Page_144.txt

ferrara_o_Page_089.txt

ferrara_o_Page_119.txt

ferrara_o_Page_044.txt

ferrara_o_Page_057.txt


Full Text





A model which recognizes that commodities may be imperfect substitutes was developed

by Paul Armington in 1969 to analyze the world trade demand for machinery and chemical

products using Japan and France as differentiated sources of production. Moving away from the

assumption of perfect substitution, Armington' s model differentiates between products within

goods representing a group of products such as machinery, meats, or appliances. Within a group

category, products are not considered perfect substitutes; they are differentiated based on their

region or country of production, but due to similarities in their core attributes, they belong to the

same category. Consequently, this model suggests that one country's beef market is likely to be

composed of demands for beef products originated in several distinct and independent regions.

The strength of these individual product demand functions is largely determined by the size of

the market in the demanding country and the size of each country's beef market is affected by

many variables with the most important including national incomes, average beef prices in the

region, population levels, and the prices of substitute goods.

Armington proposed that in order to determine the overall demand for any product in any

particular trade area and calculate market shares, some fundamental assumptions must be

considered. First, the demand for products within each good' s market is independent of those

demands for products in other goods markets (e.g., U.S. beef and Australian beef). Thus in any

exporter nation, each industry produces only one product and this product is different from the

product of any other country (i.e., separability assumption) (Lloyd and Zhang 2006). Second, the

model assumes that as long as the prices of products within a category remain constant relative to

each other, the shares of those products will not be affected by changes in the size of that market.

This assumption implies that changes in prices of competing products affect product demands

only indirectly through their influence on the total market size and that market shares are not









CHAPTER 4
EMPIRIRICAL TRADE MODEL FOR BEEF PRODUCTS

Introduction

In this chapter the conceptual framework is to estimate the demand for fresh/chilled and

frozen beef products in South Korea, Japan, Taiwan, and Hong Kong from the United States,

Australia, New Zealand and the rest of the world. Specifically, this study estimates the impacts

of economic variables (prices and expenditures) and non-economic variables (animal disease

outbreaks and beef promotions) on product demands and market shares for U.S. beef products

compared with beef from other sources. Using an Armington specification, the model suggested

for this study is based on the premise that changes in trade flows can be distinguished not only

by their kind but also according to their origin or place of production.

Beef Trade Schematic Representation

Before turning to the specific nature of the substitutability among beef products, it is useful

to employ a generalized trade system that will be adopted for the present study. Accordingly, the

schematic representation in Figure 4. 1 displays the causal relationships between prices and

quantities of beef products (fresh and frozen), two market regions (supply and demand markets)

consisting of a set of four exporting countries and four importing countries respectively, which is

set forth for a one-way trade. Figure 4. 1 shows the United States, Australia, New Zealand, and

the Rest of the World as suppliers of beef products (X.i) with prices (P4). The total supply of

beef products is constrained to equal total demand in the importing markets for beef products

(Xi.), which in this trade system is formed by South Korea, Japan, Taiwan, and Hong Kong. The

average price paid for beef in a given importing country is (Pig) and is defined in Equation 4.7.

Thus, the obj ective of this model is to obtain product demands and market shares in each

importing market for beef products differentiated by the country of origin.



































To my wife, Jaquelina; and my children, Sofia, Nicolas, and Marco.











STARTING THE BEEF TRADE SIMULATOR
SIM #8 RELATIVE PRICE ADJUSTMENTS IN TAWAIN
THIS SHOWS EACH EXPORTERS SHARE TO TAWAIN AS RELATIVE PRICES CHANGE


SET SIMNUM = 8;
ZRESETZ; SET SIMVAR = 1; SET
DO ADJ WRP = .90 TO 1.10 BY .025;
SET SIMVAR = SIMVAR+1;
SET SDI3=1; SET SDX1=1; ZBSIMZ;
ENDDO; SET H= SIMVAR;


SDI3=1; SET SDX1=1; ZBSIMZ; ? US


ZRESETZ; SET SIMVAR = 1;
DO ADJ WRP = .90 TO 1.10 BY .025;
SET SIMVAR = SIMVAR+1;
SET SDI3=1; SET SDX2=1; ZB
ENDDO; SET H= SIMVAR;

ZRESETZ; SET SIMVAR = 1;
DO ADJ WRP = .90 TO 1.10 BY .025;
SET SIMVAR = SIMVAR+1;
SET SDI3=1; SET SDX3=1; ZB
ENDDO; SET H= SIMVAR;


SET SDI3=1; SET SDX2=1; ZBSIMZ;


SIMZ;


SET SDI3=1; SET SDX3=1; ZBSIMZ;


? AUSTRALIA





? NEW ZEALAND;


STARTING THE BEEF TRADE SIMULATOR
SIM #9 RELATIVE PRICE ADJUSTMENTS IN HONG KONG
THIS SHOWS EACH EXPORTERS SHARE TO HONG KONG AS RELATIVE PRICES CHANGE


SET SIMNUM = 9;
ZRESETZ; SET SIMVAR = 1;
DO ADJ WRP = .90 TO 1.10 BY .025;
SET SIMVAR = SIMVAR+1;
SET SDX1=1; ZBSIMZ;
ENDDO; SET H= SIMVAR;

ZRESETZ; SET SIMVAR = 1;
DO ADJ WRP = .90 TO 1.10 BY .025;
SET SIMVAR = SIMVAR+1;
SET SDX2=1; ZBSIMZ;
ENDDO; SET H= SIMVAR;

ZRESETZ; SET SIMVAR = 1;
DO ADJ WRP = .90 TO 1.10 BY .025;
SET SIMVAR = SIMVAR+1;
SET SDX3=1; ZBSIMZ;
ENDDO; SET H= SIMVAR;


SET SDX1=1; ZBSIMZ;


? US


SET SDX2=1; ZBSIMZ;





SET SDX3=1; ZBSIMZ;


? AUSTRALIA





? NEW ZEALAND;


STARTING THE BEEF TRADE SIMULATOR
SIM #10 RELATIVE TOTAL DEMAND IS ADJUSTMENTS IN KOREAN


SET SIMNUM = 10;
ZRESETZ; SET SIMVAR = 1;
DO ADJWQALL = .80 TO 1.20 BY .05;
SET SIMVAR = SIMVAR+1;


SET SDIl=1; SET SDX1=1; ZBSIMZ; ? US









The loss of confidence in the safety of beef products along the entire marketing chain and

elevated levels of general concern for food safety have deeply damaged the position of American

beef products in the Japanese market. U.S. beef exports to Japan have been limited to less than

8% of the 2003 monthly shipments since the market re-opened on July, 2006 (ERS 2007). In

2007, the United States accounted only for 6% of the market and Australia, the big winner,

accounted for 82% of the import market while New Zealand and the ROW accounted for 8% and

4% respectively. Figure 2-16 also illustrates the changes in consumption within the Japanese

market. Shortage of beef supply, unfavorable media reports, together with an unsatisfied

Japanese preference for feedlot-type beef produced in the U. S. accounted for the 18.5% overall

decline in Japanese beef consumption during the period between 2003 and 2007.

Taiwan is mainly a pork-meat-consuming country and one of Asia' s largest pork

producers. Therefore, it dependence on imported beef as major source of red meat is relatively

small if compared to other Asian nations. This characteristic explains the reduced impact of the

BSE announcements on consumption levels and consumers' confidence towards imported beef

products.

As shown in Figure 2-17 consumption levels dropped a little more than one kilogram per

capital between 2003 and 2007. Taiwan opened its market for U.S. beef products in February

2006, and by the end of that year total exports of U. S. beef to Taiwan increased 176% in volume

and 142% in value over the previous year (USMEF 2007), which clearly shows the acceptance of

American beef among beef consumers in this country. Before the BSE scare, market conditions

were relatively stable and Australia accounted for more than 37.4% of the beef market in

Taiwan, followed by New Zealand, the United States, and the ROW with 31.3%, 25.3%, and

6. 1% of the import market share respectively. At the end of 2007, the United States maintained a











1000 41T
New Zealand Frlozen Beef Tolume
-Korea Japan- Taiwan -Hong K~ong


1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
Hong Kon 291 3 33 3 04 278 2 02 176 217 377 3 44 329 3 13
Tanvan 3 7 4 24 3 87 3 54 2 58 2 24 2 76 4 81 4 38 4 19 3 99
Japan 14 74 16 88 15 42 14 08 10 26 8 91 10 99 19 13 17 45 16 68 15 88
SKorea 3 56 4 71 3 91 3 24 168 126 19,4 6 1 5 05 46 4 15

Year


Figure 6-6. Estimated frozen New Zealand beef product demands across time in selected Asian
markets .

In all figures, it is very clear that as the United States started to re-gain market access in the


region during the year 2007, while Australian and New Zealand product demands started to

decline in all four markets. According to the estimates of the model, the demand for U. S. beef


products has its largest increase in Japan, 33% (fresh), and 32% (frozen), while its lowest

increase was in South Korea with only 8% (fresh) and 8% (frozen) with respect to the 2003


levels. Meanwhile, New Zealand product demands showed the largest decline in 2007 with a


range between 37% (South Korea) and 17% (all other markets) in the case of fresh products and

32% and 17% in the case of frozen products. Australia, however, did not show such a large


decline during 2007 and in both cases the decline was between 8% (South Korea) and 3.5% (all


other markets).


25



20










5



O










ZRESETZ; SET SIMVAR =0+1; SET SDX3=1; ZBSIMZ; ? NEW ZEALAND TO HONG KONG;
DOT(INDEX=H,CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10
SDYR11;
ZRESETZ; SET SIMVAR = H+1; SET .X=1; SET SDX3=1; ZBSIMZ;
ENDDOT; SET HH= SIMVAR;


STARTING THE BEEF TRADE SIMULATOR
SIM #6 RELATIVE PRICE ADJUSTMENTS IN KOREA
THIS SHOWS EACH EXPORTERS SHARE TO KOREA AS RELATIVE PRICES CHANGE


SET SIMNUM = 6;
ZRESETZ; SET SIMVAR = 1;
DO ADJ WRP = .90 TO 1.10 BY .025;
SET SIMVAR = SIMVAR+1;
SET SDIl=1; SET SDX1=1; ZB2
ENDDO; SET H= SIMVAR;


SET SDIl=1; SET SDX1=1; ZBSIMZ;


? US


ZRESETZ; SET SIMVAR = 1;
DO ADJ WRP = .90 TO 1.10 BY .025;
SET SIMVAR = SIMVAR+1;
SET SDIl=1; SET SDX2=1; ZB
ENDDO; SET H= SIMVAR;

ZRESETZ; SET SIMVAR = 1;
DO ADJ WRP = .90 TO 1.10 BY .025;
SET SIMVAR = SIMVAR+1;
SET SDIl=1; SET SDX3=1; ZB
ENDDO; SET H= SIMVAR;


SET SDIl=1; SET SDX2=1; ZBSIMZ;


SIMZ;


SET SDIl=1; SET SDX3=1; ZBSIMZ;


? AUSTRALIA





? NEW ZEALAND;


STARTING THE BEEF TRADE SIMULATOR
SIM #7 RELATIVE PRICE ADJUSTMENTS IN JAPAN
THIS SHOWS EACH EXPORTERS SHARE TO JAPAN AS RELATIVE PRICES CHANGE


SET SIMNUM = 7;
ZRESETZ; SET SIMVAR = 1; SET
DO ADJ WRP = .90 TO 1.10 BY .025;
SET SIMVAR = SIMVAR+1;
SET SDI2=1; SET SDX1=1; ZBSIMZ;
ENDDO; SET H= SIMVAR;

ZRESETZ; SET SIMVAR = 1; SET SI
DO ADJ WRP = .90 TO 1.10 BY .025;
SET SIMVAR = SIMVAR+1;
SET SDI2=1; SET SDX2=1; ZBSIMZ;
ENDDO; SET H= SIMVAR;

ZRESETZ; SET SIMVAR = 1; SET S
DO ADJ WRP = .90 TO 1.10 BY .025;
SET SIMVAR = SIMVAR+1;
SET SDI2= 1; SET SDX3= 1; ZB SIMZ;
ENDDO; SET H= SIMVAR;


SDI2=1; SET SDX1=1; ZBSIMZ; ? US





DI2=1; SET SDX2=1; ZBSIMZ; ? AUSTRALIA





DI2=1; SET SDX3=1; ZBSIMZ; ? NEW ZEALAND;













-.11 Unie States








0.00 0.002


Austrli



-- 0.599 70.575




-.00S 0.010


New Zaland




0.2590.215
- .201 .2 ,.
0.005 0.009


Parameter Value (%)


-AID IAIJ IBIJ


1.20
1.00

0.80

0.60

0.40
0.20

0.00
1.00

0.80

0.60

0.40

0.20

0.00
1.00

0.80

0.60

0.40

0.20

0.00


2002

Year


2003


2004 2005 2006 2007


1998 1999 2000


Figure 5-3. Taiwan and Hong Kong product demand parameters for beef imports (US-AU-NZ).


The BSE largest impact is, undoubtedly, on U. S. beef products showing a decline in


market participation of 57% across markets between the years 2003 and 2005. Overall, the Pii


trend shows a reduction of 83% in U. S. beef market participation between 1997 and 2007.


Figures 5-2 and 5-3 also show the other side of the BSE scare on these Asian Markets. In the


case of beef products from Australia and New Zealand, the overall upward trend on the Pii


coefficients is very clear, showing an increase of almost 38% in the case of Australia and 15% in


the case of New Zealand between 1997 and 2007. In particular, after the BSE announcements the


increment on the Pii trend was almost 22% and 14% respectively. This upward trend reflects, in









Carwell estimates that for every $500 million spent each year on U.S. agricultural promotion

programs, the average cost to farmers is U.S. $1000.

The Armington Trade Model

Using economic models to evaluate changes in agricultural trade policy generally requires

the conversion of policy changes into price effects. The Armington model uses these price shifts

to determine how policy is expected to affect output, employment, trade flows, economic

welfare, and other variables of interest (Jung 2004). The direction and magnitude of a trade

policy change on individual variables depends on the size of the shock as well as the behavioral

relationships present in the economy. When evaluating policy shifts within an economic model,

these behavioral relationships largely take the form of elasticities reflecting price responsiveness

of one set of variables to a change in a second set (McDaniel and Balisteri 2002).

Prices of goods produced in different countries do not typically move together. This

behavior was first pointed out by Armington (1969). Ever since, it has become a standard

practice among empirical trade researchers to treat goods produced in different countries

differently and to assume a constant elasticity of substitution among them. Such elasticity for

example, the elasticity of substitution between the basket of U. S. goods and that of Australian

goods is referred to as the Armington elasticity (Lloyd and Zhang 2006). As a result, a key

relationship for model analysis is the degree of substitution between goods according to their

origin. In general, knowledge of elasticities is important for policy considerations. Changes in

tariffs and taxes will affect a country' s trade opportunities, level of income, and employment.

The size of these impacts will largely depend on the magnitude of elasticities.

In 1990, Duffy, Wohlgenat, and Richardson studied the price elasticities of export demand

for U. S. cotton and estimated feedback effects of U. S. prices on competing countries. The usual

Armington framework was extended to account for finite elasticities of export supply from U.S.









Chapter 4, Equation 4-18, product demands include a set of product and market differentiation

parameters, domestic demands (Xi.), and a price ratio between average domestic prices (Pi.) and

import prices (Pg y).

This set of parameters is considered the base of the model and defines the elasticities of

demand among products competing in a specific market. Therefore, using (4-3 8) and the

corresponding dummy variables specified in Equations 4-35 through 4-37, the model calculates

their coefficients, including those for beef promotions and type of beef, and their corresponding

T-values and supportive statistics as illustrated in Table 5-1. The t-values are compared at 1.96

for a two-tail 95% confidence level, which means that any value greater than |1.96| will show

evidence against the null hypothesis that the parameter is equal to zero, and hence conclude that

the parameter is statistically significant at the 95% confidence level.

Since these are estimates from logistic and exponential regressions, one cannot

immediately infer that the coefficients indicate a dimensional change in the dependent variable.

Conclusions in this regard are presented later in the simulation chapter. In general, the statistics

of the model correspond to what one would expect given the annual data. The Durbin Watson

statistic is close to two, suggesting little or no serial correlation in the data. With respect to the R2

value, it is reasonably high for the type of model (highly non-linear regression) and data (panel)

and implies that about 76% of the variation in the (log of) product demands parameters is

explained by the (log of) relative sub-parameters, type of beef, and commodity promotions.

Therefore, from a purely statistical viewpoint, the estimated regression fits the data quite well

(Greene 2003).

Table 5-1 illustrates three sections representing the parameter estimates for the ai., aii

and figy components of the CRES function applied to the Xii product demand Equation 4-17.



























































3 See Appendix A


Pi. = C;(Pil Xii)/Xi. (4-7)

At this junction it is very important to reiterate that prices included in the data set collected

for this study are in fact product prices (Pii) in U. S. dollars for each beef product in each of the

four markets and that these prices already include all tariff leved by the importing country.

In the case of a model with a large set of products being traded, most of the equations

already presented would not have practical solutions due to the large number of parameters in the

model. A way to simplify this equation is to introduce the assumptions that (a) import demand is

separable among import sources, (b) elasticities of substitution between all pairs of products

within a group and within a market are constant (i.e., Constant Elasticity of Substitution CES),

and (c) the elasticity of substitution between any two products competing in a market is the same

as that between any other pair of products competing in the same market (Armington 1969).

Equation 4-8 represents the Armington specification that imposes the CES functional form on

the product demand function ( Equation 4-3) implying the separability among product sources

(Alston et al. 1990). In terms of utility function specification, these assumptions are equivalent to

the specification that the 0(;s 3 is a COnstant elasticity of substitution (CES) having the form:




where fig; is a share parameter, o-, is the single CES in the its market, Xi. is the demand function

for goods (i.e., beef products) in country i and Xii is the import demand in the ith market for a

product produced in the j country. In order to "collapse" the utility function (U) into Equation 4-

3 the necessary and sufficient conditions are that (a) the Marginal Rate of Substitution (MRS)

between any pair of products competing in the it" market is independent of demand for any other

products) in that market which is known as the assumption of independence, and (b) the









This study deals with total trade data without regional bi-lateral trade and re-exports (i.e.,

Japan exports to Hong Kong or U.S. imports from Australia) to avoid the problem of double-

counting and demonstrates the relative strength of each country in the international beef market.

Therefore, exports of any given country to any partner country (or importing country) are

defined as exports without intra-regional trade.

The International Beef Market

During the 13-year period under consideration, a number of significant changes have taken

place in the world trade environment. One of the most important changes is the significant shift

in the demand for resources, in particular with respect to food and energy, from western

economies to emerging Asian economies like India, China and many other nations on the Pacific

Rim. These emerging economies are absorbing a large portion of the world supply of agricultural

commodities helped by technological improvements in storage and transportation methods. For

example, the average beef consumption in places like Hong Kong and Taiwan has increased

almost 63% during the past 10 years while other developed western nations have shown an

opposite pattern during the same period. Beef exporting nations have shown a significant change

in trade flows during the past decades. Before the 2003 BSE outbreak, the U.S. was considered

among the top three beef exporting countries in the world with significant presence in Asia,

Africa, and Central America. Today, Brazil is considered the largest exporter of red meats

products followed by Australia, which clearly shows the economic impact of food safety issues

and how countries around the globe are adapting their policies to prevent future outbreaks. In

order to see the impacts of these policy changes on trade patterns, this section shows beef trade

flows between selected countries during the period 1995 to 2007 in terms of quantities traded,

market shares, prices and expenditures, and consumption levels for the years before and after the

BSE market restrictions policies took place.









The data set shows that Australian beef exports increased 23.4% and that domestic production

and consumption increased by 15.7 and 24.8 % respectively over this period. These numbers

reflect that Australia is a net beef exporter and that most of its production demand is located

outside the country. Today Australia is the second largest exporter of beef products in the world,

exporting almost 65% of its total production. Like the United States, Australia had focused its

attention on markets around the Pacific Rim where it has a clear advantage over the Rest of the

World beef exporting nations due to its proximity to most of the maj or markets in the region.

Since December 2003, Australian beef has captured two-thirds of the import market shares

in the region as it filled the large market left by the U.S. beef absence. The current situation of

absence or limited presence of U. S. beef in maj or Asian markets continues to present

opportunities for Australian products, limited supplies due to a prolonged drought, limited range

of cuts, and the increasing pressure from competitors (i.e., New Zealand) that have negatively

affected its ability to increase its market shares in the region. Finally, Australia's exports are

expected to decline by 3% during the next year as the United States regains markets in Asia and

as Australia production declines (USMEF 2007).

New Zealand

Although a small producer of beef if compared to the United States and Australia, this

nation has a significant importance as a world beef exporter due to its geographical location and

its relatively large exports levels in relation to domestic consumption and production. New

Zealand is a net beef export country, with around 79.3% of its production being exported

overseas on average over the past 13 years and showing its highest peaks right after the 2003

BSE scare in North America with more than 81% of its production being exported. The average

beef production during the period 1995 to 2007 was about 641,000 metric tons, showing a

significant increase of more than 16% in production levels after 2002, while domestic










preferences for the different qualities in beef product imports. Using EU' s beef imports data

from Mercosur countries, the study shows that TRQs have a significant impact on the average

quality of exports demonstrating that under specific tariff reduction low-quality beef (Brazil) is

maximized, while high-quality beef (Argentina) trade is maximized with large TRQs. The

research concludes that the use of tariff dispersion policies impose a proportionally higher

protection on low unit value products, and have a quality upgrading effect on imports. A cut in

these tariffs may therefore change the composition of imports towards lower unit value products.

Coffey et al. (2005) analyzed the BSE incident on the wholesale sector of the U. S. beef

industry. Specifically, the research estimated economic losses to the U.S. beef industry caused by

increased regulatory costs and reduced exports to Japan and South Korea. These losses were

estimated using excess demand and excess supply relationships coupled with price elasticity

estimates previously reported in other studies. Changes in the excess demand and supply

functions caused by the BSE cases were used to calculate changes in producer and consumer

surplus.

International Beef Promotions

Agricultural commodities are considered less differentiated goods and are generally

advertised through some type of cooperative effort that includes several sectors of the supply

chain such as producers, processors, and exporters (Forker and Ward 1993). Under this

cooperative scenario, advertising and promotion of commodities is known as generic advertising

and it was defined by Forker and Ward as the promotion of a nearly homogeneous product to

disseminate information about its underlying characteristics to existing and potential consumers

for the purpose of strengthening demand for the commodity.

Competition in the international agricultural market has become much more intense,

triggering innovation and transformation inside the industry. As a result of this, a non-segmented













Expenditures on Australia beef (5 per capital)
40.00-
Importing Countries
mKorea 'Japan 'Taiwan O Hong Kong
35.00 _34.50
32.73

30.00 -29.77






20.00 -61

15.00 14.43 15.29 15 53 141
13.13
11.46 126

10.00-


5.00-


mmm-mmmHMH

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
Hong Kong 000 000 000 225 2 11 199 200 200 159 239 352 331 362
Taiwan 0 00 0 00 0 00 3 65 3 57 3 44 3 38 3 28 3 98 4 19 433 5 32 3 96
Japan 10 05 8 78 7 04 7 29 6 81 7 76 7 55 5 61 6 89 10 47 14 68 12 98 13 16
Korea 3 08 2 69 2 68 1 24 2 80 2 88 2 60 3 92 3 69 6 62 10 20 12 89 9 03


Figure 2-20. Total expenditures on beef products from Australia by country. Source: USDA-
FAS. February 2008.

Expenditures on New Zealand Beef (5 per capital)
14.00-
Importing Countries
SKorea I Japan I Taiwan iDHong Kong

12.0 1.6

10.32
10.00-



8.00
6.91

6.00-



4.00- 3


2.17 1.87 m.3 H
2.00-



0.0 0MMuu
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
Hong Kong 0.00 0.00 0.00 3.63 2.62 2.89 2.54 2.03 2.53 2.62 1.89 2.19 2.23
Taiwan 0.00 0.00 2.10 1.91 1.78 2.48 1.83 2.05 2.64 4.19 5.08 4.69 3.34
Japan 0.73 0.72 0.56 0.56 0.40 0.38 0.37 0.29 0.37 0.77 1.19 1.11 1.03
Korea 1.43 1.15 0.70 0.18 0.36 0.56 0.46 0.81 1.37 2.74 3.49 3.11 2.40


Figure 2-21. Total expenditures on beef products from New Zealand by country. Source: USDA-
FAS. February 2008.











STARTING THE BEEF TRADE SIMULATOR
SIM #1 IS THE BASE STARTING VALUES

SET SIMNUM = 1; ? BASE
ZRESETZ;
SET SIMVAR = 1;
ZB SIMZ;


STARTING THE BEEF TRADE SIMULATOR
SIM #2 TRADE BY IMPORTER AND EXPORTER


SET SIMNUM = 2;
ZRESETZ; SET SIMVAR
ZRESETZ; SET SIMVAR
ZRESETZ; SET SIMVAR

ZRESETZ; SET SIMVAR
ZRESETZ; SET SIMVAR
ZRESETZ; SET SIMVAR

ZRESETZ; SET SIMVAR
ZRESETZ; SET SIMVAR
ZRESETZ; SET SIMVAR

ZRESETZ; SET SIMVAR
ZRESETZ; SET SIMVAR
ZRESETZ; SET SIMVAR


1; SET SDIl
2;~~~ SE D
2; SET SDIl
1;~~~ SE D
3; SET SDIl
3;~~~ SE D

1; SET SDI2
2;~~~ SE D
2; SET SDI2


SET SDX1=
S SX=
SET SDX2=
S SX=
SET SDX3=
S SX=

SET SDX1=
S SX=
SET SDX2=
S SX=
SET SDX3=


SET SDX3=


ZB SIMZ;
ZB SIMZ;
ZB SIMZ;

ZB SIMZ;
ZB SIMZ;
ZB SIMZ;

ZB SIMZ;
ZB SIMZ;
ZB SIMZ;


? US TO KOREA;
? AUSTRALIA TO KOREA;
? NEW ZEALAND TO KOREA;

? US TO JAPAN;
? AUSTRALIA TO JAPAN;
? NEW ZEALAND TO JAPAN;

? US TO TAWAIN;
? AUSTRALIA TO TAWAIN;
? NEW ZEALAND TO TAWAIN;


ZBSIMZ;
ZBSIMZ;
ZBSIMZ;


? US TO HONG KONG;
? AUSTRALIA TO HONG KONG;
? NEW ZEALAND TO HONG KONG;


STARTING THE BEEF TRADE SIMULATOR
SIM #3 US TRADE OVER TIME TO COUNTIES

SET SIMNUM = 3;
ZRESETZ; SET SIMVAR = 1; SET SDIl=1; SET SDX1=1; ZBSIMZ; ? US TO KOREA;
DOT(INDEX=H,CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10
SDYR11;
ZRESETZ; SET SIMVAR = H+1; SET .X=1; SET SDIl=1; SET SDX1=1; ZBSIMZ;
ENDDOT; SET HH= SIMVAR;

ZRESETZ; SET SIMVAR =1; SET SDI2=1; SET SDX1=1; ZBSIMZ; ? US TO JAPAN;
DOT(INDEX=H,CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10
SDYR11;
ZRESETZ; SET SIMVAR = H+1; SET .X=1; SET SDI2=1; SET SDX1=1; ZBSIMZ;
ENDDOT; SET HH= SIMVAR;

ZRESETZ; SET SIMVAR =1; SET SDI3=1; SET SDX1=1; ZBSIMZ; ? US TO TAWIN;
DOT(INDEX=H,CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10
SDYR11;
ZRESETZ; SET SIMVAR = H+1; SET .X=1; SET SDI3=1; SET SDX1=1; ZBSIMZ;
ENDDOT; SET HH= SIMVAR;

ZRESETZ; SET SIMVAR =1; SET SDX1=1; ZBSIMZ; ? US TO HONG KONG;
DOT(INDEX=H,CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10
SDYR11;









collect U.S. and international export and import data on agricultural trade. In addition, selected

demographics, national economic indicators, and information on tariff and non-tariff barriers

were collected using the FAOSTAT database from the United Nations, the U.S. Department of

Commerce, and the World Organization for Animal Health (OIE).

For each of the selected international beef markets, the World Trade Atlas database records

import shipments using the Harmonized System code of the product (HS)4 at the 0200 level, the

volume in metric tons, the CIF5 value in US dollars, the country of origin, and a text description

of the product. In addition, yearly information collected from country specific sources (i.e., U.S.

embassies, U.N., and OIE regional/country offices) through field representatives is used to

complement the original dataset. Missing data information is the result of lack of appropriate

statistical records at country levels and in our case they represent a minimal proportion of the

total data set. Finally, the total dataset for this research includes around 25,000 observations from

which market share equations are estimated using more than 2,230 observations containing

information on total import and export quantities and country's specific economic and

demographic variables.

Outline of Chapters

The importance of trade as the underlying force behind globalization has been presented

emphasizing some of the factors affecting today' s world agricultural commerce in particular

those affecting international beef markets. Chapter 1 thus provides the background for this

proj ect. Chapter 2 discusses the international beef industry and the characteristics of each import

and exporting market during the period between 1995 trough 2007. This chapter reviews each

beef supply industry, demand for beef in each importing nation, and international regulations.

4 Governments use HS codes to classify trade into discrete categories for customs duties and for cataloging
exports/imports.
5 CIF prices include cost, insurance, and freight.









Considering the pre-and post-BSE periods, the results show an extreme change in market

composition in particular when considering the case of Australia that doubles its market shares

between 2002 and 2006. In general, price elasticities of the product demands are such that those

for the South Korean market are more elastic than those for Japan, Taiwan, and Hong Kong.

These results clearly indicate the volatility of the South Korean market, while the rest of the

markets are likely to remain relatively constant.

In the case of beef promotions, the simulation results show little or no effects on product

demands. Contrarily to what was expected, this research could not find a link between

promotions and increasing levels of the demand for beef products. As promotion expenditures

increased or decreased from the mean, products demands and market shares show very small

changes indicating that the positive parameter coefficient is no different than zero. Despite

marketing efforts aimed to promote and expand the presence of U. S. beef products in the region,

the negative demand effects of food safety concerns as measured by the embedded BSE effect

across time trend has been the single factor driving all beef import demands in these four

countries, giving little room for an effective and significant impact of the commodity promotion

programs aimed to compensate the inward shift in U.S. beef demand induced by both BSE

information and longer-term consumer preference changes.

Despite the negative press and the reduced presence in Japan, Taiwan and Hong Kong,

U. S. beef is considered by consumers in these countries to be the high quality product while

Australian beef is considered to be the low quality product. However, with food safety being one

of the main drivers of the beef import demand, a high-quality characteristic may not be enough

to overcome the feelings towards the production process and the potential of food borne diseases

that are related to intensive grain-feed produced beef.












the South Korean Market. That is, the amount of products demanded during the three years


considered was 90,000 MT in Taiwan, while in Hong Kong this amount was about 70,000 MT.


In addition, relative price elasticity estimates for Taiwan and Hong Kong show unit elastic


values, thus as beef prices increase from 10 % below the mean or average price to 10 % above


the mean prices there is an identical decrease across time in the quantity of beef product


demanded, of 14% in the case of U.S. beef, 19% in the case of Australian beef, and 18% in the


case of New Zealand beef.

1000 1\T
30.00-
Taiwan
Price Quantity Rlesponse
00 9 01 511.1
25.00-



20.00-



15.00 -- -



10.00 ------



5.00 --------



0.00 h
US 1998 AU 1998 NZ 1998 US 2002 AU 2002 NZ 2002 US 2006 AU 2006 NZ 2006
0.9 16 90 16 08 6 57 16 03 13 65 3 46 0 12 19 67 6 50
115 63 14 37 5 92 14 82 12 19 3 12 0 11 17 57 5 85
1.1 14 54 12 98 5 39 13 79 11 02 2 84 0 10 15 88 5 33
Elasticity -1 01 -1 02 -1 02 -1 01 -1 02 -1 02 0 00 -1 02 -1 02

Adjusted Price


Figure 6-9. Taiwan beef market demands and price elasticities in 1998, 2002, and 2006.

In summary, these simulation results correspond to the descriptive statistics which were


shown in Chapter 2 to be the largest import and export markets. Results show that product


demand responses to variation in the average market prices were more extreme after the 2003


BSE announcements, in particular in the larger markets, such as South Korea and Japan.










negative impact on beef demand, mainly because of the negative publicity from the press and in

particular the TV coverage of the issue. They results also showed that as consequence of the

negative media, pork consumption increased significantly.

Adda (2002) investigated the effects of past consumption of risky goods on current

consumption patterns in France, using the BSE crisis as a natural experiment. He found that new

health information interacts with prior exposure to risks. Consumers with intermediate levels of

past consumption decreased their demand for beef and sought higher quality products, while

low- and high-stock consumers did not alter their behavior after the crisis. In a number of

experiments and surveys, consumers have indicated that they would be willing to pay more for

food with lower risks of disease. Hence, previous studies have found that U.S. consumers were

willing to pay a premium of 15% to 30% per meal to reduce their risk of becoming ill from their

meal and that consumers are willing to pay a premium for reduced pesticide residues in produce

(Buzby 2003).

In a previous study, Jin and Koo (2003) suggested a three-way approach to the BSE effect

on consumer's preferences using contingent evaluation methods, analyzing structural changes in

consumers' or producers' welfare, and evaluating the economic consequences of the outbreaks.

A non-parametric approach was used by them to show that there has been an ongoing structural

change in Japanese consumers' preferences for meat and that consumers' tastes for meat have

systematically moved away from beef to its substitutes. Ishida et al. (2006) investigated the

impacts of the BSE and Bird Flu on the demand for meat products in Japan using the Almost

Ideal demand system. As expected, they found evidence that the appearance of BSE and Bird Flu

cases in Japan deeply affected the structural demand for beef and chicken and have a positive

impact on the demand for their closest substitutes (e.g., pork and fish products). This research









market shares in the region. In all four markets, the BSE scare represented a total U.S. beef

export value loss of $2.4 billion during the 2003 to 2004 period (US1VEF 2007). Since then,

Australia has dominated the fresh and frozen beef market of the region, while New Zealand has

increased its presence in all beef markets on a much smaller scale. For example, in South Korea,

Australian fresh beef products represent 89.5% of the shares, while New Zealand is second with

only 8. 1% of the product demand shares. With respect to frozen products, the shares for

Australian beef are about 47% and 5% in the case of New Zealand. In relative terms, these

numbers represent an increase in product demands of 49% (fresh) and 61% (frozen) for

Australia; and 64% (fresh) and 57% (frozen) in the case of New Zealand products.

In the Japanese beef market, Australian shares are quite less compared to the South Korea

market showing only 24.5% of the total fresh market and 17.4% of the total frozen market.

Comparing pre-and post-BSE shares, this number represents a relative increase of 36% in

product demand shares in both the fresh and frozen market. In the case of New Zealand, the

results show market share levels of 8. 1% (fresh) and 5.8% (frozen), which represent a slight

increase with respect to pre-BSE product demand levels. Less expensive Australian and New

Zealand beef has helped these two countries to compete in the Japanese market, but strong

consumer preferences towards grain-feed production type of beef have helped beef products

coming from other parts of the world (i.e., South America) to establish a significant presence in

Japan. In addition, reports from the United States Department of Agriculture have shown a slow

but positive increase in the demand for U.S. beef products since 2007, which has become a main

focus of interest for U. S. producers and exporters due to the fact that Japan was the most

important market for U.S. beef products in general before December 2003.












Considering product demand levels for beef products that originated in Australia and New


Zealand between 2002 and 2006, the figures above clearly illustrate this trend. In all these


scenarios it is clear that the absence of U. S. beef products has helped the expansion of Australian


products elsewhere, in particular in South Korea were consumers have set a strong position


against beef imports from North America.

1000 1\T
30.00-
Hong Kong
Price Quantity Rlesponse
00.9 01 511.1
25.00-


20.00-


15.00-


10.00 ---


5.00-



0.00 -
US 1998 AU 1998 NZ 1998 US 2002 AU 2002 NZ 2002 US 2006 AU 2006 NZ 2006
0.9 13 27 12 62 5 16 12 59 10 71 2 72 0 09 15 44 5 10
1 12 27 11 28 4 65 11 64 9 57 2 45 0 09 13 79 4 60
1.1 11 41 10 19 423 108X3 865 223 008 12 46 4 18
Elasticity -1 02 -1 02 -1 02 -1 01 -1 02 -1 02 0 00 -1 02 -1 02

Adjusted Price


Figure 6-10. Hong Kong beef market demands and price elasticities in 1998, 2002, and 2006.

This set of scenarios yields useful information to policy makers and beef industry


representatives attempting to understand the impact of restrictive trade policies affecting the


normal flow of agricultural commodities and that are based on arbitrary food safety concerns as


expressed in several occasions by World Organisation for Animal Health (OIE)




SBased on the Terrestrial Animal Health Code, this international regulatory agency specifies all health measures to
be used by the veterinary authorities of importing and exporting countries to avoid the transfer of agents pathogenic
for animals or humans, while avoiding unjustified sanitary barriers. The OIE officially announced the U.S. as a
"controlled risk" country in May 2007.









showed that in the Japanese market, Bird Flu concerns negatively impacted beef s market shares,

although BSE fears increased consumer demand for chicken products. Empirical results also

showed that both impacts do not persist over time and that in the short-run the impact dissipates

depending on the characteristics of the disease (e.g., incubation period, cure rate, and infection

risk) and the ability of the government to efficiently respond to an outbreak.

Trade and Beef Markets Issues

Global food trade will likely increase due to expected increases in income levels around

the world, improved transportation networks, and growing populations requiring greater and

safer quantities of food. In addition, the creation of the General Agreement on Tariff and Trade

(GATT) has integrated global markets into an overall framework that affects not only the volume

of goods traded but also the diversity of products flowing across nations. As a result, this

ongoing process of trade liberalization has raised consumers' concerns about the safety of food

products (Bureau et al. 2002).

The USDA-ERS published a report (Regmi 2001) that studied structural changes in

international trade and consumption as a result of food safety incidents. It argued that following

the resolution of the problem that caused a maj or international incident, consumer perceptions

about the implicated food product and about the exporting country's ability to produce safe food

may be slow to change, and these perceptions have a lasting influence on food demand and

global trade. Consumers value a safe food supply, and recently food safety scares like the BSE or

problem or the E. coli outbreak have raised awareness about food safety issues.

Food safety regulations and the perception of risk are different among countries. Even if

the food safety risks are the same across countries, countries may perceive and handle these risks

differently, which can lead to persistent trade frictions and even reduce food trade (Buzby 2003).

Although little disruption to trade has occurred for food safety reasons (considering the total











corresponding market shares values above, at, and below the mean, while Figures 6-11 and 6-12

illustrate the values at the mean levels.

Table 6-5. Estimated variations in market demand distribution
United States Australia New Zealand
PRE POST PRE POST PRE POST
Fresh Frozen Fresh Frozen Fresh Frozen Fresh Frozen Fresh Frozen Fresh Frozen
S.K.
+20 0.697 0.273 0.001 0.001 0.432 0.175 0.883 0.461 0.031 0.019 0.086 0.046
0 0.714 0.269 0.001 0.001 0.456 0.176 0.895 0.463 0.032 0.020 0.091 0.046
-20 0.725 0.266 0.001 0.001 0.471 0.453 0.916 0.420 0.033 0.061 0.093 0.042
JP
+20 0.202 0.143 0.357 0.001 0.155 0.110 0.246 0.174 0.054 0.038 0.082 0.058
0 0.201 0.143 0.445 0.001 0.155 0.110 0.245 0.174 0.054 0.038 0.082 0.058
-20 0.201 0.049 0.511 0.047 0.154 0.174 0.245 0.168 0.054 0.067 0.081 0.056
TW
+20 0.449 0.321 0.089 0.003 0.345 0.246 0.546 0.390 0.120 0.086 0.182 0.130
0 0.449 0.321 0.112 0.003 0.345 0.246 0.546 0.389 0.120 0.086 0.182 0.130
-20 0.449 0.108 0.128 0.105 0.345 0.388 0.546 0.374 0.120 0.149 0.182 0.124
H.K.
+20 0.449 0.321 0.070 0.003 0.345 0.246 0.546 0.390 0.120 0.086 0.182 0.130
0 0.449 0.321 0.088 0.003 0.345 0.246 0.546 0.390 0.120 0.086 0.182 0.130
-20 0.449 0.109 0.101 0.105 0.345 0.388 0.546 0.375 0.120 0.150 0.182 0.124



Looking at the simulation results for the United States before 2003, it is clear that the

United States was the leader in percentage of product demanded in all four importing markets

and that the share of fresh beef was larger than the share for frozen products. That is, in South

Korea the difference between U.S. beef and products originated in Australia was 25.8% and

9.3% for fresh and frozen products respectively; in Japan the difference in product demand

shares was 4.6% and 3.3 %, and in comparison to New Zealand products the difference between

market shares was 68.2% for fresh and 24.9% frozen; and in Taiwan and Hong Kong this

difference was 10.4% and 7.5% in each market. Figure 6-11 illustrates that in 2003 New Zealand

beef, in all markets, has the smallest shares in terms of product demands representing a

combined share (fresh and frozen products) of only 5.2% in South Korea and 9.2% in Japan,

while in Taiwan and Hong Kong the combined share is 20.5% in each market.









In(Xg y) = eoij + e11] (In(Pg y) In(Pi.)) + 021j In(Xi.) + uii (4-22)

where the first component, Goij, is a constant of integration that represents the effects of

distribution and differentiating coefficients on market shares; the second component, 81i is the

estimated value of the product demand relative price elasticity and is given by myi; the third

component, 92ij, is the estimate value of the product demand market size elasticities; and the last

component, uii, is a white noise error of the estimable model and allows for error in the

adjustment of the inputs to their utility maximization level. Thus, comparing Equations 4-18 and

4-21 the following identities are obtained:




8luj = (1/(aii 1)) = 7my and (4-24)

82ij= Ke~ag 1) 1).(4-25)

Equation 4-22 and its corresponding identities clearly show that the parameters

90tjl 8luj, and 92ij arT HOH-linOrT, HOn-constant and may change over time; that is, these

parameters are some measure of the coefficients of the cross function (Sparks and Ward 1992).

The adjustment process captured with these relations are the most revealing part of the model,

since it shows whether or not the parameter is moving in a particular direction representing the

changes in beef demand. Thus the system for beef trade as specified for econometric estimation

will obtain elasticities of substitution to analyze the impact of the parameter coefficients included

in the CRES function.

Recalling the previously defined Equations 4-7 and 4-22, the price elasticity can be

obtained substituting them back into Equation 4-22 as follows:

In(Xii) = Boil + Bzzy In(Pii) Bzzy In(Pii Xi/Xi.) + 821j In(Xi.). (4-26)









The a and P coefficients have particularly important meanings in the CRES function. First,

the p shows that each country has an initial share of the import markets and those shares can

differ for a fixed level of import demand. Clearly, P does not just happen. It must depend on

many conditions such as non-tariff restrictions, previous trade history, infrastructure, food safety

concerns, preferential treatment, etc. Second, the product category also differs across countries

and shares change as the imports Xii change. Both ail and ai. reflect those differences and like

the p's they do not just happen. They reflect true quality and other attribute differences among

exporting countries as well as perceptions about a particular product Xii from country j in

country i. For example, information influences perception and knowledge about a product' s

attributes. Anything that enhances this knowledge also impacts the ax's (e.g., generic advertising,

product attributes). Later in the chapter, more specific forms for capturing the levels of both p's

and a's are developed showing that both coefficients may change over time.

Given the CRES market demand function, Equation 4-9, the following optimization steps

will define the product demand function for beef products. The first order condition for optimum

product mix implies

Ps y = Pi. (aX,./aXg y), (4-12)

where Pi. is the price level in the ith market and


=x .i~ i (4-13)


Thus, the partial derivatives in Equation 4-13 depend only on ratios of quantities of

products demanded in the its market, and these ratios in turn depend only on ratios of the

product prices which mean that market shares must depend only on relative prices of the

products in the market (Armington 1969). The fulfillment of this condition determines the











accounted for one-third of the beef consumption in Japan. Consequently, domestic production


and imports from different sources, in particular Australia and New Zealand, have increased


since then, but these sources of beef have been incapable of meeting the Japanese demand as


show in Figure 2-5, resulting in a tight supply situation. Today, according to estimates from the


USDA-FAS, Japanese beef imports are expected to increase more than 10% between 2007 and


2008. The U. S. presence in the Japanese beef market has been almost insignificant compared to


pre-BSE levels due to trade restrictions that constrained American exports to cattle under 21


months old, in addition to high U. S. beef prices and consumer anxiety.

C IF Pri ces for Fres h B eef in Ja pan ($ per Kg)
12-
U.S. o AU INZ --World Av.


10-











9- 9 20 01 02 20 00 05 20

8-a

7-ue27 vrg I rcsfrfehbe pout nJpnfo 99t 06 ore

6-AFS ebur 08

5- sepnieAsrla efhsbe a o atrafcigU .mre atcpto

4-teJpns akt speetdi iue -,Aeia efhsbe omrilzda

3-hrpie nrlto oAsrla ef- ewe 9 n 5 ihr-dpnigo h



























































I


y Regon
m 1995 m 2007


Volume, Share, and Growth of Global Production

Tables 2-1, 2-2, and Figure 2-1 show the world total production by region, the share of

each region, and the rate of growth for the years 1995 and 2007 in terms of quantities and

percentages respectively.

The world beef production totaled 48.53 millions of metric tons (M.MT) during the year

1995 and 54.48 M.MT during the year 2007, representing an increase of almost 11%. The

Americas region concentrated more than one half of the world production with more than 50% of

the share in 1995 and almost 53% in 2007, where the South American portion has shown the

largest boost in production share with almost a 4% increase. In particularly, the North American

region, which includes the U.S., Canada, and Mexico emerged as the largest producing region,

while the South American region that includes Brazil, Argentina, Paraguay, etc., appears as the

world's second largest producing region during the same years, Figure 2-1.

M orld Beef Production by Regon (M.MT)


16.00

12.00

8.00


m 1995 12007


S America N. America Asia* East Asia Eur~opean U Oceania ROM
Regions

.-1. Total beef production and shares by region. Source: USDA-FAS. January 2007.


Figure 2


Share of Total M orld Production b


rl









allowing the conversion of these changes into price effects. The Armington specification

assumes that imports originating from different sources are imperfect substitutes of each other,

that the demand functions show constant elasticities of substitution (CES), and the elasticity of

substitution between any two products competing in a market is the same as that between any

other pair of products competing in the same market.

As in Preckel, Cranfield, and Hertel (2005) and Femenia and Gohin (2007), the most

convenient way of representing the explicit form of the Armington CES product demand

function is:

u~xiXn>= (CEPi.i.-P P A-1


where u denotes utility and xi denotes a vector of differentiated goods by their origin, and /7 and

p are the distribution and substitution parameters of the CES utility function. In order to simplify

the interpretation, Armington (1969) "collapsed" the Utility function assuming that any given

quantity-index function,0gi2, has the generalized CES form as previously defined. Suppressing

the arguments of u and rearranging Equation A-1 leads to the following implicit expression of

the same relationship:


Xi. = Oi(xi, ,...,xin)= ( ?=14t.xi-P) P, i=1, 2,...,n. A-2)

Since the utility is an ordinal measure, there is no consequence to normalize the pig parameters

by imposing their sum equal to one (1 ..z= Pi = 1). Thus, based on the theory of the utility-

maximizing consumer, where any given quantity of a good is to be obtained at least money cost,

we know that the marginal rate of substitution (MRS) between two goods must equal the ratios

of their prices (Varian 1992):


O r is assumed to be linear and homogeneous, thus all demands for Xi. products are the same (Armington 1969).
SThe CES technical relationship is defined as: X,. = 74 fit; X~'r









In Taiwan and Hong Kong, the post-BSE markets show similar shares for Australian beef,

54.5% for fresh products, and 38.5% for frozen products, which represent an increase with

respect to pre-BSE levels of 20. 1% and 14.3% respectively. In the case of New Zealand beef,

Taiwan and Hong Kong also show similar market results. That is, New Zealand products

represent 18. 1% of shares for fresh beef and 13.1% of the shares for frozen beef in these two

markets, which correspond to 6.2% and 4.4% increase in fresh and frozen market shares with

respect to pre-B SE levels.

This chapter concentrated on a more detailed interpretation of the coefficients estimates

presented in Chapter 5. Simulations were conducted for each of the four importing beef markets

on the functional relationship presented in Chapter 4, except for simulations representing the

Rest of the World. That is, the parameters estimated with the non-linear procedure were used to

simulate the reactions of the dependent variable to changes in the explanatory variables, holding

all other variable constant (ceteris paribus). Three types of simulations were carried out on the

product demand equations. First, the product demands were analyzed across time and the effect

of the BSE announcements measured; second, the average prices were varied from 10 % below

the mean or average price to 10% above the mean prices (1997); and third, product demand

shares were simulated when the average quantity demanded varied from 20% below the mean or

average price to 20% above the mean prices (1997). The simulation results for the beef trade

model proposed in Chapter 4 make economic sense and yield insights into the trends observed in

Chapter 2. This speaks very well of the model; it captures the real world trends occurring in beef

trade during the 1997 to 2007 period.










prices and values, total market size of the industry, and expenditures on promotions and other

marketing activities.

Yearly export and import data collected from the World Trade Atlas (Global Trade

Information Services, Inc.) during the 1995 to 2007 period is used in this research. The quantities

and prices of beef products from the four producing regions to the four importing countries will

be analyzed at the H.S. 0201 and 0202 levels that define meat of bovine animals as fresh or

chilled and frozen. Market shares for each region will be estimated within the construct of the

model to allow for the estimation of consistent parameter estimates. The analysis will clearly

illustrate how consumers in selected markets differentiate products by their origin of production

and how promotions can affect market shares for U.S. products.

Percentage of global market share
60.096
-Percentage of World Exports (U.S., Australia, and New Zealand)




40.090
34.8%











1980 1983 1986 1989 1992 1995 1998 2001 2004 2007
Years

Figure 1-3. Beef trade participation of selected countries from 1980 to 2007. Source: USDA-
ERS. November 2007.






2 Harmonized System code of the product (HS).









characteristics of the results. Model estimates are used to develop a sensitivity analysis and

examine the impacts of trade policies on beef consumption in selected Asian markets.

Adapting the Armington Model to Beef Trade

The Armington model is used to estimate demand elasticities between products from

different sources and then used to simulate the effect of exogenous demand shocks from normal

marketing efforts, from food safety scares such as BSE, and from trade restrictions. This

approach is flexible and provides cross-price elasticities between imports from different sources

using estimates of the aggregate price elasticity of demand for imports (Jung 2004). Furthermore,

demand elasticities as well as estimated coefficients of non-economic variables can be used to

formulate effective policies targeted towards expanding sales and market shares for U.S. beef.

Price elasticities measure the responsiveness of trade flows to price changes. Elasticities of

substitution provide the cross-price elasticity between products from different origin countries

and measure the degree to which price changes could influence market share among the

importing countries to a specific importing country. Since one of the obj ectives of this research is

to measure the impact of restrictive policies on beef trade, the size of those impacts largely

depends on the magnitude of the elasticity (McDaniel and Balistreri 2002).

The Armington specification presents a two-stage budgeting procedure in which

consumers allocated their total expenditures in a sequence of steps. In the initial step, a consumer

or buyer decides how much of a particular product defined as Xie- to buy (Equation 4-1). Next,

given the total amount demanded, the buyers decide how much to import from each country

(Equation 4-2). Using i to represent the four demand regions and j to represent the four supply

regions, the product import demand functions for beef for each region can be represented as in

Equation 4-1 where the dot denotes the summation over all beef imports. This general demand

function assumes that market demands for beef products are functions of the average price in the










SET SDIl=1: SET SDX1=1: ZBSIMZ:
ENDDO: SET H= SIMVAR:


ZRESETZ: SET SIMVAR = 1: SET SDIl=1: SET SDX2=1: ZBSIMZ: ? AUSTRALIA
DO ADJWQALL = .80 TO 1.20 BY .05;
SET SIMVAR = SIMVAR+1:
SET SDIl=1: SET SDX2=1: ZBSIMZ:
ENDDO: SET H= SIMVAR:

ZRESETZ: SET SIMVAR = 1: SET SDIl=1: SET SDX3=1: ZBSIMZ: ? NEW ZEALAND;
DO ADJWQALL = .80 TO 1.20 BY .05;
SET SIMVAR = SIMVAR+1:
SET SDIl=1: SET SDX3=1: ZBSIMZ:
ENDDO: SET H= SIMVAR:


STARTING THE BEEF TRADE SIMULATOR
SIM #11 RELATIVE TOTAL DEMAND IS ADJUSTMENTS IN JAPAN


SET SIMNUM = 11:
ZRESETZ: SET SIMVAR = 1: SET
DO ADJWQALL = .80 TO 1.20 BY .05;
SET SIMVAR = SIMVAR+1:
SET SDI2=1: SET SDX1=1: ZBSIMZ:
ENDDO: SET H= SIMVAR:

ZRESETZ: SET SIMVAR = 1: SET S
DO ADJWQALL = .80 TO 1.20 BY .05;
SET SIMVAR = SIMVAR+1:
SET SDI2=1: SET SDX2=1: ZBSIMZ:
ENDDO: SET H= SIMVAR:

ZRESETZ: SET SIMVAR = 1: SET S
DO ADJWQALL = .80 TO 1.20 BY .05;
SET SIMVAR = SIMVAR+1:
SET SDI2=1: SET SDX3=1: ZBSIMZ:
ENDDO: SET H= SIMVAR:


SDI2=1: SET SDX1=1: ZBSIMZ: ? US





jDI2=1: SET SDX2=1: ZBSIMZ: ? AUSTRALIA





jDI2=1: SET SDX3=1: ZBSIMZ: ? NEW ZEALAND;


STARTING THE BEEF TRADE SIMULATOR
SIM #12 RELATIVE TOTAL DEMAND IS ADJUSTMENTS IN TAWAIN


SET SIMNUM = 12;
ZRESETZ: SET SIMVAR = 1: SET
DO ADJWQALL = .80 TO 1.20 BY .05;
SET SIMVAR = SIMVAR+1:
SET SDI3=1: SET SDX1=1: ZBSIMZ:
ENDDO: SET H= SIMVAR:

ZRESETZ: SET SIMVAR = 1: SET S
DO ADJWQALL = .80 TO 1.20 BY .05;
SET SIMVAR = SIMVAR+1:
SET SDI3=1: SET SDX2=1: ZBSIMZ:
ENDDO: SET H= SIMVAR:

ZRESETZ: SET SIMVAR = 1: SET S


SDI3=1: SET SDX1=1: ZBSIMZ: ? US





jDI3=1: SET SDX2=1: ZBSIMZ: ? AUSTRALIA





jDI3=1: SET SDX3=1: ZBSIMZ: ? NEW ZEALAND;















ROW 50.0%
Foreign Beef Consumption per Capita in K~g
25.00-

20.60 ustralia 5.0%

20.00-
17.4 N. Zealand 12.0%

US 33.0%
15.00-








0.00 '




Pre-BE ba Pos-BSEban mport Marlat Shares for Beef Products in Hong Kong
2003 2007
Year


Figure 2-18. Hong Kong imported beef consumption per-capita and market shares. Source:
USDA-FAS. February 2008.


China's impressive economic growth and Hong Kong' s stable macro-economic indicators


(per capital income growth and currency appreciation) contributed to a steady increase in the


demand for imported beef products during the following years. In late 2005, Hong Kong


resumed imports from the U.S. of boneless beef products from cattle under 30 months of age


only, while main China declared itself open for U. S. beef in 2006. In that year, Hong Kong


imported 20 times more than the rest of China, which corresponded to 80,000 metric tons versus


4,800 metric tons respectively (FAS 2007). This difference was also translated in terms of


consumption levels of imported beef where China consumed only 5.4 kg per capital of imported


beef, while Hong Kong consumed 20.6 Kg during 2007. In terms of market participation, the


ROW group has been the largest winner from the U.S. export ban absorbing 81% of the share









United Kingdom, Burton and Young (1996) investigated the impact of BSE on the demand for

beef and other meats using an AIDS model which included indices of media coverage of BSE.

They found that the influence of negative press had significant effects on the allocation of

consumer expenditures among the meats. A short-run impact was identified that accounted for a

large portion of the perceptible drop in the market share of beef in the early 1990s. Results also

showed that there also appears to be a significant long-run impact ofBSE, which by the end of

2003 had reduced the beef market share by 4.5%.

Kinnucan et al. (1997) investigated the impact of health information and generic

advertising on U.S. meat demand using a Rotterdam specification and concluded that this

demand is affected by prices, expenditures, and health information, but the effect of generic

advertising is less clear and not robust. Health-information elasticities in general were larger in

absolute value than price elasticities, which suggest that small percentage changes in health

information have larger impacts on meat consumption than equivalently small percentage

changes in relative prices.

In 1998, Latouche, Rainelli, and Vermersch conducted a survey using a contingent

valuation method to analyze consumer behavior in France after the BSE outbreak in the United

Kingdom. Their survey revealed that the BSE disease raises the problem of loss of public

confidence, in addition to the fact that beef consumers expected greater transparency from the

industry and that they would accept paying for a system that guarantees the safety of the beef

products consumed (willingness to pay).

Similarly, Verbeke and Ward (2001) investigated fresh meat consumption in Belgium

during the period from 1995 through 1998 using an AIDS model which incorporated a media

index to measure the BSE impact. They found that health safety scares have a strong and









funded by a specific firm with the intent of benefiting that firm's demand by differentiating the

product from other suppliers in order to increase the market share of the brand within the same

industry (Ward 1997).

For example, previous market studies conducted by the USMEF (2007) on beef demand in

Japan have shown that the most important characteristics of food products were that they taste

good and be guaranteed safe to eat, while the most important characteristics of beef products

were that they look fresh, not have a lot of waste, be certified as USDA inspected, and be free of

chemical residues and foodborne hazards. New marketing programs sponsored by the US1VEF

and the United States beef council promotion programs (e.g., "Iowa Beef" and "Beef from

Nebraska") are reinforcing these attributes, emphasizing the characteristics of the product,

adding value, and improving presentation in order to build a positive perception about branded

products.

In the case of U. S. beef exports, producers and trade agents have been able to differentiate

American products from other sources (e.g., Australia and New Zealand) due to continued efforts

to keep the "U.S. brand" recognition in key markets such as Japan, Korea, Taiwan, and Hong

Kong. If safety is a concern, experimentation is not an option and consumers look for assurance

about the safety. Quality assurance through industry seals and government inspection may solve

some of the problem. Even with safe products, some attributes cannot be judged through

experience. These credence attributes must be identified through means other than consumption,

and branded beef products may capture shares of a market through a consistent message about

one or more credence attributes (Ward and Lambert 1993). For example, brands emphasizing

U. S. produced beef may capture some loyalty even with the generally homogeneous nature of the

product group. Thus, specific beef products within a generally common product category may









BIOGRAPHICAL SKETCH

Oscar Ferrara was born and raised in Asuncion, Paraguay. In 1992 he earned a B.S. degree

in agricultural engineering at the Universidad Nacional de Asuncion in Paraguay. In August

2001 he earned a B.S. degree in applied economics from the University of Minnesota and in

August 2005 he earned a M.S. degree in food and resource economics at the University of

Florida. In November 2007 he was admitted to candidacy for Doctor of Philosophy in food and

resource economics at the University of Florida. His research interests are focused on

agricultural marketing, trade and food safety.












TABLE OF CONTENTS


IM Le

ACKNOWLEDGMENT S .............. ...............4.....


LI ST OF T ABLE S ........._..... ...............7..____ ......


LIST OF FIGURES .............. ...............8.....


AB S TRAC T ............._. .......... ..............._ 10...


CHAPTER


1 INTRODUCTION ................. ...............12......... .....


Importance of Agricultural Trade ........_................. ............_........1
International Beef Market ........_................. ........_._ .........1
International U. S. Beef Promotions ................. ...............19......... ....
Problem Statement and Research Objective ........_................. ........._._ ....... 2
Scope..................... ..............2
Research Methodology .............. ...............24....
Armington Trade Model .............. ...............25....
D ata Set............... ...............29..
Outline of Chapters ........._._.... ...............30..._ .........


2 WORLD BEEF TRADE AND PRODUCTION ...._.__... ........_.. ...._._.. .........3


Introducti on ........._...... .......... .. ...............32.....
International Beef Trade Data .............. ...............32....
The Intemnational Beef Market. ........._........ .... .....__... ...............34....
Volume, Share, and Growth of Global Production .............. ...............35....
Beef Trade Patterns ...................... ...............37
Pacific Rim Demand for Beef Products. ........._._........... ...............42...

Republic of South Korea .............. ...............43....
Japan ........._ _. ....... ._. ... ....... .............4
Republic of Taiwan and Hong Kong ........._._.... ...............48.._._.. ....
International Beef Supply .............. ...............52....
United States............... ...............52.
A ustralia .............. ...............55....
New Zeal and .............. ..._.. ...............56....
Rest of the World.. ............. .... .......... ....__ ..........__ .....__ ..............58
Bovine Spongiform Encephalopathy (B SE): Impact on Beef Trade ................. ................. 59
Consumption Levels and Market Shares ................ ...............59........... ...
Expenditures on Beef by Country of Origin ................. ...............65........... ..
Pre and Post-BSE Pattern of Trade .............. ...............69....











STARTING THE BEEF TRADE SIMULATOR
SIM #21 RELATIVE DEMAND ADJUSTMENTS AND TIME FOR THE~ US

SET SIMNUM = 21;
ZRESETZ; SET SIMVAR = 1; SET SDIl=1; SET SDX1=1; ZBSIMZ; ? US
DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11;
ZRESETZ; DO ADJ_WQALL = .80 TO 1.20 BY .05;
SET SIMVAR = SIMVAR+1;
SET .X=1; SET SDIl=1; SET SDX1=1; ZBSIMZ;
ENDDO; ENDDOT;

ZRESETZ; SET SIMVAR = 1; SET SDI2=1; SET SDX1=1; ZBSIMZ; ? US
DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11;
ZRESETZ; DO ADJ_WQALL = .80 TO 1.20 BY .05;
SET SIMVAR = SIMVAR+1;
SET .X=1; SET SDI2=1; SET SDX1=1; ZBSIMZ;
ENDDO; ENDDOT;

ZRESETZ; SET SIMVAR = 1; SET SDI3=1; SET SDX1=1; ZBSIMZ; ? US
DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11;
ZRESETZ; DO ADJWQALL = .80 TO 1.20 BY .05;
SET SIMVAR = SIMVAR+1;
SET .X=1; SET SDI3=1; SET SDX1=1; ZBSIMZ;
ENDDO; ENDDOT;

ZRESETZ; SET SIMVAR = 1; SET SDX1=1; ZBSIMZ; ? US
DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11;
ZRESETZ; DO ADJWQALL = .80 TO 1.20 BY .05;
SET SIMVAR = SIMVAR+1;
SET .X=1; SET SDX1=1; ZBSIMZ;
ENDDO; ENDDOT;


STARTING THE BEEF TRADE SIMULATOR
SIM #22 RELATIVE DEMAND ADJUSTMENTS AND TIME FOR AU

SET SIMNUM = 22;
ZRESETZ; SET SIMVAR = 1; SET SDIl=1; SET SDX2=1; ZBSIMZ; ? AU
DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11;
ZRESETZ; DO ADJWQALL = .80 TO 1.20 BY .05;
SET SIMVAR = SIMVAR+1;
SET .X=1; SET SDIl=1; SET SDX2=1; ZBSIMZ;
ENDDO; ENDDOT;

ZRESETZ; SET SIMVAR = 1; SET SDI2=1; SET SDX2=1; ZBSIMZ; ? AU
DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11;
ZRESETZ; DO ADJWQALL = .80 TO 1.20 BY .05;
SET SIMVAR = SIMVAR+1;
SET .X=1; SET SDI2=1; SET SDX2=1; ZBSIMZ;
ENDDO; ENDDOT;

ZRESETZ; SET SIMVAR = 1; SET SDI3=1; SET SDX2=1; ZBSIMZ; ? AU
DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11;
ZRESETZ; DO ADJ_WQALL = .80 TO 1.20 BY .05;
SET SIMVAR = SIMVAR+1;













Finally, Figure 2-22 shows expenditure levels of each importing nation on beef products


from countries clustered in the Rest of the World (ROW) category. Until 1997 Hong Kong was


part of the Commonwealth of England and all of it imports were made through the British


authority. Since 1998, Hong Kong became part of China and most of it imports started to arrive


from different places in the world, in particular from Latin American countries, which after the


2003 BSE scare through 2007 remained the maj or beef supplier of Hong Kong in terms of


expenditures per capital. The ROW region has been an important source of beef products for


Japan during the years previous to the BSE announcement, but since then this region has lost


importance for Japanese traders due mainly to the lack of enforcement of international food


safety regulations and OIE standards.


Expenditures on Rest of the lWorld Beef ($ per capital)
24.00230
Importing Countrir "0
m Korea = Japan = Taiwan c I I..I1' lIl*


2"-**~


21.00


18.00


,I


I ~~


1111


1.1~1
1.11~


1.1'.1


12.00-


9.00-


6.33
6.00-
5.24 47







1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
Hong Kong 0.00 0.00 0.00 7.86 8.39 13.18 11.47 9.73 9.84 15.25 13.23 18.86 21.65
Taiwan 0.00 0.00 1.05 1.16 1.13 1.72 1.34 0.91 1.32 0.51 0.41 0.83 0.63
Japan 5.07 6.06 3.34 4.18 3.83 4.27 3.52 2.38 1.91 1.15 0.36 0.42 0.57
Korea 0.17 0.27 0.34 0.22 0.49 1.53 0.55 0.78 0.46 0.07 0.24 0.37 0.22


Figure 2-22. Total expenditures on beef products from the ROW by country. Source: USDA-
FAS. February 2008.










EQSUB EQ5 EQ2:
EQSUB EQ6 EQ2 EQ7:
EQSUB EQ1 EQ4 EQ5 EQ6:

PARAM KO .75 KI1 .75 KI2 .75 KI3 .75;
PARAM GO .75 GX1 .75 GX2 .75 GX3 .75 GF1 .75 GG1 .75 GI1 .75 :
? GI1 .75 GI3 .75 GI2 .050;
PARAM HO .75 :
? HI1 .75 : ? HI2 .75; ? HI3 .75 :
PARAM HYR2 .75 HYR3 .75 HYR4 .75 HYR5 .75 HYR6 .75 HYR7 .75 HYR8 .75 HYR9 .75
HYR10 .75 HYR11 .75 HYR1X1 .75
HYR2X1 .75 HYR3X1 .75 HYR4X1 .75 HYR5X1 .75 HYR6X1 .75 HYR7X1 .75 HYR8X1 .75
HYR9X1 .75 HYR10X1 .75 HYR11X1 .75
HYR2X2 .75 HYR3X2 .75 HYR4X2 .75 HYR5X2 .75 HYR6X2 .75 HYR7X2 .75 HYR8X2 .75
HYR9X2 .75 HYR10X2 .75 HYR11X2 .75
HYR2X3 .75 HYR3X3 .75 HYR4X3 .75 HYR5X3 .75 HYR6X3 .75 HYR7X3 .75 HYR8X3 .75
HYR9X3 .75 HYR10X3 .75 HYR11X3 .75;

SELECT (WYEAR>1996) & (MISS(WQQ)=0) & (MISS(WQALL)=0) & (MISS(WRP)=0) & (WRP>.50);
TREND TTT:
CORR DI1 DI2 DI3:
LSQ(HETERO) EQ1:
PRINT SCOEF: MAT NR = NROW(0 COEF): PRINT NR:
? SHOW ALL:
ANALYZ EQ4 EQ5 EQ6 EQ2 EQ3 EQ7:
MSD WQQ WQALL:

SET KKO = a COEF(1);
SET KKI 1 = a ~COEF(2):
SET KKI2 = a ~COEF(3);
SET GGO = a COEF(4);
SET GGX1 = a COEF(5);
SET GGX2 = a COEF(6);
SET GGX3 = a ~COEF(7);
SET GGI1 = a COEF(8);
SET GGF1 = a COEF(9);
SET GGG1 = a COEF(10);
SET HHO = a ~COEF(11);
SET HHYR2 = a ~i2COEF(12):
SET HHYR3 = a COEF(13);
SET HHYR4 = a COEF(14);
SET HHYR5 = a COEF(15);
SET HHYR6 = a COEF(16);
SET HHYR7 = a COEF(17);
SET HHYR8 = a COEF(18);
SET HHYR9 = a COEF(19);
SET HHYR10 = a COEF(20);
SET HHYR11 = a COEF(21);
SET HHYR1X1 = a COEF(22):
SET HHYR2X1 = a COEF(23);
SET HHYR3X1 = a COEF(24);
SET HHYR4X1 = a COEF(25);
SET HHYR5X1 = a COEF(26);
SET HHYR6X1 = a COEF(27);
SET HHYR7X1 = a COEF(28);
SET HHYR8X1 = a COEF(29);










(e.g., price, color, size). Alternatively, experienced goods are those where the attributes can be

observed at the time of consumption (e.g., taste, tenderness, fat content) and influence

consumers' satisfaction with a particular meat product. Finally, credence goods include attributes

that are not explicitly observed in the product (e.g., safety: antibiotic and pesticide residues;

health: cholesterol levels; environmental: "green" production techniques/conditions) but

consumer credibility of the claims made about the product is influenced by a high level of

confidence in the effectiveness of the systems in charge of creating and monitoring the product

(Codron, Stems and Reardon 2003).

Purchase decisions are based on predictions of product performance. Consumers base their

predictions in part on product cues and are accurate to the extent that they have properly learned

the relationship between the cues and performance. If consumers leamn the relationship between

product attributes and quality, they will differentiate among products that possess different

attributes and treat as commodities those brands that share the same attributes. Once the

predictive rule is learned, it may be applied to any existing or new product that possesses the

same set of attributes (Van Osselaer and Alba 2000).

Advertising and promotion provide this information facilitating purchasing decisions, and

in some cases they even change the underlying preference function for a particular product or

service. Depending on the characteristics of the product, potential buyers, and the scope and

beneficiaries of the marketing campaign, there are two types of advertising and promotion of

commodities: generic alheill iving. a cooperative effort among producers designed to collectively

increase the primary demand (i.e., the size of the pie) of a product by advertising the attributes

and characteristics of the product, without influencing the market share of any producer (e.g.,

Beef checkoff program, Florida citrus, Washington apples); and brandadvertising, designed and









EXPORT DEMAND AND PROMOTIONS OF U.S. BEEF: IMPACT OF FOOD SAFETY
AND PROTECTIONISM.

















By

OSCAR FERRARA


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2008









CHAPTER 3
LITERATURE REVIEW

Introduction

This chapter presents a review of previous research on areas related to agricultural trade,

food safety issues and policies, commodity advertising and promotion programs, and the

Armington model and its applications. Since this study focuses on international trade and

marketing the discussion in this chapter will provide a foundation for evaluating some of the

issues affecting market access efforts and import demand for beef products from the United

States, Australia, New Zealand, and the Rest of the World in markets such as South Korea,

Japan, Taiwan, and Hong Kong

International Beef Demand and Food Safety

International beef consumption has increased considerably over the last three decades due

to trade liberalization and more effective food safety policies. Since food is a necessity,

consumers value the fact that their food is free of toxins, foreign material, and pathogens. Food

safety concerns have increased as wealth has risen. Now that many consumers in the

industrialized world have adequate quantities of food, they (or their governments) can spend

resources to ensure that their food is safer (Jin et al. 2004).

As of December 2007, the world has already been impacted by several food-borne disease

outbreaks where BSE, Avian Influenza (AI), Salmonella, and E. coli are among the most

notorious and controversial. In the United States, the BSE disease remains a serious concern to

the cattle and beef industry as traditional U.S. beef importers such as Korea and Japan have not

yet fully opened their markets to American beef products. Coffey et al. (2005) argue that in the

event of a widespread BSE occurrence there will be a large decline in domestic and foreign

demand since consumers' reactions to BSE cases around the world have permanently changed









of the parameter means that the two products are dissimilar or, that they are weak substitutes

(Kapuscinski and Warr 1999).

The imposition of the CRES technical relationship on the markets clearly shows that all

elasticities of substitution (aii) vary proportionately with the common factor of proportional

change That is, the elasticity of substitution oil and the value share Sii of the jth beef product

in the ith beef market are represented as



where
Sq = Pii Xii/C JPii Xi (4-11)

These elasticities play a significant role in trade modeling, especially when analyzing the

impact of trade policies. For example, when a tariff applied to beef imports is increased to

protect the domestic beef production in a country such as South Korea, this change automatically

raises the domestic price of the imported beef. Nevertheless, the effect that this change in the

tariff has on the price of the domestically produced beef is what determines its domestic resource

allocation effects (Armington 1969). In the special case of specific tariff applied to U. S. beef

products competing in Asian markets, if all competing beef products are perfect substitutes, then

the price of beef from the United States will necessarily change (increase) by the same

proportion as that of the other suppliers change (decrease). However, if the goods are imperfect

substitutes, prices may not change by the same proportion. Thus, the impact that changes in trade

policy have on the str-ucture of the markets (e.g., market shares) basically depends on the degree

of substitutability between imported beef, and this is what the model that is suggested for this

study captures (Kapuscinski and Warr 1999).

5 Previous studies by Mukerji (1963), Gorman (1965), Sato (1967), and in particular the paper by Hanoch (1971)
suggested that all ot; could vary proportionately, with a conunon factor being defined as: 1/ Ci Sijaii = 1/il,
where a is a weighted average of {ai} with the factor shares (Si} as weight.











2-20 Total expenditures on beef products from Australia by country. ........._..._.. ............... .67

2-21 Total expenditures on beef products from New Zealand by country. ........._..... .............67

2-22 Total expenditures on beef products from the ROW by country ............... .............. .68

2-23 Pre-BSE ban trade scenarios for fresh/chilled and frozen beef in selected Asian
nations. .............. ...............70....

2-24 Post-BSE ban trade scenarios for fresh/chilled and frozen beef in selected Asian
nations ........... __..... ._ ...............71....

4-1 Schematic representation of the Armington trade system for beef products in selected
countries of the Pacific Rim region. ..........._ .....__ ...............94.

5-1 South Korean product demand parameters for beef imports (US-AU-NZ) ................... ..121

5-2 Japan product demand parameters for beef imports (US-AU-NZ). ............... ..............122

5-3 Taiwan and Hong Kong product demand parameters for beef imports (US-AU-NZ). ...123

6-1 Estimated fresh U.S beef product demands across time in selected Asian markets........129

6-2 Estimated frozen U. S beef product demands across time in selected Asian markets......130

6-3 Estimated fresh Australian beef product demands across time in selected Asian
markets ................. ...............131................

6-4 Estimated frozen Australian beef product demands across time in selected Asian
markets ................. ...............132................

6-5 Estimated fresh New Zealand beef product demands across time in selected Asian
markets. ........... _. .... __ ...............133..

6-6 Estimated frozen New Zealand beef product demands across time in selected Asian
markets. ........... _. .... __ ...............134..

6-7 South Korea beef market demands and price elasticities in 1998, 2002, and 2006.........136

6-8 Japan beef market demands and price elasticities in 1998, 2002, and 2006. ........._........137

6-9 Taiwan beef market demands and price elasticities in 1998, 2002, and 2006. ........._......139

6-10 Hong Kong beef market demands and price elasticities in 1998, 2002, and 2006..........140

6-11 Pre-B SE market distribution of fresh and frozen beef products. ................ ................. 141

6-12 Post-B SE market distribution of fresh and frozen beef products. ................ ................143










ZRESETZ; DO ADJ WRP = .90 TO 1.10 BY .025;
SET SIMVAR = SIMVAR+1;
SET .X=1; SET SDI2=1; SET SDX2=1; ZBSIMZ;
ENDDO; ENDDOT;

ZRESETZ; SET SIMVAR = 1; SET SDI3=1; SET SDX2=1; ZBSIMZ; ? AU
DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11;
ZRESETZ; DO ADJ WRP = .90 TO 1.10 BY .025;
SET SIMVAR = SIMVAR+1;
SET .X=1; SET SDI3=1; SET SDX2=1; ZBSIMZ;
ENDDO; ENDDOT;

ZRESETZ; SET SIMVAR = 1; SET SDX2=1; ZBSIMZ; ? AU
DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11;
ZRESETZ; DO ADJ WRP = .90 TO 1.10 BY .025;
SET SIMVAR = SIMVAR+1;
SET .X=1; SET SDX2=1; ZBSIMZ;
ENDDO; ENDDOT;


STARTING THE BEEF TRADE SIMULATOR
SIM #20 RELATIVE PRICE ADJUSTMENTS AND TIME FOR NZ

SET SIMNUM = 20;
ZRESETZ; SET SIMVAR = 1; SET SDIl=1; SET SDX3=1; ZBSIMZ; ? NZ
DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11;
ZRESETZ; DO ADJ WRP = .90 TO 1.10 BY .025;
SET SIMVAR = SIMVAR+1;
SET .X=1; SET SDIl=1; SET SDX3=1; ZBSIMZ;
ENDDO; ENDDOT;

ZRESETZ; SET SIMVAR = 1; SET SDI2=1; SET SDX3=1; ZBSIMZ; ? NZ
DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11;
ZRESETZ; DO ADJ WRP = .90 TO 1.10 BY .025;
SET SIMVAR = SIMVAR+1;
SET .X=1; SET SDI2=1; SET SDX3=1; ZBSIMZ;
ENDDO; ENDDOT;

ZRESETZ; SET SIMVAR = 1; SET SDI3=1; SET SDX3=1; ZBSIMZ; ? NZ
DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11;
ZRESETZ; DO ADJ WRP = .90 TO 1.10 BY .025;
SET SIMVAR = SIMVAR+1;
SET .X= 1; SET SDI3= 1; SET SDX3= 1; ZB SIMZ;
ENDDO; ENDDOT;

ZRESETZ; SET SIMVAR = 1; SET SDX3=1; ZBSIMZ; ? NZ
DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11;
ZRESETZ; DO ADJ WRP = .90 TO 1.10 BY .025;
SET SIMVAR = SIMVAR+1;
SET .X=1; SET SDX3=1; ZBSIMZ;
ENDDO; ENDDOT;









country and time period. Using Chapter 4's notation, the equation used to estimate the

corresponding product demands can be written in its simple form as follows:

In WQQ = In(Xij) = Goij + 8tij (In(Pii) In(Pi.)) + 921j In(Xi.). (5-1)

Each of the theta (0kij) was estimated separately and is some function of ai., aii and Pii as

set forth in the previous chapter. Once estimates of ai., aii and pii are known, then the 9's are

determined. Equation 5-1 is estimated with a pooled-time-series-cross-sectional estimator and is

double-log form. After their calculation, these parameters were incorporated into Equation 5-1 to

obtain the corresponding product demands. Note that for a given set of 8's, one can easily show

the corresponding elasticities. The resulting model had one equation for each country and type of

beef product (Fresh/Chilled or Frozen). The obj ective is to compare the results from each

country and draw conclusions regarding the behavior of their parameters in terms of their values,

signs, significance, and consistency with the economic theory.

The reminder of this chapter will discuss the result of the non-linear single equation used

to estimate the parameters of the product demand equations. All model estimates were obtained

using TSP software, as well as the statistics indicating the performance of the model. The

magnitudes of the estimated parameters or elasticities will be discussed and contrasted with the

economic theory as they indicate the responsiveness of the dependent variable (product demand)

to the explanatory variables. In general, the results of the estimation are good in that they make

economic sense and generally correspond with what one would expect given the trading situation

described in earlier chapters.

Model Results

The following section describes the parameter estimates for the product demands functions

(Xii) or as defined in the TSP programming, the WQQ dependent variable. As presented in










Mutondo, J. E., and S. Henneberry. "Competitiveness of U.S. Meats in Japan and South
Korea: A Source Differentiated Market Study." Paper presented at the American
Agricultural Economics Association, AAEA, Portland, OR, July 29- August 1,
2007.

Nicholson, W. M~icroeconontic Theory: Basic Principles and Extensions. Eighth edition.
USA. South-Western College/Thomson Learning, 2002.

Peterson, H. H., and Y. J. Chen. "The Impact of BSE on Japanese Retail Meat Demand."
Agribusiness 21, No. 3 (2005): 313-27.

Piggott N. E., J. A. Chalfant, J. M. Alston and G. R. Griffith. "Demand Response to
Advertising in the Australian Meat Industry." American Journal ofAgricultural
Economics 78, No. 2. (1996): 268-279.

Poulin, D., and A. K. Boane. "Mad Cow Disease and Beef Trade." edited by Yvan
Gervais, 7. Ottawa, Canada: Statistics Canada, 2003.

Preckel, P. V., J. Cranfield, and T. W. Hertel. "Implicit Additive Preferences: A Further
Generalization of the CES." West Lafayette, Indiana: Dept. of Agricultural
Economics, Purdue University, 2005.

Regmi, A. "Changing Structure of Global Food Consumption and Trade." Agricultural
Economic Report WRS-01-1.Washington DC, 2001.

Reinert, K. A., and D. W. Roland-Holst. "Armington Elasticities for United States
Manufacturing Sectors." Journal ofPolicy Modeling 14, No. 5 (1992): 631-39.

Roig-Franzia, M. "A Culinary and Cultural Staple in Crisis." The Washington Post,
Saturday, January 27, 2007.

Saito, M. "Armington Elasticities in Intermediate Inputs Trade: A Problem in Using
Multilateral Trade Data." Canadian Journal ofEconontics Revue Canadienne
d'econontique 37, No. 4 (2004): 1097-117.

Sato, K. "A Two-Level Constant-Elasticity-of- Sub stitution Production Function." The
Review of Econontic Studies 34, No. 2 (1967): 201-18.

Satyanarayana, V., and D. Johnson. "Credit Guarantee Programs and U.S. Markets Share
in Selected Wheat Import Markets." Agricultural Econontics Report, No. 13.
Fargo ND, 1998.

Shiells, C. R., and K. A. Reinert. "Armington Models and Terms-of-Trade Effects: Some
Econometric Evidence for North America." The Canadian Journal ofEconontics,
Revue Canadienne d'Econontique 26, No. 2 (1993): 299-316.









CHAPTER 7
SUMMARY AND CONCLUSIONS

This dissertation focuses on the economic impact of food safety issues, in particular the

Bovine Spongiform Encephalopathy (BSE), on four of the most important international beef

markets and its repercussion on U.S. beef exports. Private and governmental efforts to re-gain

market access and promote the attributes of U. S. beef products in terms of quality and safety

have captured the attention of this study looking at the efficacy of these campaigns. This chapter

serves as a brief summary of the findings of this research, where conclusion and implications are

discussed.

To review briefly the dissertation, Chapters 1 and 2 examine the international beef market

and discuss the importance of agricultural trade and beef promotions. The literature review is

presented in Chapter 3, where previous research on beef demand, food safety, trade, and

promotions are discussed, as well as previous applications of the Armington model. The

theoretical basis of the model is presented and developed in Chapter 4, where the specific CRES

functional representation, along with a non-linear trade model, is defined to measure the

variations on product demand across time. In Chapter 5, the results of the econometric estimation

are presented and discussed. Figures of the actual data observed along with tables with the

parameter coefficients predicted by the model and the statistics indicating the performance of the

model are included. Chapter 6 presents the simulation results and their implications for beef

trade. Simulations were carried out on product demand and market share functional relationships

for each of the beef-importing markets.

All product demands calculated using the estimated parameters from Chapter 5 and

simulated in Chapter 6 show that model's specifications capture the general direction taken by

the actual data presented in Chapter 2. In addition to that, Chapter 5 and 6 include a number of









In the case of ag.~ both the differential intercept and dummy coefficients are statistically

insignificant for all parameters; however, the signs of the parameters for Korea and Japan are

negative indicating an inverse relationship between these estimates and the parameter ag..

With respect to the aii estimates for U. S. products in South Korea, the results show

negative and insignificant values, while all exporting countries show positive and significant

estimates. In the particular case of the United States, the estimates are positive and very

significant when comparing with the mean but at this point it does not indicate either the size or

importance of the observed variable or its impact on the elasticities. However, these results could

indicate a tendency for beef importing countries to become somewhat more responsive to the

origin of the product and changes in consumers' perceptions of imported beef. Dummy variables

for Japan, Taiwan, and Hong Kong were initially included in the model to see whether these

countries had different impacts on ai. and ail parameters. However, these variables were all

statistically insignificant and showed clear symptoms of multicollinearity; therefore they were

not reported.

The type of beef, defined by the parameter DF l, is statistically significant, but again the

effect of this variable will be discussed in Chapter 6. However, it is possible to assume that the

form of the product (fresh or frozen) has a substantial indirect impact on aii, which should be

translated in the elasticities between fresh and frozen beef products. On the other hand, estimates

for beef promotions (GGl) show a negative and insignificant value with respect to the mean (first

year), which somehow refutes initial expectations of the effect of this variable on the demand for

differentiated beef products.

The Pii can be interpreted as influencing the share that the exporter j has in the ith import

market when Pii equals Pi Theoretically, Pii must be positive, should sum "on average" to









measure the impact of restrictive policies on beef trade, the size of those impacts largely depends

on the magnitude of the elasticity (McDaniel and Balistreri 2002).

The methodology suggested for this research distinguishes products by place of origin and

uses a Constant Ratio of Elasticity of Substitution (CRES) functional form to bring prices,

quantity, and other factors into the model, in order to measure trade allocation and demand

preferences for beef products. In this model, total import demand for beef will be determined for

each country and then it will be independently allocated among competing sources of supply

(Sparks and Ward 1992). However, this model represents a non-linear system with a set of

equations that are non-linear in the parameters and their corresponding restrictions within each

set of equations.

Econometric procedures will be used to estimate product demands and market share

equations to assure consistency results. The resulting model and its estimates will be used to

develop a sensitivity analysis to examine the effects of changes in selected factors on beef trade

and consumption.

Armington Trade Model

In the international food market, countries are assumed to behave competitively and

commodities like beef are usually seen as homogeneous goods. That is, similar products from

different origins are considered perfect substitutes and the corresponding price ratios are constant

in that particular market. However, reality shows that consumers tend to behave and adopt food

products based on well defined personal preferences constructed upon perceptions of the

attributes of the product which, at the end, make competing products behave as imperfect

sub stitutes.



3 The assumption of perfect substitution among products implies the presence of infinite elasticities of substitution.










Import prices were regressed on world product prices for all of the years of the data and the

estimated relationship was then used to estimate an import product price for a year when it was

missing. Finally, missing values in the original data set were approximated through linear

interpolation between existing quantities (Chiang 1997). Manipulations were carried out on these

data sets, after which they were merged by year and by country to create the final data set. These

manipulations resulted in a data set that shows space as well as time dimensions, which are

typical characteristics of panel data sets where the same cross-sectional unit (such as a country)

is surveyed over time. Thus, import quantities are expressed in thousands of metric tons and

values in thousands of U. S. dollars per metric ton. Both quantities and values are used to derive

import prices. Given the nature of the data, prices are treated as exogenous variables and already

include the corresponding tariff schedules for beef products originated outside the country.

For any given year, the data on beef trade represent a cross-sectional sample. Then, in the

case of fresh beef, for any given year there are 16 observations and 1 1 time series observations;

thus there is in all (16xl l) = 176 (pooled) observations on fresh beef (Guj arati 2003). Thus,

combining cross-sectional and time series data into a pooling data set gives more degrees of

freedom, reduces or eliminates the issue of multicollinearity among variables, and minimizes the

bias that might result from aggregation.

The rest of the data consist of country-specific statistics such as GDP, CPI, population

levels, total food imports, and country domestic beef production and consumption levels that

were obtained from the UN and FAO data bank and arranged in a country by country basis.

Product Demand Equation

The product demand equation WQQ = Xii is calculated assuming that there is a product

differentiation among beef products and that the rate of substitutability is no constant across










from early 2004 through 2007. Th U.S. market participation decreased to 7% while Australia and

New Zealand kept a modest 6% of the market respectively.

Expenditures on Beef by Country of Origin

In this section, expenditure levels on imported beef are used to analyze the position and

relevance of each exporting nation. Figures 2-19 through 2-22 show related statistics in terms of

U.S. dollars per capital for both the pre- and post-BSE announcements.

Expenditures on U.S. Beef ($ per capital)
40.00-
Importing Countries 36.90
I Kilr.- I Tapan I Taiwn-n TT~nne K~nn

30.00 -



20.00-

15.00-

10.00 -90







0.00 1
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
BHong Kong 303 559 577 586 582 844 781 746 928 780 005 088 378
Taiwan 164 2 15 259 297 266 287 245 231 280 191 1 17 274 337
Japan 1374 1565 1100 1196 1035 1193 1074 698 770 387 000 004 102
Korea 720 674 637 373 598 1051 680 1252 17 12 200 008 000 088

Figure 2-19. Total expenditures on beef products from the U.S. by country. Source: USDA-FAS.
February 2008.

In the case of expenditures on U.S. beef products, Figure 2-19 clearly illustrates that

Japanese and Korean consumers have a manifest preference for grain-feed U. S. beef products.

Before the BSE announcements, these two countries were considered among the top three

destinations for U.S. beef. During the year 2000, both Japan and Korea increased their

expenditures on American beef by almost 14% and 43% due to the steady economic growth in










In 1995, the next largest share was for the European Union (EU) region with 18.5%

followed by East Asia (10.3 %), Oceania (4.8%), Asia (2.6%), and the rest of the regions of the

world with 13%. In 2007, however, East Asia (15.7%) surpassed the EU (14.7%) as the second

largest producing region of the world followed by Oceania, Asia, and the rest of the world. With

the exception of the EU and some small producing regions included in the ROW category, all

other regions of the world show important advances in terms of quantity produced which

guarantees a healthy supply and demonstrates the increasing importance of this commodity in

people's diet around the globe.

Table 2-1. Compared world production levels and trade from 1995 to 2007
1995 2007 2007/1995
Region 1000 MT % of World 1000 MT % of World Ratio
S. America 9,975 0.206 13,235 0.243 1.327
N. America 14,363 0.296 15,514 0.285 1.080
Asia* 1,239 0.026 2,724 0.050 2.199
East Asia 5,003 0.103 8,574 0.157 1.714
European U 8,985 0.185 8,000 0.147 0.890
Oceania 2,347 0.048 2,921 0.054 1.245
ROW 6,626 0.137 3,521 0.065 0.531
World 48,538 1.000 54,489 1.000 1.123
Source: FAS-USDA. January 2008. Asia* represents values of South and Southeast Asia.

Table 2-1 shows the imports to exports ratio 2007/1995, which is interpreted as a growth

rate indicator for production levels. Thus, production grew globally at a rate of 1.123 from 1995

to 2007 or in other words, the 2007 production level was 1.123 times the 1995 level. The Asian

region shows the largest improvement in terms of production and that traditional beef-producing

regions, like North America and the EU, have been surpassed by competing regions, like South

America and Oceania showing an important shift in the world beef supply. Using Figure 2-2 to

support the discussion, production levels have increased at a rate of 1.2, 0.71, 0.33, and 0.25

times in Asia, East Asia, South America, and Oceania respectively, while the North American

region, the EU, and the rest of the world (ROW) have shown small or negative increments during

the same period.










axii_ Pi fi.1+p i
= ~ =, / 1 2..n (A-3)


Solving Equation A-3 for X .,


Xg. X ,i=12,..,n.(A-4)


Using this function, Equation A-2 can be expressed as a relation between Xi. and Xii, and their

corresponding prices. Rearranging terms,

1 1
= i=,2,..,n(A-5)


from which it follows that the elasticity of substitution between Xii and any other product


competing in the same market is equal to the constant = ai, which denotes the elasticity of
1+p

substitution (Varian 1992)4. Thus, Equation A-4 can be expressed as follows:


Xg. PI = =12.., (A-6)


where ai (0 I ai I oo) defines the degree of substitution among products. For example, at the

limiting cases if ai = 0, the products are perfect complements or if ai = oo, the products are

perfect substitutes (Armington 1969). Substituting Equation A-2 into Equation A-6, and writing

pi in terms of ai we have:










4 The elasticity of substitution (ES) between Xi and Xi is:












Volume of Trade in 1995, 2003, and 2007 (1000 MT)


3500


3000-
Imports (Dark) and Exports (Light)


2500-


2000-


1500


1000-







South North Asia* East European Oceania ROW

America America Asia Union
Region




Figure 2-3. Pre and post- BSE beef trade levels by region. Source: USDA-FAS. February 2008.

Exporting regions like North America, South America, Oceania, and Asia revealed


significant levels of growth in their export markets, although in the case of the North America

this trend drastically changed after 2003. Other regions such as Europe and the ROW presented


an opposite trend, which is interpreted as a loss in market participation due to the negative


impact of food safety issues affecting their supply chain, like in the case of the BSE epidemic in

the United Kingdom with almost 1,000 new cases per week January 1993 (OIE 2007). Figure 2-3


presents the import and export quantities for years under consideration. During the year 1995, the


maj or import regions were North America and East Asia followed by the European Union. These

three regions imported 1257, 1295, and 494 thousands of metric tons respectively representing

29.5% 30.4%, and 1 1.6% of the world imports respectively; during the year 2003, 2048


(39.6%), 1495 (28.9%), and 549 (10.6%) thousands of metric tons respectively; and in the year
1Percentage share of world imports/ exports for beef products.









set on promotions expenditures, which turn out to be one the most important limitations in this

study due to the lack of applicability at the country level.

Using a combination of logistic and exponential functions, the aig, aii and Pity parameters

were obtained and applied to the product demand functional form specified in Chapter 4. The

resulting econometric model represents a non-linear system that shows estimates and supportive

statistics indicating that this trade system fits well the trade data (R-square = 0.763) and explain

the economics behind the international beef trade in the region. The resulting Durbin Watson

statistic values indicate little serial correlation among the parameters of the model. With the

exception of the econometric results on international promotions, the resulting trade model's

estimates resemble the economic trends of the beef trade in the region.

Because product demands are a function of product' s price relative to the average market

price and the size of the beef market in a particular country, any changes in market demands also

affect product demands. Thus, for each market, two set of simulations were conducted on the

product demands, where the trade volumes from each of the exporting countries was simulated

over time (1997 to 2007), relative prices were varied from 10% below to 10% above of their base

or mean price (US: $1.34 per MT; AU: $0.94 per MT; and NZ: $1.02 per MT), and market

demands were varied from 20% below to 20% above of their base or average quantity demanded

in each market (KR: 99.58 MT; JP: 286.55 MT; TW: 32.19 MT; HK: 25.26 MT). All product

demands show a negative slope in terms of price variations, that is as the average or mean price

increases, the quantity of beef products demanded decreases. For the simulations where the

average price was varied, the product demands for South Korea, Japan, Taiwan, and Hong Kong

reflect a negative relationship between price and quantity demanded as expected by the

economic theory.










Buzby, J. C. "International Trade and Food Safety: Economic Theory and Case Studies."
Agricultural Economic Report. No. 828.Washington DC, 2003.

Cameron, A. C., and F. A. G. Windmeij er. "An R-Squared Measure of Goodness of Fit
for Some Common Nonlinear Regression Models." Journal ofEconometrics 77,
No. 2 (1997): 329-42.

Capps, O., R. Tsai, R. Kirby, and G.W. Williams. "A Comparison of Demands for Meat
Products in the Pacific Rim Region." Journal ofAgricultural and Resource
Economics 19, No. 1 (1994): 210 -24.

Carwell, R. T. "The Importance of Dynamics in Commodity Advertizing Demand
Systems." Ph.D. Dissertation, University of Saskatchewan, 2005.

Chiang, A. C. Fundamental Methods~~d~~dd~dd~ of2athematical Economics. Third edition. New
York, NY: McGraw-Hill/Irwin, 1987.

Codron, J. M., J. A. Sterns, and T. Reardon. "Strategic Choices in Produce Marketing:
Issues of Compatible Use and Exclusion Cost." Journal ofFoodDistribution
Research Volume 34, No. 03 (2003): 1-12.

Coffey, B., J. Mintert, S. Fox, T. Schoeder, and L. Valentin. "The Economic Impact of
BSE on the U.S. Beef Industry: Product Value Losses, Regulatory Cost, and
Consumer Reactions." In Kansas State University Agricultural Experiment
Station and' Cooperative Extension Service, 64: Kansas State University, 2005.

Cong, L.T., H. M. Kaiser, and W. Tomek. "Export Promotion and Import Demand for Us
Red Meat in Selected Pacific Rim Countries." Agribusiness 14, No. 2 (1998): 95-
105.

Davis, G. C., and N. C. Kruse. "Consistent Estimation of Armington Demand Models."
American Journal ofAgricultural Economics 75, No. 3 (1993): 719-23.

Devadoss, S., D. W. Holland, L. Stodick, and J. Ghosh. "A General Equilibrium
Analysis of Foreign and Domestic Demand Shocks Arising from Mad Cow
Disease in the United States." Journal ofAgricultural and Resource Economics
31, No. 2 (2006): 441-53.

Dyck, J.H., and K.E. Nelson. Structure of the Global Markets for Meat." In Agriculture
Information Bulletin, edited by Economic Research Service (ERS) Market and
Trade Economics Division, 24. Washington, DC: United States Department of
Agriculture, 2003.

Femenia, F., and A. Gohin. "Estimating Price Elasticities of Food Trade Functions: How
Relevant Is the Gravity Approach? ". 7RADEAG Working Papers, edited by INRA
France, 37. 2007.









Within this framework, there are several relationships operating to equilibrate the demand

and supply of beef products and their most general functional representations are discussed

below. Since this model assumes that import demand must equal supply we have an equilibrium

condition from which two restrictions can be obtained:

Xi. = Ci Xii, and Xi = Cixi X, (4-4)

Differences in import prices among demanding (importing) countries are driven by

asymmetries in cost of insurance and freight, quality of products, market structures, non-tariff

barriers to trade, and information. Non-tariff barriers (or NTB) are mechanisms that impede the

flow of international goods by imposing unilateral and arbitrary restraints to normal trade. The

most common NTB are import quotas, exchange controls, subsidies, boycotts or bans, technical

barriers, and voluntary restraints among others.

A tariff is a tax levied by the government in country i on goods imported into that country

from country j (or import duty). The main obj ective of a tariff rate is to make country i products

more competitive by increasing the price at which the goods from country j are sold. The effect

of tariff rates on prices of products from country j in market country i is quantified as:

Pi1 = (1 + Tr y) Cr y (4-5)

where Pii is the market price of the products, Tii is the cost of tariff expressed in percentage

terms, and Cii represents import prices (CIF) that are a function of FOB prices and a proxy

variable (V) created to capture the variation in cost over time:

Cii = f (Fiy, V). (4-6)

Then, the average price paid for beef in a given importing country is defined as follows:


2 A product produced and consumed domestically does not incur in cost associated with shipping and barriers to
entry. Consequently, this price is assumed to be equal to the average of all export prices in a particular market
(Sparks and Ward 1992).









produces exactly y units of outputs) and differs between factors (Varian 1992). The choice of the

CRES rather than the CES is justified because it allows the estimation of parameters without

restricting the substitutability to be the same between every product, and consequently fully

exploiting the attributes of the Armington's theoretical framework (Sparks and Ward 1992).

The CRES technical relationship determines the functional nature of the product demands

and market shares from competing supply regions. For this study, we assume that the CRES

function permits variations of the elasticities of substitution among competing beef products in a

market i to vary by a constant proportion, thus allowing for differences in substitutability

between products of different origin within a market. This assumption increases the flexibility of

the model by reducing the number of parameters to be estimated.

Following earlier specifications used by Mukerji (1963), Hanoch (1971), and in particular

Sparks and Ward (1992), the derivation of the product demand functional form from the CRES

technical relationship is developed as follows:




where the Xi.is the product import demand for beef in region i in period t, and Xii is the ith

market' s import demand for beef from country j. The Pii is the share or distribution parameter

and is a function of the exogenous variables affecting the system (e.g., market structure), while

the parameters agyi and ati~are substitution parameters that are functions of variables like generic

advertising, country origin, and information.

The degree of substitutability between imported beef from different sources of supply is

captured by the CRES. Consequently, the higher the value of this parameter, the closer the

degree of substitution. In other words, a high value of this parameter means that beef products

from all sources are considered by consumers to be virtually identical. Conversely, a low value






























-
U S.Fresh Jeef Yolume
-
-S Korea Japan- Taiwan -Hong Kong
-

-

-

-



-

-

-

-

-

-


type of beef (see Table 1-2 for a full description of the restrictions). However, during the year



2006, a gradual re-opening of trade was translated into an increased product demand for U. S.


beef products across countries, in particular in Japan where consumers have shown a marked



preference for imported, grain-feed, beef from the United States.


11.41


1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
Hong Kog14 76 12 27 12 54 12 13 12 18 11 64 11 35 3 82 () (3 () ( 3 71
Tanvan 18 8 15 63 15 96 15 44 15 51 14 82 14 45 4 87 () (3 0)11 4 72
Japan 74 96 62 31 63 66 61 58 61 84 59 11 57 61 19 4 0)13 0)45 18 83
S Korea 129 (3 84 93 89 14 82 68 83 48 75 38 71 11 6 (5 () ( 5 66

Year



Figure 6-1. Estimated fresh U.S beef product demands across time in selected Asian markets.


As of September 2007, the increase in U. S. product demand represented 32% of the pre-



BSE levels; and market prediction by the USDA, Foreign Agricultural Service and the US1VEF



showed a positive trend in the volume of beef exported, as well as in the value of export to Japan



for the incoming years.



In the case of South Korea, the situation is far more complex and driven by policy rather



than market demand. As previously mentioned, South Korea is the largest importer/consumer of



beef in the region and even though the complete ban on U. S. beef has been lifted by the


150

140

130

120

110

100

90

80

70

60

50

40

30

20









































-EastAsia -SouthAsia -SoutheastAsia














I I I i I I I I I I I I I I I I I I I I I I I I I I


increases in demand especially in Asian-Pacific markets where the explosive economic


expansion has contributed to an increase per capital income, encouraging reductions in trade

restrictions and a shift in dietary preferences for animal-source proteins (Capps et al. 1994).

Economic development brings with it increased household income and consequently

higher utility levels. With increased incomes, households can purchase more superior quality

foods, such as beef. As in the cases of Japan and Korea, the growth in meat consumption pushes

up prices of domestic beef, and increases the possibility of a market for imported beef. The

importance of dietary changes to beef imports is shown by the case of Pacific Rim countries

where import growth reflected the rapid increase in beef consumption that occurred there

between 1980 and 1995, when consumption per capital increased almost six times. Consumption

increased faster than production, and imported beef supplied the difference.


Domestic Consumption (1000 MT)


10,000
9,000
8,000
7,000
6,000
5,000
4,000


1980 1985 1990 1995 2000 2005
Years


Figure 1-1. Domestic consumption of beef products in Asian regions. Source: FAS-USDA.
November 2007.

While meat consumption growth appears to have leveled off since the late 1990s in

western developed regions (e.g., Europe and the United States), meat consumption is currently

growing in maj or parts of East, Southeast, and South Asia as economic expansion proceeds









CHAPTER 6
SIMULATION ANALYSIS

Introduction

The previous chapter concentrated on a general discussion of the econometric estimates

obtained using the Armington model and the CRES assumption about the elasticity of

substitution. In this chapter, a more detailed analysis of the significance of the different

parameters and sub-parameters estimates from Chapter 5 will be presented. Thus, the obj ective is

to simulate the model specified in Chapter 4 in order to evaluate the effect of changes in relative

prices, quantities demanded across time, and commodity promotions on product demands across

four beef markets in the Pacific Rim region. Under different price, promotion, and trade flow

scenarios, this section will compare the product demands and market shares relative to variations

in the mean or average price or quantity.

The simulation process involves a set of preliminary steps that were conducted using the

TSP program, where all the variables of the model are set to the mean level (all dummy variables

are set to zero) in order to compare the changes in the dependent variable when one of the

independent variables varies, and all other input parameters are held constant at their mean value.

The following step is then to predict product, market share demands, and elasticities using

calculated relative prices, commodity promotions, and quantities at their mean values. When

calculating these values for each of the variables, their respective means were included and

multiplied then by an adjusted parameter set to one for simulation purposes (i.e., relative price

= MEANRP*AD _RP .

The simulations measured the strength of the relationship between the independent and

dependent variables. Tables 6-1 through 6-4 illustrate the base values estimated for each beef

exporter in each beef market.









































Xz;(Product Demand) X~j(Product Demand) X,i(Product Demand) X4;(Product Demand)
Xzy/X,.(Mkt. Shares) X~j/X2.(Mkt. Shares) X,i/X,.(Mkt. Shares) X4;/X,.(Mkt. Shares)



Figure 4-1. Schematic representation of the Armington trade system for beef products in selected
countries of the Pacific Rim region.

Prices are the crucial link to allocate products and their impact on product demands, and

market shares are presented in Equations 4-17 and 4-19 respectively and are considered

exogenous variables. Within this framework, there are several relationships operating to

equilibrate product demand and supply. The resulting econometric model represents a non-linear

system and parameters restrictions within each set of equations. Econometric procedures will be

used to estimate product demands equations and assure the consistency and un-biased












Equations 4-18 and 4-20 specify the functional forms of the product demand and market

share equations imposed by the CRES technical relationship on the markets and hence non-linear

estimation techniques must be used to quantify the parameters (Sparks and Ward 1992)6. Using

the log function to express the Armington model's product demand functions, Equation 4-18, the

following form is obtained:

In(Xii) = xJi (In ai. In aii In #g y) + us; (In Pii InPg.)






In(Xii) = i-1) (In ag. In a1;-In #g ) + (i1) (In Pg; In





If ai, ail, and pii are in turn functions of events and changes, then clearly the product

demand is highly non-linear and must be solved using a non-linear technique. Since the obj ective

of the model is to estimate coefficients of demand, the first estimate is of the differing and

substituting country intercept and slope coefficients using exponential and logistic function, and

the results are incorporated into Equation 4-21 to obtain the corresponding estimates of the

model. Since the double-log functional form is used to estimate each country product demand,

the final coefficients were the estimated elasticities of the corresponding variables. Therefore, for

convenience and to incorporate the impact of promotions, food safety and other factors into the

analysis, the final product demands Equation 4-21 can be written as follows:



6 See Chiang 1987 for detailed explanation of the log function properties.












Percentage of Variation
1.60-
o% Change

1.20
1.20-



0.80 -07





00.08
0.005

0-0.1



-0.40-
-0.47


-0.80
S. America N. America Asia* East Asia European U Oceania ROW

Region


Figure 2-2. Relative variation in growth rate of production (2007/1995). Source: FAS-USDA.
January 2008.


Beef Trade Patterns


This section presents a general description of each region in terms of trade, trends, and


ratios between imports and exports volumes. Figure 2-3 and Table 2-2 are used to compare levels


of trade and explain the significance of each region in terms of imports and exports. The dataset


shows that beef trade has expanded over time and that market shares between regions have


changed. Rising demand for meat products in different parts of the world and increasing


concerns about the safety of the beef supply chain have prompted nations to re-define trade


alliances in order to satisfy beef consumers and ensure the quality and safety of the products


being traded. Using trade statistics for the seven regions previously defined, it is possible to


identify trade trends for the years before and after the BSE outbreak in December 2003.










SET SDX2=1; ZBSIMZ;
ENDDO; SET H= SIMVAR;

ZRESETZ; SET SIMVAR = 1; SET SDX3=1; ZBSIMZ; ? NEW ZEALAND;
DO ADJ FOREIGN = .70 TO 1.30 BY .05;
SET SIMVAR = SIMVAR+1;
SET SDX3=1; ZBSIMZ;
ENDDO; SET H= SIMVAR;


STARTING THE BEEF TRADE SIMULATOR
SIM #18 RELATIVE PRICE ADJUSTMENTS AND TIME FOR THE US

SET SIMNUM = 18;
ZRESETZ; SET SIMVAR = 1; SET SDIl=1; SET SDX1=1; ZBSIMZ; ? US
DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11;
ZRESETZ; DO ADJ WRP = .90 TO 1.10 BY .025;
SET SIMVAR = SIMVAR+1;
SET .X=1; SET SDIl=1; SET SDX1=1; ZBSIMZ;
ENDDO; ENDDOT;

ZRESETZ; SET SIMVAR = 1; SET SDI2=1; SET SDX1=1; ZBSIMZ; ? US
DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11;
ZRESETZ; DO ADJ WRP = .90 TO 1.10 BY .025;
SET SIMVAR = SIMVAR+1;
SET .X=1; SET SDI2=1; SET SDX1=1; ZBSIMZ;
ENDDO; ENDDOT;

ZRESETZ; SET SIMVAR = 1; SET SDI3=1; SET SDX1=1; ZBSIMZ; ? US
DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11;
ZRESETZ; DO ADJ WRP = .90 TO 1.10 BY .025;
SET SIMVAR = SIMVAR+1;
SET .X=1; SET SDI3=1; SET SDX1=1; ZBSIMZ;
ENDDO; ENDDOT;

ZRESETZ; SET SIMVAR = 1; SET SDX1=1; ZBSIMZ; ? US
DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11;
ZRESETZ; DO ADJ WRP = .90 TO 1.10 BY .025;
SET SIMVAR = SIMVAR+1;
SET .X=1; SET SDX1=1; ZBSIMZ;
ENDDO; ENDDOT;


STARTING THE BEEF TRADE SIMULATOR
SIM #19 RELATIVE PRICE ADJUSTMENTS AND TIME FOR AU

SET SIMNUM = 19;
ZRESETZ; SET SIMVAR = 1; SET SDIl=1; SET SDX2=1; ZBSIMZ; ? AU
DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11;
ZRESETZ; DO ADJ WRP = .90 TO 1.10 BY .025;
SET SIMVAR = SIMVAR+1;
SET .X=1; SET SDIl=1; SET SDX2=1; ZBSIMZ;
ENDDO; ENDDOT;

ZRESETZ; SET SIMVAR = 1; SET SDI2=1; SET SDX2=1; ZBSIMZ; ? AU
DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11;









demand for beef products. Further, without much loss in generality the previous Equations 4-29,

4-30, and 4-31 can be also written as:

ai. = K(Z), (4-32)

agyi = G(Z) and (4-33)

Pii = H(Z) (4-34)

assuming that 0 I ai., aii I 1, and Pii 2 0; and that the (Z) variable represents a set of dummies

and variables representing promotions and the type of beef. In order to take into account the

"uniqueness" effect of each parameter, this study suggests the use of a logistic regression to

capture the impact of each of the factors affecting ail, asi and an exponential form to estimate the

effect of the variables affecting PgJ. In both cases, a set of dummy variables are introduced in the

model to understand fully the effect of each parameter. For the sake of clarification and without

distracting the attention from the main model specification, the logistic function can be written as

F(z) =; and F(z) = e-zi, (4-35)

where F(z) represents each of the parameters ail, ai. and Pii ; and the "input" (zi) indicates a

regression model containing the respective subset of variables and dummy variables representing

the market and product characteristics. The logistic function is useful because it can take as an

input, any value from negative infinity to positive infinity, whereas the output is confined to

values between 0 and 1. The variable (zi) represents the exposure to some set of trade factors,

while F(z) represents the likelihood of a particular parameter, given that set of factors. The

exponential function is used because F(z) in this case is always positive (a required assumption

for Pii parameters), although it gets arbitrarily close to zero. Thus, the parameter agyi, ai., and Pii,

can be written as:









international beef markets. According to this report, the economic significance of animal disease

outbreaks is also influenced by the degree of consumer response: fears that the disease can

spread to humans can lead to sharp drops in consumption as was reported in Pacific Rim nations.

Similarly, Peterson and Chen (2005) explored the impact of BSE on the Japanese retail

meat demand using a Rotterdam model that allowed beef product differentiation and found that

effectively, the scare affected the demand for domestic beef (wagyu beef), domestic dairy beef,

and imported beef from the United States and Australia. According to this research, Japanese

retail meat demand underwent a two-month transition period following the initial announcement

and continued to adjust subsequently until it reached a new state within five months of the BSE

announcement.

Devadoss et al. (2006) employed a general equilibrium model to analyze the economic

impact of the BSE outbreak on the U.S. cattle-and beef-related industries using different demand

scenarios in the domestic and international market. This study found that only if domestic

demand declines significantly will the economic hardship in the U.S. industry be very large.

Using a trade model that allows for imperfect substitution between goods, three different

scenarios were analyzed in which the decline in foreign demand was in all cases 90 % and the

decline in the U.S. market was 0%, 10%, and 25% respectively. Results showed that the impact

of the BSE outbreak even in the worst-case scenario was not as damaging as it was in Canada,

whose industry depends largely on the foreign market and exports prices dropped significantly.

In the case of the United States, since it only exports 10 % of its beef, reduction in foreign

demand did not have a long term effect on the domestic industry in general, although cattle

futures and beef cash prices declined 19% in the days following the BSE announcements.









demands for beef products in selected Asia's countries using a model of product differentiation

by country of origin and to estimate empirically the impact of U. S. beef promotions on re-

capturing pre-ban market levels. Based on Armington's specification (1969), the model for this

research assumes a two-stage budgeting allocation and a product's imperfect substitutability

between export sources. Hence, this study presumes that beef produced in the United States

presents unique attributes (i.e., superior quality and taste) and that foreign consumers can

perceive this difference from competing beef sources.

Departing from Armington's Constant Elasticity of Substitution (CES) specification, this

work proposes the use of a Constant Ratio Elasticity of Substitution (CRES) that allows

elaticities to vary proportionally while the substitutability is not necessarily identical. This

research includes eight countries divided into four exporting and four importing countries

respectively. Based on countries' import/export yearly data, the entire dataset includes almost

2000 observations with information recorded over time across selected countries during an 11-

year period (January 1995 to December 2006). Resulting product demand, market share, and

elasticity estimates determine the degree of substitution among competing beef products and

provide the coefficients to measure the economic impact of trade barriers and effectiveness of

market access and promotion policies in four maj or markets for U. S. beef: Japan, Republic of

Korea, Hong Kong, and Taiwan.












3 LITERATURE REVIEW .............. ...............73....


Introducti on ........._. ._...... .. ...... .. ........ ...............73.
International Beef Demand and Food Safety ...._.. ................ ............ ...... 7
Trade and Beef Markets Issues ................. ...............78......._ ....
International Beef Promotions ..............._ ...............81..._..._......
The Armington Trade Model ........._._._..... ..... ...............88...


4 EMPIRIRICAL TRADE MODEL FOR BEEF PRODUCTS ......___ .........._ ...........93


Introducti on ................ ... ........... ...............93......
Beef Trade Schematic Representation............... .............9
Adapting the Armington Model to Beef Trade ................. ...............95........... ..
The Constant Ratio Elasticity of Substitution .............. ...............100....
Analysis of Trade Variations ................ ...............107...............

5 EMPIRICAL RESULT S ................. ...............113...............


Introducti on ................. ...............113................
Data Set................. .... ..............11
Product Demand Equation ................. ...............114................
M odel Results ................. ...............115......... ......


6 SIMULATION ANALY SIS ................. ...............125................


Introducti on .................. .. ......... ...............125......
Total Product Demand Trend .............. ...............128....
Product Demand Simulations .............. ... ..... ..............13
Pre- and Post-BSE Market Distribution Simulations .............. ...............141....


7 SUMMARY AND CONCLUSIONS .............. ...............146....


APPENDIX


A CONSTANT ELASTICITY OF SUBSTITUTION (CES) ................ ................ ...._.152


B TSP PROGRAM: ARMINGTON SPECIFICATION AND SIMMULATIONS ................157


LIST OF REFERENCES ................. ...............176...............


BIOGRAPHICAL SKETCH ................. ...............183......... ......



































Republic of Taiwan
--Production -Im ports--Consumption



n 81 73 04 00 101 00 85 co 107 87 101 100 110
ts 75 o7 88 85 co 85 80 01 101 82 05 104 105
n 6 6 o 5 5 5 5 5 o 5 o 5







Hong Kong/ Republic of China
-Production -Im ports -Consum ption
-1


year considered. As in the Korean market, Japanese consumers have shown a strong preference

towards American beef, which is translated into higher price levels. Following the same price

trend, beef products from New Zealand have been priced above the average of U. S. and

international price levels showing an important increase, in particular, after the December 2003

BSE scare. This circumstance has positively affected the demand for beef from grass-feed cattle

since cattle produced under this condition are less likely affected by BSE.

Republic of Taiwan and Hong Kong

Total Beef Production, Imports, and Consum ption (1000 1\T)


1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
Consumption 93 80 00 82 80 05 03 101 105 102 100 111 114
Imports 05 50 51 03 71 78 78 87 02 88 05 o~ 100
Itoduction 28 21 15 10 18 17 15 14 13 14 14 14 14
Year

Figure 2-8. Taiwan and Honk Kong beef market indicators. Source: USDA-FAS. November
2007.

Figure 2-8 includes data from the Republic of Taiwan and Hong Kong given that both

countries reflect marked similarities in production, imports, and consumption trends. First, both


620
510
40
30
20
10
50


120
110
80


620
10
40
30
20
10
50










ZRESETZ; SET SIMVAR = H+1; SET .X=1; SET SDX1=1; ZBSIMZ;
ENDDOT; SET HH= SIMVAR;


STARTING THE BEEF TRADE SIMULATOR
SIM #4 AUSTRALIA TRADE OVER TIME TO COUNTIES

SET SIMNUM = 4;
ZRESETZ; SET SIMVAR = 1; SET SDIl=1; SET SDX2=1; ZBSIMZ; ? AUSTRALIA TO KOREA;
DOT(INDEX=H,CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10
SDYR11;
ZRESETZ; SET SIMVAR = H+1; SET .X=1; SET SDIl=1; SET SDX2=1; ZBSIMZ;
ENDDOT; SET HH= SIMVAR;

ZRESETZ; SET SIMVAR =0+1; SET SDI2=1; SET SDX2=1; ZBSIMZ; ? AUSTRALIA TO JAPAN;
DOT(INDEX=H,CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10
SDYR11;
ZRESETZ; SET SIMVAR = H+1; SET .X=1; SET SDI2=1; SET SDX2=1; ZBSIMZ;
ENDDOT; SET HH= SIMVAR;

ZRESETZ; SET SIMVAR =0+1; SET SDI3=1; SET SDX2=1; ZBSIMZ; ? AUSTRALIA TO TAWIN;
DOT(INDEX=H,CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10
SDYR11;
ZRESETZ; SET SIMVAR = H+1; SET .X=1; SET SDI3=1; SET SDX2=1; ZBSIMZ;
ENDDOT; SET HH= SIMVAR;

ZRESETZ; SET SIMVAR =0+1; SET SDX2=1; ZBSIMZ; ? AUSTRALIA TO HONG KONG;
DOT(INDEX=H,CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10
SDYR11;
ZRESETZ; SET SIMVAR = H+1; SET .X=1; SET SDX2=1; ZBSIMZ;
ENDDOT; SET HH= SIMVAR;


STARTING THE BEEF TRADE SIMULATOR
SIM #5 NEW ZEALAND TRADE OVER TIME TO COUNTIES

SET SIMNUM = 5;
ZRESETZ; SET SIMVAR = 1; SET SDIl=1; SET SDX3=1; ZBSIMZ; ? NEW ZEALAND TO KOREA;
DOT(INDEX=H,CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10
SDYR11;
ZRESETZ; SET SIMVAR = H+1; SET .X=1; SET SDIl=1; SET SDX3=1; ZBSIMZ;
ENDDOT; SET HH= SIMVAR;

ZRESETZ; SET SIMVAR =0+1; SET SDI2=1; SET SDX3=1; ZBSIMZ; ? NEW ZEALAND TO JAPAN;
DOT(INDEX=H,CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10
SDYR11;
ZRESETZ; SET SIMVAR = H+1; SET .X=1; SET SDI2=1; SET SDX3=1; ZBSIMZ;
ENDDOT; SET HH= SIMVAR;

ZRESETZ; SET SIMVAR =0+1; SET SDI3=1; SET SDX3=1; ZBSIMZ; ? NEW ZEALAND TO TAWIN;
DOT(INDEX=H,CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10
SDYR11;
ZRESETZ; SET SIMVAR = H+1; SET .X=1; SET SDI3=1; SET SDX3=1; ZBSIMZ;
ENDDOT; SET HH= SIMVAR;










production sector of "cooperative goods" with little or no branding history has become a more

sophisticated industry. In this contest, commodity advertising is used to influence market shares

and, where necessary, to increase the volume of information about product attributes demanded

by consumers (Ward and Lambert 1993). Agents, along the food supply chain, have learned how

to differentiate and increase the value of their products (e.g., live animals and meat products) in

order to build demand and increase financial profits for the firm. Advertising and promotions,

which often highlight production and marketing practices, are examples of the methods that

companies and individuals have utilized to set their product apart in order to meet the needs and

desires of specific consumer segments (Allen and Pierson 1993). The fresh beef market is a good

example of this practice where exporting countries and their agents have learned to differentiate

their beef cuts using the product' s particular attributes as central message in their marketing

campaigns (e.g., U.S. "We Care" and Australian "Clean and Safe" beef).

In markets such as Japan, the USMEF launched its "We Care" campaign to help rebuild

consumer confidence in U. S. beef using a range of advertising and consumer events. According

to the USMEF, the obj ective of the campaign was to counter the negative emotions generated by

the media by putting a "human face" on the U.S. industry. The campaign highlighted the

concerns of the entire U. S. beef industry in delivering safe and quality beef products for

consumers and retailers, and provided updated information on BSE prevention systems being

implemented in U.S. plants (USMEF, 2007).

Commodities in general, and in our case beef products, can be differentiated according to

how consumers acquire information about them and how this information affects consumers'

perceptions towards the products' performance. For search goods consumers have an active

participation searching out the product attributes which can be observed at the time of purchase









Future research will require more detailed data on beef promotion expenditures including

data from competing countries such as Australia and New Zealand in order to fully analyze the

relation between promotions and beef demand. In addition, the own nature of an international

commodity promotion program should be re-considered. Since the basic idea of such programs is

to expand the total demand for the commodity and not the demand for products from a particular

country, in an international scenario, it' s impossible to apply this concept without encountering

the problem of free riders. These countries get the benefits of increasing beef product demands

without an active participation in the promotion of their products. That could be one of the

explanations for the lack of significance of the promotion parameters in this research. Future

research must be able to disaggregate total international expenditures on beef promotions to fully

address their impact in a particular market.

Previous research on international commodity promotions conducted by Kinnucan and

Zheng (2004) and Piggott et al. (1996) suggested that market demands are insensitive to changes

in advertising expenditure, and that unless the beef exporter country has a surprising degree of

market power in a particular international market, the promotional campaigns may not have been

profitable, which clearly reflect the promotion results in this research, in particular in the case of

U.S. beef exports after the BSE announcements.

In all four markets considered in this study current market restrictions require that all

imported beef be produced from animals between 30 and 20 months of age or younger at the

time of slaughter and the ability of U. S. beef exporters to provide specific beef cuts could be

used as a marketing tool by turning around these restrictions into competitive advantages. The

highly integrated and efficient U. S. beef supply chain can easily satisfy the preferences of these

consumers in terms of age and type of beef cut in particular in sophisticated markets such as











than 40%. These price fluctuations might be a consequence of the presence or absence of beef


products from competing nations in the South Korean market. Despite the higher prices, beef


products from New Zealand have shown an important participation in the Korean market,


although during the past eight years a strong market competition has forced its prices to drop


more than 44% to a point where today's prices for New Zealand beef are 22% below the average


beef import price.

CIF Prices for Fresh Beef in Korea ($ per Kg)
12-
U.S. o AU INZ --World Av.


10-











9- 9 20 01 02 20 00 05 20

8-a

7-ue25 vrg I rcsfrfehbe pout nKrafo 99t 06 ore

6-AFS ebur 08

5-pa











1999mpio 2000n196 n 2001 2002 2003r 200 2005 2006 mr hn .% Hwvr









one, and each estimate should be in the (0, 1) interval (Armington 1969). However, when their

impact across time on U.S. market shares is illustrated, the range is above one in the first year

(Figures 5-1 through 5-3). This circumstance is the result of including the Hz(.x (DX, DYR,)

term in the model specification, which allows the market share parameter for beef imported from

the United States to have a higher initial value.

At this stage, it is important to mention that the BSE effect is nested on the time trend as

shown in the last portion of Table 5-1 and that the significance of these parameters is relatively

more important in the evaluation than the sign of the dummies. In the case of Pii, the base used

to compare the estimates is the Rest of the World. With this in mind, it is clear that U. S market

shares have been affected by the BSE scare when comparing the t-values after the BSE

announcements. All Pii sub-parameters show insignificant values with respect to the mean,

except for the first year and the post-B SE coefficients. Also, the effect of the B SE

announcements in North America shows a significant effect on the Australian and New Zealand

market participation compared to previous years. The Eigures below illustrate the significance of

the estimates in Table 5-1 and their impact on the parameters ai aii nd Pii across time.

One of the most revealing aspect of these figures is the fact that in each import market the

parameter values of asi are constant across beef suppliers and time. That is, all ai 's values for

the South Korean beef market are about 52%; for the Japanese market these values are about 2%,

and for Taiwan and Hong Kong the ai. values are 1% respectively. In the particular case of the

ai. parameters there are two point to be considered: first, it is suitable to think that regardless of

the origin of the beef product and how much is available for consumption, markets demands

show constant rate of substitution between beef and all other goods in that market and that

consumers substitute beef for other goods at a proportional rate across time; and second, they









technical relationship determines the functional nature of the product demands and market

shares, while the other relationships in the model remain unaffected by the use of the CRES

function (Sparks and Ward 1992).

The Constant Ratio Elasticity of Substitution

An important aspect of the CES assumption is that it implies the separability between

different import sources which entails that the consumer' s decision process may be viewed as

occurring in two stages (Varian 1992; Webb et al. 1989). In the initial stage, the importer decides

how much of the good Xi. is going to be imported (Equation 4-1); and then, in the second stage

(Equation 4-2), given the total amount imported, the importer decides how much product to

import from each country Xii (Davis and Kruse 1993). The single CES restricts responses of the

import demand of each product to the price change relative to the price index for the good to be

the same for all products (Winters 1984; Weatherspoon and Seale 1995). Therefore, the issue of

substitutability must be addressed in order to formulate a more specific system.

Previous studies applied to agricultural trade analysis have raised questions about several

properties of the Armington model. Studies by Alston et al. 1990; Moschini et al. 1994; Seale et

al. 1992; Weatherspoon and Seale 1995; and Winters 1984 all showed the biased character of the

Armington specification when it is based on the CES functional form. Furthermore, Yang and

Koo 1993 suggested a less restrictive set of assumptions on demand relationships than those of

the Armington model in order to avoid inconsistent parameter estimations.

In order to bring prices, quantities, and other factors into the model, this study suggests a

functional form: one-product, many-suppliers demand function using a more flexible form, the

Constant Ratio of Elasticity of Substitution (CRES) (Mukerji 1963, Hanoch 1971). That is, its

Elasticity of Substitution (ES) varies along a mapping of isoquants (i.e., all input bundles that













-.11 Unie States








.00 0.002 100


Austrli



0.599 70.575




-.021 0.021


New Zaland




0.259
0.20 --- --../ """ ---*0.215
0.021- -- 0.021
0 ago0.122


Parameter Value (%)


-AID IAIJ IBIJ


1.20
1.00

0.80
0.60

0.40
0.20
0.00
1.00

0.80

0.60

0.40

0.20


1.00

0.80

0.60

0.40

0.20

0.00


1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

Year


Figure 5-2. Japan product demand parameters for beef imports (US-AU-NZ).


With respect to the Pii parameters, all figures show a clear fluctuation across time and


across beef suppliers. Comparing the Pii coefficient results for the United States, Australia, and


New Zealand, the minimum and maximum Pii value for each of these suppliers were 0.2% (2005


to 2006) and 1 11% (1997) for U. S. beef products; 20% (1997) and 60% (2005) for Australian


beef; and 12% (2002) and 26% (2004) for New Zealand beef. In addition, although it is


impossible at this point to identify the full impact of these coefficients, it is evident that the BSE


announcements in 2003 have an important effect on the Pii trend.










International Beef Supply

The present research is based on a trade model specification that follows the Armington

model of product differentiation by country of origin using four exporting regions to formulate a

beef trade model that compares beef products. Traditionally, the three maj or suppliers of beef

products in the Pacific Rim region have been the United States, Australia, and New Zealand. All

other beef suppliers were aggregated and included in the model as part of the Rest of the World

category with countries such as Brazil, Canada, Mexico, Argentina, and the EU community

included in this last group. In the following sections each country or region is discussed showing

production capacity, consumption, imports (M) and exports (X) measured in metric tons (MT).

United States

As of December 2007, the United States is considered the third largest exporter of beef

products behind Brazil and Australia, and at the same time, it is also considered the largest

importer of beef, specifically ground beef imports from Australia and New Zealand. According

to official trade data on import and export volumes, the U. S. is clearly a net beef importer. The

ratios of imports-to-exports, during the last six years, support this statement in Table 2-3.

Table 2-3. U.S. beef imports and exports volumes between years 2002 and 2007 (1000 MT)
Year Imports (M) Exports (X) Ratio (M/X) %/Production
2002 1459.66 1109.94 1.31 9.0
2003 1363.50 1142.15 1.19 9.6
2004 1668.77 208.65 7.99 1.9
2005 1632.48 316.15 5.16 2.8
2006 1399.33 518.91 2.69 4.4
2007 1384.36 649.09 2.13 5.9
Source: USDA-ERS. February 2008.

Since 2002, the United States has imported, on average, 3.41 times more than it has

exported showing peak import levels right after the BSE outbreak. This represents the logical

consequence of the ban on U. S. beef products overseas and the increasing domestic demand in









CHAPTER 2
WORLD BEEF TRADE AND PRODUCTION

Introduction

The purpose of this chapter is to present an overview of the world beef market based on

descriptive statistics and trade data gathered from public and private international organizations.

In particular, in this segment the objective is to describe trade flows between a particular set of

countries that were previously defined as regions of interest. The discussion that follows includes

a description of the data set with specifications on classification and characteristics of the traded

beef products. In addition, country-specific demographics will be presented accompanied with an

analysis of relative prices, expenditures on beef products by country, volume traded, market

shares, and values of export-imports during the period 1995 to 2007.

International Beef Trade Data

For this research, the data set is based on international trade statistics from four importing

and three exporting countries who reported quantities, import-export values, and average prices

for beef products in their respective markets. All trade data are expressed in metric tons (MT)

and values in thousands of U. S. dollars. The data set for world meat trade was obtained from The

United States Department of Agriculture, the Foreign Agricultural Service, and the Economic

Research Service using the World Trade Atlas Program. This data set was developed by Global

Trade Information Services, Inc. (GTI, 2007) to provide information on international

merchandise trade to governments and corporations.

Existing studies on international food trade suggest that agricultural world trade can be

divided into regions using the criteria of geographic differences, the pattern of trade flows, and

political boundaries. Accordingly, today' s international beef trade scenario is divided into seven

regions: South and North America, Asia which includes South and South East Asia, East Asia,










SELECT WEXP=K & WTYPE=L;
WQALL=WQ.FALL; ?THIS IS THE TOTAL QUANTITY EXPORTED TO WIMP BY TYPE;
WQQ = WQ.F_.X; ?THIS IS THE QUANTITY (MIL MTONS) EXPORTED FROM WEXP TO WIMP BY
TYPE;
WVV = WV.F_.X; THIS IS THE VALUE ($1000) EXPORTED FROM WEXP TO WIMP BY TYPE;
WSH = WSH.F_.X; ? THIS IS THE MARKET SHARE OF WEXPij EXPORTED TO WIMP BY TYPE;
WRP = WRP.F_.X; ? THIS IS THE RELATIVE PRICE OF WEXPij RELATIVE TO THE AVERAGE
PRICE TO WIMP BY TYPE;
SELECT 1;
ENDDOT; ENDDOT;

DOT(VALUE=L,CHAR=F) 1 2 3 4;
DOT(INDEX=K,CHAR=X) 1 2 3 4;
XF = (WIMP=L & WEXP=K);
MSD(WEIGHT=XF,NOPRINT) WRP;
PRINT L K ~MAN;
ENDDOT; ENDDOT;

WFOR1 = EXP(WFOREIGN/1000);
WFOR2 = EXP(WFOREIGN/10000);
WFOR3 = EXP(WFOREIGN/1000000);

MSD WFOR1 WFOR2 WFOR3;
SET CV1 = ~MAN(1) / @STDDEV(1);
SET CV2 = ~MAN(2) / @STDDEV(2);
SET CV3 = ~MAN(3) / @STDDEV(3);

PRINT CV1 CV2 CV3;
HIST(DISCRETE) WEXP WIMP;


CREATING WIMP AND WEXP DUMMY VARIABLES

DOT(VALUE= J) 12 3 4;
DI.=(WIMP= J); ? DUMMY VARIABLES FOR THE IMPORTING COUNTRIES;
DX.=(WEXP=J); ? DUMMY VARIABLES FOR THE EXPORTING COUNTRIES;
ENDDOT;

DOT(CHAR=I,VALUE=J) 1 2 3 4;
DOT(CHAR=X,VALUE=K) 1 2 3 4;
D.I.X=(WIMP= J & WEXP=K);
ENDDOT; ENDDOT;

DOT(VALUE=J) 1 2;
DF.=(WTYPE=J); ? DUMMY VARIABLES FOR THE FRESH VERSUS FROZEN;
ENDDOT;


TIME SPECIFIC VARIABLES SUCH AS BSE

SET YRBSE=2003;
B SE=(WYEAR>=YRB SE);
T 1= 0;
T1=(WYEAR-1994)*(WYEAR<=YRBSE) + (WYEAR>YRBSE)*T1(-1);
T2=B SE* (WYEAR-YRB SE);




















LL 1117 Unite States
1.00-

0.80-
0.0 -.56 0.599 0.581


0.282
0.20-
0.002
0.00 ,-
1.00-
Austrli
0.80-
0.646
0.0 -.60P 0.589 *- '0.629
/ .599 1 0.575



0.20 029

0.00
1.00-
New haland
0.80-
0.600 0.580 0670.620
0.60-

0.40 _-0.511 0.515
0.259
0.20 --0.00 ---- .. """- --*0.215

0.00 0.121

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

Year


Figure 5-1. South Korean product demand parameters for beef imports (US-AU-NZ).


In the case of the aii's coefficients, it is clear that only in the case of the South Korean


Market there is some variation across time, although the range of this variation is only about 2


percentage points. For example, Figure 5-1 shows ail values for the United States, Australia,


and New Zealand. In the case of U.S. beef, these values vary between 56% and 58%; between


61% and 63% in the case of Australian beef; and between 60% and 62% in the case of beef from


New Zealand. In the rest of the markets the fluctuation of the aij's coefficients is almost


insignificant (+/- 1%), which is clearly shown in Figures 5-2 and 5-3.


have a measurable impact on the elasticity parameters only in the case of the South Korean beef


market.


Parameter Value (%)


-AID -AIJ -BIJ









having Japan and Korea as its main export destinations. Imports from the ROW region decreased

46% in all countries but in Hong Kong where traders concentrated their imports on alternative

sources, such as South America, resulting in a significant 50% increase during the year 2007.

Finally, although beef products from North America were totally banned shortly after the

December 2003 BSE announcement, the United States managed to maintain a small market

participation, especially after 2005.

Today U. S. beef exports are benefited by the strong economic growth of the Asian region

as rising incomes lead to greater demand of high quality beef (USMEF 2007). For example,

record U. S. beef exports of fresh/chilled and frozen beef to Taiwan during 2007 have surpassed

2003 volumes by 26%, for a total of 19,000 metric tons. In contrast, Japan, Korea, and Hong

Kong today only represent 8% of the 2003 United States total annual volume of exports; while

before the BSE scare they represented 97% of total U. S. beef exports. This significant difference

is the result of trade restrictions that allowed only boneless beef and no variety meats under 21

months of age in Japan and under 30 months of age in the rest of the markets (USMEF 2007). As

a consequence of those restrictions, the United States has lost almost 40 % of the region' s market

for fresh/chilled and frozen beef between December 2003 and the end of 2007, which represents

a loss in value and quantity traded of $1.627 billion and 431,000 metric tons respectively.

According to USMEF and USDA-FAS estimations, U.S. beef exports globally are

expected to continue to increase, especially in Japan, Korea, China, and Russia fueled by

continued economic growth, increasing incomes, and, more important, the OIE announcement

that declares the United States as a "controlled risk" BSE country that will help increase market

access for U.S. beef in all countries.





















































-


-


-


-


-


-


-


Rest of the World


Brazil, Canada, Argentina, and other beef-producing countries around the world have



shown a solid growth in their exports levels, in particular after the year 2003 with an increase of


almost 28%. Economic growth in Asia, Eastern Europe, and nations from other parts of the



developing world has boosted the demand for beef products which has also triggered global



production. International trade has benefited from this economic bonanza as Figure 2-13 clearly


illustrates a steady and positive trend of those indicators during the past 13 years. Thus,


international beef trade, world consumption, and production levels have shown growth levels of



more than 35%, 9.6%, and 12.6% respectively during the 1995 to 2007 period.


LMillions of Metric Tons (M. MT)
50-
Row
45 -Production-Consumptionl-Exports


1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
Exports 3.41 3.15 3.47 3.04 3.34 3.27 3.2 3.74 3.78 4.7 5.13 5 5.26
Consumption 35.09 33.68 35.42 35.54 36.55 36.18 35.5 36.74 35.86 36.41 37.44 38.16 38.81
ftoduedion 34.61 33.47 34.79 34.55 35.34 35.45 35.02 36.14 35.28 37.27 38.33 38.89 39.6

Year


Figure 2-14. Rest of the world production, consumption, and exports of fresh/chilled and frozen
beef from 1995 to 2007. Source: USDA-FAS. February 2008.


40


35


30


25


20


15


10



O












and 2008. The figure shows the impact of the complete ban on U. S. beef products that is


translated into a large and positive difference in Australian product demands between the pre-


BSE years and the year 2006. This difference represents more than 150% increase in product


demands. In the case of New Zealand, there was a decline of 3% in product demand between


1998 and 2006. As the relative prices of exported beef increase from 10 % below the mean or


average price to 10 % above the mean prices there was an identical decrease across time in the


quantity of beef product demanded of 28.8% in the case of U. S. beef, 41.9% in the case of


Australian beef, and 38.9% in the case of New Zealand beef.

1000 luT
100.00-
Japan
Price Quantity R~esponse
0-0.9 101 511.1
80.00-




60.00 ---




40.00 ------




20.00 -------




0.00 ------
US 1998 AU 1998 NZ 1998 US 2002 AU 2002 NZ 2002 US 2006 AU 2006 NZ 2006
0.9 67 37 64 27 26 25 63 91 54 55 13 83 0 48 78 60 25 95
1 62 31 57 41 23 65 59 11 48 73 12 47 0 45 70 22 23 38
1.1 57 96 51 88 21 52 54 98 44 03 11 34 0 41 63 44 21 28
Elasticity -1 01 -1 02 -1 02 -1 01 -1 02 -1 02 0 00 -1 02 -1 02

Adjusted Price


Figure 6-8. Japan beef market demands and price elasticities in 1998, 2002, and 2006.


The product demand responses to variation in the mean prices are more extreme in the case


of Australian beef, where in all cases a 10% reduction in the price of beef originated in this









Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

EXPORT DEMAND AND PROMOTIONS OF U.S. BEEF: IMPACT OF FOOD SAFETY
AND PROTECTIONISM.

By

Oscar Ferrara

August 2008

Chair: Ronald W. Ward
Maj or: Food and Resource Economics

Trade disruptive policies have been used for a long time to address a large number of

politic, economic, and food safety issues. Policy shocks (i.e., embargoes or export bans) have

significant short-and long-term effects on trade relations and production of agricultural

commodities, shaping the economies of producing countries as well as the direction and

magnitude of trade flows. The consequences of the first case of BSE in North America in

December 2003 exemplify how food safety issues significantly affect consumer preferences and

producer welfare. After the announcement of the first case of BSE in North America in

December 2003, almost immediately 53 countries banned imports from the United States to

prevent the disease from entering their countries and to safeguard human health. The resulting

negative shift in aggregate market demand for U. S. beef products and the loss of export markets

were the immediate consequences of either government-imposed trade restrictions or changes in

import demands.

Today, U.S. beef exports show an increasing trend thanks in part to trade liberalization

policies and information campaigns (i.e., promotions) aimed to expand the presence of U. S.

products in maj or beef importing countries. The obj ective of this research is to analyze import










SELECT WYEAR>1996;
DUMMY(PREFIX=DYR) YEAR;
DOT DYR1-DYR11;
MSD(WEIGHT=.) YEAR; ENDDOT;

DOT(CHAR=X) 1 2 3 4;
DOT(CHAR= #) 1-11;
DYR.# .X=DYR.# DX.X;
ENDDOT; ENDDOT;
SELECT 1;

AI= 1.00;
AIJ=1.00;
BIJ=1.00;
THO=1.00;
TH1=1.00;
TH2=1.00;

LWQQ = LOG((WQQ+.001)/1000);
LSHWQQ = LOG( (WQQ+.001)/ WQALL );

THE NEW ARMINGTON MODEL WITH DATA STACKED USING DUMMIES
TO IDENTIFY WIMP AND WEXP

?FRML EQ1 LOG(WQQ + .001) -[ THO + TH1*( LOG(WRP) ) + TH2*LOG(WQALL) ];
FRML EQ1 LWQQ = [ THO + TH1*( LOG(WRP) ) + TH2*LOG(WQALL/1000) ];
?FRML EQ1 LSHWQQ = [ THO + TH1*( LOG(WRP) ) + (TH2-1)*LOG(WQALL/1000) ];

? THERE CAN BE SEVERAL ALTERNATIVE SPECIFICATION FOR AI AIJ AND BIJ;

? 0 < Aij < 1.0 USING A LOGISTIC FUNCTION;
IDENT EQ2 AIJ = 1 / [1+ EXP[GO + GX1*DX1 + GX2*DX2 + GX3*DX3 + GIl*DI1 + GFl*DF1
+ GG1*"(WFOREIGN/10000) ] ] ;? + GI3*DI3 + GI2*DI2;

? 0 < AI < 1.0 USING A LOGISTIC FUNCTION;
IDENT EQ7 AI = 1/[ 1 + EXP[ KO + KIl*DI1 + KI2*DI2 ]]; ? + KI3*DI3;

? Bij > 0; USING AN EXPONENTIAL FUNCTION;
IDENT EQ3 BIJ = [EXP[
[HO + HYR2*DYR2 + HYR3*"DYR3+ HYR4*DYR4 + HYR5*DYR5 + HYR6*DYR6
+ HYR7*DYR7 + HYR8*DYR8 + HYR9*DYR9 + HYR10*DYR10 + HYR11*DYR11]
+ [HYR1X1 + HYR2X1*DYR2 + HYR3X1*DYR3+ HYR4X1*DYR4 + HYR5X1*DYR5
+ HYR6X1*DYR6 + HYR7X1*DYR7 + HYR8X1*DYR8 + HYR9X1*DYR9 + HYR10X1*DYR10
+ HYR11X1*DYR11]*DX1
+ [HYR2X2*DYR2 + HYR3X2*DYR3+ HYR4X2*DYR4 + HYR5X2*DYR5 + HYR6X2*DYR6
+ HYR7X2*DYR7 + HYR8X2*DYR8 + HYR9X2*DYR9 + HYR10X2*DYR10
+ HYR11X2*DYR11]*DX2
+ [HYR2X3*DYR2 + HYR3X3*"DYR3+ HYR4X3*DYR4 + HYR5X3*DYR5 + HYR6X3*"DYR6
+ HYR7X3*DYR7 + HYR8X3*DYR8 + HYR9X3*DYR9 + HYR10X3*DYR10
+ HYR11X3*DYR11]*DX3 ] ];
? + (HI1*DI1) + (HI2*DI2) + (HI3*DI3) ;? *(DX1=0 & DX2=0 & DX3=0)

IDENT EQ4 THO = [1/(AIJ-1)]*[ LOG(AI) LOG(AIJ) LOG(BIJ) ];
IDENT EQ5 TH1 = [1/(AIJ-1)];
IDENT EQ6 TH2 = [ 1/( AIJ -1) ]*(AI -1);
EQSUB EQ4 EQ2 EQ3 EQ7;










DO ADJWQALL = .80 TO 1.20 BY .05;
SET SIMVAR = SIMVAR+1;
SET SDI3= 1; SET SDX3= 1; ZB SIMZ;
ENDDO; SET H= SIMVAR;


STARTING THE BEEF TRADE SIMULATOR
SIM #13 RELATIVE TOTAL DEMAND IS ADJUSTMENTS IN HONG KONG


SET SIMNUM = 13;
ZRESETZ; SET SIMVAR = 1; S
DO ADJWQALL = .80 TO 1.20 BY .05;
SET SIMVAR = SIMVAR+1;
SET SDX1=1; ZBSIMZ;
ENDDO; SET H= SIMVAR;

ZRESETZ; SET SIMVAR = 1; S
DO ADJWQALL = .80 TO 1.20 BY .05;
SET SIMVAR = SIMVAR+1;
SET SDX2=1; ZBSIMZ;
ENDDO; SET H= SIMVAR;

ZRESETZ; SET SIMVAR = 1; S
DO ADJWQALL = .80 TO 1.20 BY .05;
SET SIMVAR = SIMVAR+1;
SET SDX3=1; ZBSIMZ;
ENDDO; SET H= SIMVAR;


IET SDX1=1; ZBSIMZ; ? US





ET SDX2=1; ZBSIMZ; ? AUSTRALIA





ET SDX3=1; ZBSIMZ; ? NEW ZEALAND;


STARTING THE BEEF TRADE SIMULATOR
SIM #14 RELATIVE TOTAL FOREIGN PROMOTIONS IS ADJUSTMENTS IN KOREA


SET SIMNUM = 14;
ZRESETZ; SET SIMVAR = 1; SET SDIl
DO ADJ FOREIGN = .70 TO 1.30 BY .05;
SET SIMVAR = SIMVAR+1;
SET SDIl=1; SET SDX1=1; ZBSIMZ;
ENDDO; SET H= SIMVAR;

ZRESETZ; SET SIMVAR = 1; SET SDIl
DO ADJ FOREIGN = .70 TO 1.30 BY .05;
SET SIMVAR = SIMVAR+1;
SET SDIl=1; SET SDX2=1; ZBSIMZ;
ENDDO; SET H= SIMVAR;

ZRESETZ; SET SIMVAR = 1; SET SDIl
DO ADJ FOREIGN = .70 TO 1.30 BY .05;
SET SIMVAR = SIMVAR+1;
SET SDIl=1; SET SDX3=1; ZBSIMZ;
ENDDO; SET H= SIMVAR;


=1; SET SDX1=1; ZBSIMZ;


? US


=1; SET SDX2=1; ZBSIMZ; ? AUSTRALIA





=1; SET SDX3=1; ZBSIMZ; ? NEW ZEALAND;


STARTING THE BEEF TRADE SIMULATOR
SIM #15 RELATIVE TOTAL FOREIGN PROMOTIONS IS ADJUSTMENTS IN JAPAN









Bovine Spongiform Encephalopathy (BSE): Impact on Beef Trade

This section presents an analysis of the impact of 2003's BSE announcement on maj or

beef markets, the consequences for American beef exports, and how this food safety issue

affected patterns of trade. To facilitate the interpretation, each importing market is analyzed from

the consumption and market share perspectives, while country-specific expenditures per capital

on beef are used to understand BSE consequences in exporting nations. Figures 2-15 through 2-

18 illustrate pre- and post-BSE per capital consumption of imported beef, as well as market share

information for Japan, Korea, Taiwan, and Hong Kong. Figures 2-19 through 2-22 introduce the

amount of U. S. dollars expended on imported fresh/chilled and frozen beef products in each of

the four importing market with respect to a specific exporting country. Finally, Figures 2-23 and

2-24 present a pre- and post-BSE trade volume matrix that represents trade patterns between

each of the eight areas considered for this research.

Consumption Levels and Market Shares

According to the USDA-FAS dataset used in this study, the Republic of South Korea was

considered the third largest market for U.S. beef products before the BSE announcement and

accounted for roughly 19% of all U.S. beef exports on a volume basis and 22% on a value basis.

In numbers, Korea imported more than $ 816 million worth of U. S. beef products during 2003,

which represented a consumption level of 17.42 Kg (38.3 lb) per capital.

The United States accounted for 69% of Korea' s beef market, followed by Australia, New

Zealand, and the rest of the world (ROW) with 21%, 8%, and 2% of the import market share

respectively (Figure 2-15). The same data indicated that more than 90% of Korea' s imports from

the United States and its competitors were concentrated in the following categories according to

their total volume: frozen boneless and bone-in beef, fresh/chilled beef, and frozen edible offal.









International U.S. Beef Promotions

Purchase decisions are based on predictions of product performance. Consumers base their

predictions in part on product cues and are accurate to the extent that they have properly learned

the relationship between the cues and performance (Van Osselaer and Alba 2000). If consumers

learn the relationship between product attributes and quality, they will differentiate among goods

that possess different attributes and treat as close substitutes those that share the same attributes.

During the past twenty years, the U.S. beef industry has experienced a slow transition from

a traditional commodity-selling perspective to a more contemporary marketing-driven sector.

This evolution has encouraged the industry to differentiate its products, review production

processes, reorganize its value chain, and use a mix of price and non-price marketing strategies

in order to increase market share. Competition has become much more intense and the demands

for red meat products, in particular fresh, chilled, and frozen beef reveal the distinctive

characteristics in product quality and production technology across exporting countries.

Although only 12 % of the beef produced in the United States is exported, international market

participation has an important impact on the industry because it raises the value of all products. It

increases the number of customers which consequently boosts prices through a positive shift in

aggregate demand and it allows packers to sell differentiated products according to preferences

in each international market, thus obtaining higher rents from those customers willing to pay

global prices (Kinnucan et al. 1995).

Advertising and promotion provide useful information necessary to facilitate purchasing

decisions, and in some cases they are even used to change the underlying preference function for

a particular product or service. Commodities in general, and specifically beef products, can be

differentiated according to how consumers acquire information and how this information affects

consumers' perceptions towards the products' performance. In places like Korea, Japan, Hong










(FAOSTAT 2007). Rising consumer demand for beef, aided by trade liberalization and changes

in technology (i.e., shipping and storage) has helped beef to grow across the globe (Figure 1-1).

Beef trade mainly is in cuts or parts, not in the form of live animals or carcasses which generate

extra revenue per animal. The slaughter of a meat animal automatically generates a full set of

muscle meat cuts, as well as trimmings, offal, and other byproducts (Table 1-1). The value of a

carcass is the composite value of the cuts and other products taken from it, and the derived value

of a meat animal is the composite value of the carcass and byproducts from the animal, less

processing and transaction costs (Dyck and Nelson 2003).

Table 1-1. Top 5 exported beef cuts during 2006. Ranked by the extra value generated.
Cut Domestic % U.S. Value International Total Extra
Type Use Export ($kg) Value ($kg) Value ($ M)
Short Ribs Trim 57 1.39-2.09 4.00-10.00 338

Tongue Pet Food 70 0.22 9.92 328
Outside Skirt Trim 61 2.09 3.50-6.00 166

Short Plate Trim 68 0.96 1.74-2.65 63

Chuckeye Roll As is 11 2.62-2.68 3.50-6.00 43
Source: USMEF. September 2007.

In recent years, U. S. beef exports represented 12.4% of the production generating an

additional value equivalent to 18% of the total value of a feed steer or the equivalent of an

increase of almost one billion dollars out of the top five export beef cuts. Forecasts from the U. S.

Meat Export Federation (USMEF) show that for the next five years a total increase of almost

90% in the volume of beef exported is expected and that Japan, Hong Kong, Taiwan, and Korea

are among the top five destinations (USMEF 2007).

Trade issues have become increasingly important for the U. S. beef industry, in particular

those referred to as food safety have significant effects, either through government-imposed

trade restrictions or changes in import demand (Verbeke and Ward 2001; Jin and Koo 2003).









CHAPTER 1
INTRODUCTION

During the past three decades, globalization and trade liberalization have contributed to

reducing the isolation of domestic economies but also have permanently changed the channels of

food distribution, raising important questions about the safety and quality of agricultural goods.

Political and economic changes have forced countries around the world to abandon the

complexity of old trade structures and incorporate open mechanisms in which legal and

monitoring protocols are highly standardized. Under this new world order, transactions are set to

be dynamic in order to encourage the exchange of goods and services produced across the

borders and reduce government participation. During the past two decades, the development of

new technologies in areas related to communication and logistics has facilitated, innovated, and

consequently accelerated the way that countries trade around the world. In addition, a number of

international institutions have played an important role in promoting free trade including the

World Bank, International Monetary Fund (IMF), and General Agreement on Tariffs and Trade

(GATT). The latter, was succeeded in 1995 by the World Trade Organization (WTO) (Jung

2004). In addition, Regional Trade Agreements (RTAs) such as the Asia Pacific Economic

Cooperation (APEC), European Union (EU), the Mercado Comun del Sur (MERCOSUR), and

North American Free Trade Agreement (NAFTA) have boosted trade among countries within a

particular region of the world.

International commodity markets represent an important focus of attention for the U.S.

agricultural sector. Recognizing the significance of the factors affecting the flow of agricultural

products is necessary in order to take advantage of the current commodity trade structure and

predict its future evolution. New international trade policies, food security issues, and shifting










log wii = o- log bi + (1 o-) I j ,2,.. (A-15)

where wi denotes the market share of imports from source j. No expenditure term in the right

hand side in Equation A-15 implies homotheticity assumption. That is, the change in importer's

expenditure does not affect the market share (Yang and Koo 1993). As a result, all expenditure

elasticities within a group are equal and unitary and import market shares change only in

response to relative price changes.










Kong, and Taiwan, U.S. beef is highly appreciated because of its particular characteristics of

taste, fat content, tenderness, and consistency which have influenced consumers' preferences for

more than two decades (USMEF 2007). Studies before the BSE outbreak have shown that the

most important characteristics for consumers of U. S. beef in these markets are that it tastes good,

it looks fresh, does not have a lot of waste, and that it is USDA certified and inspected.

Marketing oriented to recuperate market shares should reinforce these attributes, emphasizing the

characteristics of the product in order to re-build the positive perception about U. S. beef

products. Promoting U.S. beef as a brand might be necessary to reach pre-ban levels of

consumption in these countries using product differentiation based on the attributes of the

product rather than simply selling beef as a commodity. The U. S. beef checkoff promotion

programs have been an instrument for providing consumers with information about the attributes

of U. S. beef. As in many promotion programs, the main obj ective of the beef checkoff program

is to increase the demand for beef products in the domestic market as well as internationally. The

USMEF, the international arm of the U. S. beef board, has launched several marketing campaigns

abroad to keep U.S. beef products in front of consumers in markets that are still closed or to

increase existing demand. Lately, efforts are aimed to overcome the trade barriers imposed by

the EU on U. S. beef over the issue of using hormones in cattle production, and age restrictions

imposed by several countries in the Pacific Rim region.

Problem Statement and Research Objective

International beef trade has shifted from a domestic or sometimes regional commercial

activity to a global network of buyers and sellers that continues to grow in complexity and

importance. The international marketplace for beef products is influenced by several aspects

including changes in consumer demand and preferences, national and international trade policies,









industries and governments are constantly seeking and enforcing the policies necessary to

participate in the international market while protecting local markets from foreign competition.

Many studies from different countries have focused their attention on beef trade and the

economic impact of the policies designed to address food safety issues. For example, in 2003

Poulin and Boame from the Canadian Business and Trade Statistics Department focused their

efforts to measure the economic impact of the May 2003 BSE announcement on cattle

production, domestic prices, and beef exports to United States and Mexico, since Canada

exported more than 92 % of its production to these countries. As a consequence of this outbreak,

beef exports fell virtually to zero during the following months affecting entire supply chain from

the breeders to the meat packing plants. The immediate loss in terms of trade represented almost

$ 1 billion, with most of this loss concentrated in the fresh beef U. S. market where 30% of the

beef imported came from Canada. Other impacts were found at the domestic level where the

collapse of exports triggered an increase in beef supplies and a fall in prices that generated a 50%

drop in cattle prices, although consumers did not see a significant drop in the price of beef at the

retail price which clearly shows a price asymmetry effect of the embargo.

Differences in what food products countries want and what they will accept in imported

food ultimately affect patterns of food demand and global trade, and complicate the development

of workable trade rules that are acceptable to different trading partners (Buzby, 2003).The effect

of trade barriers on consumer demand for traded goods of different quality was approached by

Bureau et al. (2005) under the assumption that specific tariffs and tariff-rate quotas (TRQ) have

an effect in the composition of imported goods due to a variation of the price ratios between

products of different quality. The study used a standard Constant Elasticity of Substitution (CES)

function to deal with differentiation, where the parameters represented consumers' relative









transition from the generic idea of selling beef as a commodity to a campaign involving market

segmentation and target marketing of specific consumer segments. According to Phillip Kotler

(2002), market segmentation may be described as an assumption that all consumers are unique

and the needs of individuals may not be satisfied with a mass marketing approach. Similarly,

target marketing may be defined as a market segment profied by its demographic, economic,

and psychographic characteristics so that marketing opportunities may be evaluated. The precise

identification of a product's functional benefits and attributes, as well as the demographic and

non-demographic variables influencing consumer selection and decision-making process, would

be useful in developing marketing and merchandising strategies.

Chakravarti and Janiszewski in 2004 examined the influence of generic advertising on

primary demand and brand preferences. They concluded firstly that generic advertising can

increase or decrease the perceived differentiation among competing branded products and, thus,

influence consumers choice. Second, increases in differentiation occur because generic

advertising increases or decreases the weight consumers place on differentiating or non-

differentiating attributes. Generic advertisements that discussed a differentiating attribute

decreased access to information about the non-differentiating attribute, which resulted in an

increase in the importance of the differentiating attribute and increased price responsiveness.

The importance of U. S. beef promotion programs within the four Asian nations considered

in this study differs considerably and likely has evolved over time representing major

implications for the beef industry and its marketing strategies. Brands may segment the market

and may or may not grow total demand for the product category. Such programs are often

mandatory, where producers or exporters of a promoted good (i.e., beef) must contribute to the

marketing programs through a "check-off' system designed to avoid free-riders. In 2005,









Pre and Post-BSE Pattern of Trade

Figures 2-23 and 2-24 present trade quantities by country and by trade partner nation for

the years 2003 and 2007, which represent pre-BSE and today's beef scenarios. Each figure

represents a coordinate system indicating three axes that represent trade quantities by importing

and exporting country for the pre-and post-BSE announcements periods. Thus, the planes X, Y,

and Z represent the demand, supply, and the quantities traded. Each market is identified with a

particular color. As an example, looking at Figure 2-23, and the U. S. and Korea as beef trade

partners, during the pre-BSE year 2003 Korea imported 224,000 metric tons of fresh/chilled and

frozen beef products from the U. S. This comprised 21.7% of the region' s total beef imports

within this category for that particular year. This also represented 68.7% of Korea' s total

fresh/chilled and frozen beef imports and 46% of all U.S. exports to the region during that year.

Following the same criteria to analyze trade relations among participant nations, Figures 2-23

and 2-24 are contrasted and discussed in the following pages.

Recall that before the BSE scare the total supply of beef products in this category was

about 1,031,000 metric tons and the United States was by far the largest exporter to the region.

Figure 2-23 reveals that more than 47% of all beef imported in this region originated from the

United States and that Australia had a little more than 33% followed by the ROW and New

Zealand with 13% and 7% respectively. In terms of value and volume traded, Japan was the main

market for U. S. products but Australia was the leading beef supplier of this country during that

year. New Zealand' s exports to Korea represented 38% of its total exports during that year, while

exports to Japan represented 26%. In the case of Taiwan and Hong Kong, the figures show that

they obtained a large percentage of their beef from the same set of suppliers, with Australia

being the largest supplier to Taiwan and the ROW region in Hong Kong.




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EJGUG4W85_348XAN INGEST_TIME 2011-04-22T19:51:52Z PACKAGE UFE0022607_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES



PAGE 1

1 EXPORT DEMAND A ND PROMOTIONS OF U.S. BEEF: IMPACT OF FOOD SAFETY AND PROTECTIONISM. By OSCAR FERRARA A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2008

PAGE 2

2 2008 Oscar Ferrara

PAGE 3

3 To my wife Jaquelina ; and my children Sofia, Nicolas, and Marco

PAGE 4

4 ACKNOWLEDGMENTS My de epest gratitude goes t o Jaquelina Sofia, Nicolas, and Marco for their understanding and constant encouragement I thank Dr. Ronald W. Ward for his generosity and friendship. As mentor and professor, he helped me to pul l this project together. His valuable guidance and patience are deeply appreciated. I would also like to address very special thanks to Dr. Charles Adams, Dr. Jeff B urkhardt and Dr. Jim Morris for their generous assistance.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 4 LIST OF TABLES ................................ ................................ ................................ ........................... 7 LIST OF FIGURES ................................ ................................ ................................ ......................... 8 ABSTRACT ................................ ................................ ................................ ................................ ... 10 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .................. 12 Importance of Agricultural Trade ................................ ................................ ........................... 13 International Beef Market ................................ ................................ ................................ ....... 14 International U.S. Beef Promotions ................................ ................................ ........................ 19 Problem Statement and Research Objective ................................ ................................ ........... 20 Scope ................................ ................................ ................................ ................................ ....... 22 Research Methodology ................................ ................................ ................................ ........... 24 Armington Trade Model ................................ ................................ ................................ ......... 25 Data Set ................................ ................................ ................................ ................................ ... 29 Outline of Chapters ................................ ................................ ................................ ................. 30 2 WORLD BEEF TRADE AND PRODUCTION ................................ ................................ .... 32 Introduction ................................ ................................ ................................ ............................. 32 International Beef Trade Data ................................ ................................ ................................ 32 The International Beef Market ................................ ................................ ................................ 34 Volume, Share, and Growth of Global Production ................................ ......................... 35 Beef Trade Patterns ................................ ................................ ................................ ......... 37 Pacific Rim Demand for Beef Products ................................ ................................ .................. 42 Republic of South Korea ................................ ................................ ................................ 43 Japan ................................ ................................ ................................ ................................ 45 Republic of Taiwan a nd Hong Kong ................................ ................................ ............... 48 International Beef Supply ................................ ................................ ................................ ....... 52 United States ................................ ................................ ................................ .................... 52 Australia ................................ ................................ ................................ .......................... 55 New Zealand ................................ ................................ ................................ .................... 56 Rest of the World ................................ ................................ ................................ ............. 58 Bovine Spongiform Encephalopathy (B SE): Impact on Beef Trade ................................ ...... 59 Consumption Levels and Market Shares ................................ ................................ ......... 59 Expenditures on Beef by Country of Origin ................................ ................................ .... 65 Pre and Post BSE Pattern of Trade ................................ ................................ ................. 69

PAGE 6

6 3 LITERATURE REVIEW ................................ ................................ ................................ ....... 73 Introduction ................................ ................................ ................................ ............................. 73 International Beef Demand and Food Safety ................................ ................................ .......... 73 Trade and Beef Markets Issues ................................ ................................ ............................... 78 Internation al Beef Promotions ................................ ................................ ................................ 81 The Armington Trade Model ................................ ................................ ................................ .. 88 4 EMPIRIRICAL TRADE MODEL FOR BEEF PRODUCTS ................................ ................ 93 Introduction ................................ ................................ ................................ ............................. 93 Beef Trade Schematic Representation ................................ ................................ .................... 93 Adapting the Armington Model to Beef Trade ................................ ................................ ....... 95 The Constant Ratio Elasticity of Substitution ................................ ................................ ...... 100 Analysis of Trade Variations ................................ ................................ ................................ 107 5 EMPIRICAL RESULTS ................................ ................................ ................................ ...... 113 Introduction ................................ ................................ ................................ ........................... 113 Data Set ................................ ................................ ................................ ................................ 113 Product Demand Eq uation ................................ ................................ ................................ .... 114 Model Results ................................ ................................ ................................ ....................... 115 6 SIMULATION ANALYSIS ................................ ................................ ................................ 125 Introduction ................................ ................................ ................................ ........................... 125 Total Product Demand Trend ................................ ................................ ............................... 128 Product Demand Simulations ................................ ................................ ............................... 135 Pre and Pos t BSE Market Distribution Simulations ................................ ........................... 141 7 SUMMARY AND CONCLUSIONS ................................ ................................ ................... 146 APPENDIX A CONSTANT ELASTICITY OF SUBSTITUTION (CES) ................................ .................. 152 B TSP PROGRAM: ARMINGTON SPECIFICATION AND SIMMULATIONS ................ 157 LIST OF REFERENCES ................................ ................................ ................................ ............. 176 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ....... 183

PAGE 7

7 LIST OF TABLES Table page 1 1 Top 5 exported beef cuts during 2006. ................................ ................................ .............. 16 1 2 Trade status and BSE trade impact in millions of U.S. dollars by beef product ............... 17 2 1 Compared world production le vels and trade ................................ ................................ .... 36 2 2 Ratio import to export quantit ies by regions ................................ ................................ ...... 40 2 3 U.S. beef i mports and exports volumes ................................ ................................ ............ 52 5 1 Parameter estimates a nd t values ................................ ................................ ..................... 118 5 2 Summary statistics product demands parameters ................................ ............................ 124 6 1 South Korea: estimated base values ................................ ................................ ................. 126 6 2 Japan: estimated base values ................................ ................................ ............................ 126 6 3 Taiwan: estimated base values ................................ ................................ ......................... 126 6 4 Hong Kong: esti mated base values ................................ ................................ .................. 126 6 5 Estimated variations in market demand distribution ................................ ....................... 142

PAGE 8

8 LIST OF FIGURES Figure page 1 1 Domestic consumption of beef products in Asian regions.. ................................ .............. 15 1 2 Economic impact of the BSE Ban across U.S. beef export products as January, 2006.. ................................ ................................ ................................ ................................ .. 18 1 3 Beef trade participation of selec ted countries ................................ ................................ ... 23 2 1 Total beef production and shares by region.. ................................ ................................ ..... 35 2 2 Relative variation in growth rate of production.. ................................ ............................... 37 2 3 Pre and post BSE beef trade levels by region. ................................ ................................ .. 38 2 4 Republ ic of Korea beef market indicators.. ................................ ................................ ....... 44 2 5 Average CIF prices for fresh beef products in Korea. ................................ ....................... 45 2 6 Japan beef market indicators.. ................................ ................................ ............................ 46 2 7 Average CIF prices for fresh beef products in Japan. ................................ ........................ 47 2 8 Taiwan and Honk Kong beef market indicators. ................................ ............................... 48 2 9 Average CIF prices for fresh beef products in Taiwan ................................ ...................... 51 2 10 Average CIF prices for fresh beef products in Hong Kong. ................................ .............. 51 2 11 U.S. production, consumption, and exports of fresh/chilled and frozen beef. ................... 54 2 12 Australia production, consumption, and exports of fresh/chilled and froz en beef. ........... 55 2 13 New Zealand production, consumption, and exports of fresh/chilled and frozen beef. .... 57 2 14 Rest of the world (ROW) pr oduction, consumption, and exports of fresh/chilled and frozen beef. ................................ ................................ ................................ ........................ 58 2 15 Korea imported beef consumption per capita and market shares.. ................................ .... 60 2 16 Japan imported beef consumption per capita and market shares. ................................ ...... 61 2 17 Taiwan imported beef consumption per capita and market shares ................................ .... 63 2 18. Hong Kong imported beef consumption per capita and market shares ............................. 64 2 19 Total expenditures on beef prod ucts from the U.S. by country. ................................ ........ 65

PAGE 9

9 2 20 Total expenditures on beef products from Australia by country. ................................ ...... 67 2 21 Total expenditures on beef products from New Zealand by country. ................................ 67 2 22 Total expenditures on beef products from the ROW by country.. ................................ ..... 68 2 23 Pre BSE ban trade scenarios for fresh/chilled and frozen beef in selected Asian natio ns. ................................ ................................ ................................ ............................... 70 2 24 Post BSE ban trade scenarios for fresh/chilled and frozen beef in selected Asian nations ................................ ................................ ................................ ................................ 71 4 1 Schematic representation o f the Armington trade system for beef products in selected countries of the Pacific Rim region. ................................ ................................ .................. 94 5 1 South Korean product demand parameters for beef imports (US AU NZ). .................... 121 5 2 Japan product demand parameters for beef imports (US AU NZ). ................................ 122 5 3 Taiwan and Hong Kong product demand parameters for beef imports (US AU NZ). ... 123 6 1 Estimated fresh U.S beef product demands across time in selected Asian markets. ....... 129 6 2 Estimated frozen U.S beef product demands a cross time in selected Asian markets. ..... 130 6 3 Estimated fresh Australian beef product demands across time in selected Asian markets ................................ ................................ ................................ ............................. 131 6 4 Estimated frozen Australian beef product demands across time in selected Asian markets. ................................ ................................ ................................ ............................ 132 6 5 Estimated fresh New Zealand beef product demands across time in selected Asian markets. ................................ ................................ ................................ ............................ 133 6 6 Estimated frozen New Zealand beef product demands across time in selected Asian markets. ................................ ................................ ................................ ............................ 134 6 7 South Korea beef market demands an d price elasticities in 1998, 2002, and 2006. ........ 136 6 8 Japan beef market demands and price elasticities in 1998, 2002, and 2006. ................... 137 6 9 Taiwan beef market demands and price elasticities in 1998, 2002, and 2006. ................ 139 6 10 Hong Kong beef market demands and price elasticities in 1998, 2002, and 2006. ......... 140 6 11 Pre BSE market distribution of fresh and frozen beef products. ................................ ..... 141 6 12 Post BSE market distribution of fresh and frozen beef products. ................................ ... 143

PAGE 10

10 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy EXPORT DEMAND A ND PROMOTIONS OF U.S. BEEF: IMPACT OF FOOD SAFETY AND PROTECTIONISM. By Oscar Ferrara August 2008 Chair: Ronald W. Ward Major: Food and Resource Economics Trade disruptive policies have been used for a long time to address a large number of politic, economic, and food safety issues. Polic y shocks (i.e. embargoes or expo rt bans) have significant short and long term effects on trade relations and production of agricultural commodities, shaping the economies of producing countries as well as the direction and magnitude of trade flows. The c onsequences of the first case of BSE in North America in December 2003 exemplify how food safety issues significantly affect consumer preferences and producer welfare. After the announcement of the first case of BSE in North America in December 2003 almos t immediately 53 coun tries banned imports from the United States to prevent the disease from entering their countries and to safeguard human health. The resulting negative shift in aggregate market demand for U.S. beef products and the loss of export marke ts were the immediate consequence s of either government imposed trade restrictions or changes in impor t demands. Today, U.S. beef exports show an increasing trend thanks in part to tr ad e li be r a li za ti on p ol ic ie s and information campaigns (i.e. promotions) aimed to expand the presence of U. S. products in major beef importing countries. The objective of this research is to analyze import

PAGE 11

11 demands for be ef pro ducts in selected countries using a model of product differentiation by country of origin and to estimate empirically the impact of U.S. beef promotions on re capturing p re ban market levels. Based on ), the model for this research assumes a two stage budgeting allocation and a product's imperfect substitutability between export sources. Hence, this study presume s that beef produced in the Un ited States presents unique attributes (i.e. superior quality and taste) and that foreign consumers can perceive this difference from competing beef sources. work pr oposes the use of a Constant Ratio Elasticity of Substitution (CRES) that allows elaticities to vary proportionally while the s ubstitutability is not necessar i ly identical. This research includes eight countries divided in to four exporting and four importin g countries tire dataset includes almost 2 000 observations with information recorded over time across sele cted countries during an 11 year period (January 1995 to December 2006). Resulting product demand m arket share and e lasticity estimates determine the degree of substitution among competing beef products and provide the coefficients to measure the economic impact of trade barriers and effectiveness of market access and promotion policies in four m ajor markets for U.S. beef: Japan, Republic of Korea, Hong Kong, and Taiwan.

PAGE 12

12 CHAPTER 1 INTRODUCTION During the past three decades, g lobalization and trade liberalization have contributed to reducing the isolation of domestic economies but also have permanently changed the channels of food distribution, raising important questions about the safety and quality of agricultural goods. Poli tical and economic changes have forced countries around the world to abandon the complexity of old trade structures and incorporate open mechanisms in which legal and monitoring protocols are highly standardized. Under this new world order, transactions ar e set to be dynamic in order to encourage the exchange of goods and services produced across the borders and reduce government participation. During the past two decades, the deve lopment of new technologies in areas related to communication and logistics h as facilitated, innovated and consequently accelerated the way that countries trade around the world. In addition, a number of international institutions have played an important role in promoting free trade including the World Bank, International Monetar y Fund (IMF), and General Agreem ent on Tariffs and Trade (GATT). The latter, was succeeded in 1995 by the World Trade Organization (WTO) (Jung 2004). In addition, Regional Trade Agreements (RTAs) such as the Asia Pacific Economic Cooperation (APEC), Europe an Union (EU), the Mercado Comun del Sur (MERCOSUR), and North American Free Trade Agreement (NAFTA) have boosted trade among countries within a particular region of the world. International commodity markets represent an important focus of attention for the U.S. agricultural sector. Recognizing the significance of the factors affecting the flow of agricultural products is necessary in order to take advantage of the current commodity trade structure and predict its future evolution. New international trade policies, food security issues, and shifting

PAGE 13

13 agricultural commodities have become highly dependent on international markets. International beef trade has a long history of up and downs, but recent decades have seen fast growth of trade in terms of volume and value. Changes in trade policies have stimulated this trend; and associated with these changes, changes in diets, distribution technology, and multinational business structures have changed the global scenar io of beef trade either as causal factors or as con sequences of liberalized trade (USMEF 2007). Importance of Agricultural Trade in every corner of the world from developed to developing countries, and it has also affected production and distribution of agricultural products around the world. For example, soaring world prices of wheat have caused a 30 % increase in the prices of pa sta in Italy; and in Mexico, during the year 2006 unprecedented high prices of corn they almost quadruplicated have led to the worst tortilla crisis in its modern history 1 (Gumbel 2007, Roig Franzia 2007 ) Sanitary and protectionist trade barriers have strongly influenced agricultural trade. Sanitary standards are extremely important determinants of food related trade. For example, the distinction between countries free of foot and mouth disease (FMD) and those that are not free largely defines world tr ade in fresh, chilled, or frozen beef and pork. The danger of transmitting diseases to humans has also led to segmentation of the beef trade based on sanitary rules. The ce barred beef exports from much of Europe to other parts of the wo rld. In the U.S. case, after December 2003's outbreak, it closed the most important markets for U.S. beef around the world (OIE 2007). 1 Increasing prices are the result of a reduced sup ply. In the case of wheat (up 60% over the past year), this increase is due to a persistent drought in Australia, weak production years in Ukraine and Argentina, and in the case of corn it is due to an increased use of this commodity in the production of e thanol in the U.S.

PAGE 14

14 After almost three years of BSE related trade restrict ions and absence from key beef consuming markets, the beef industry is seeing some normalization in trade flows although it is still bearing many costs of increased product requirements and li mited product eligibility ( FAS USDA 20 07). Finally, in the last few ye a rs consumers in some markets focused their concerns on production oriented processes that emphasize issues related to animal welfare (e.g., the use of hormones and antibiotics, or the presence of genetically modified ingredients in feed rations) wh ich increasingly have become a basis for regulatory barriers to trade (Dyck and Nelson 2003). Recent advances in trade liberalization emphasize open access to world markets and expanding trade of agricultural commodities and processed food products. The ev olution of world and regional trade agreements have lowered protectionist barriers and facilitate d the flow of commodities. The 1994 WTO Agreement on Agriculture negotiated in the Uruguay Round addressed specific commitments to improve market access and re duce trade distorting subsidies in agriculture. Under this agreement, non tariff trade barriers are being converted into tariff equivalents, which gradually are being reduced or eliminated over some period of time. Import tariffs are considered more transp arent than non tariff measures, and because of that they have been a preferred policy instrument in multi lateral trade negotiations (Miljkovic and Jin 2006). However, significant protectionist barriers still remain, such as high tariffs, tariff rate quota s, and non tariff barriers, which prevent or inhibit significant potential trade in meats and are designed to discourage imported beef from competing with domestic products. International Beef Market Foreign demand for U.S. beef products has become an impo rtant factor affecting the U.S. domestic beef industry. Although the domestic market remains the dominant source of overall meat demand exports continue to account for a growing share of U.S. meat production. Despite higher prices, international beef cons umption has climbed as global economic growth supports

PAGE 15

15 increases in demand especially in Asian Pacific markets where the explosive economic expansion has contributed to an increase per capita income, encouraging reductions in trade restrictions and a shift in dietary preferences for animal source proteins (Capps et al. 1994). Economic development brings with it increased household income and consequently higher utility levels. With increased incomes, households can purchase more superior quality foods, suc h as beef. As in the cases of Japan and Korea, the growth in meat consumption pushes up prices of domestic beef, and increases the possibility of a market for imported beef. The importance of dietary changes to beef imports is shown by the case of Pacific Rim countries where import growth reflected the rapid increase in beef consumption that occurred there between 1980 and 1995, when consumption per capita increased almost six times. Consumption increased faster than production, and imported beef supplied t he difference. Figure 1 1 Domestic c onsumption of beef products i n Asian r egions. Source: FAS USDA. November 2007. While meat consumption growth appears to have leveled off since the late 1990s in western developed re gions (e .g., Europe and the United States ), meat consumption is currently growing in major parts of East, Southeast, and South Asia as economic expansion proceeds 0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000 9,000 10,000 1980 1985 1990 1995 2000 2005 Years Domestic Consumption (1000 MT) East Asia South Asia Southeast Asia

PAGE 16

16 (FAOSTAT 2007). Rising consumer demand for beef, aided by trade liberalization and changes in technol ogy (i.e., shipping and storage) has helped beef to grow across the globe (Figure 1 1) Beef trade mainly is in cuts or parts, not in the form of live animals or carcasses which generate extra revenue per animal. The slaughter of a meat animal automaticall y generates a full set of muscle meat cuts, as well as trimmings, offal, and other byproducts (Table 1 1). The value of a carcass is the composite value of the cuts and other products taken from it, and the derived value of a meat animal is the composite v alue of the carcass and byproducts from the animal, less processing and tra nsaction costs (Dyck and Nelson 2003). Table 1 1 Top 5 e xported beef c uts during 2006. Ranked by the e xtra value generated Cut Type Domest ic Use % Expor t U.S. Value ($kg) International Value ($kg) Total Extra Value ($ M) Short Ribs Trim 57 1.39 2.09 4.00 10.00 338 Tongue Pet Food 70 0.22 9.92 328 Outside Skirt Trim 61 2.09 3.50 6.00 166 Short Plate Trim 68 0.96 1.74 2.65 63 Chuckeye Roll As is 11 2.6 2 2.68 3.50 6.00 43 Source: USMEF. September 2007. In recent years, U.S. beef exports represented 12.4% of the production generating an additional value equivalent to 18% of the total value of a feed steer or the equivalent of an increase of almost one bi llion dollars out of the top five export beef cuts. Forecasts from the U.S. Meat Export Federation (USMEF) show that for the next five years a total increase of almost 90% in the volume of beef exported is expected and that Japan, Hong Kong, Taiwan and Ko rea are among the top five destinations ( USMEF 2007). Trade issues have become increasingly important for the U.S. beef industry, in particular those referred to as food safety have significant effects, either through government imposed trade restrictions or changes in import d emand (Verbeke and Ward 2001; Jin and Koo 2003).

PAGE 17

17 After the discoveries of BSE commonly referred to as mad cow disease, in the United Kingdom in 1998, Japan in 2001, and Canada and the United States in 2003, the response of the inter national community has permanen tly affected the world market for beef and led to a significant drop in beef import demand. These BSE cases prompted a number of tradit ional U.S. beef importing countries to ban almost all products from the United States, ope ning new markets or in some cases, incr easing existing market shares of competing beef producing nations like Australia, New Zealand, Brazil, and Argentina. The USDA ERS reported signific ant differences between the Pre and Post BSE ban of U.S. beef export s to these traditional markets showing an impress ive loss in value of almost $2.4 billion dollars. On the other hand, 2006 data from the USMEF show losses of approxim ately $ 1 billion or a total of 630,000 metric tons, which clearly shows the importance of these markets for the U.S. industry and the need to recuperate its position as primary supplier of beef products (Table 1 1 and Figure 1 2). Table 1 2. Trade status and BSE trade impact in millions of U.S. dollars by beef product Country Total Val Bone in Bone less Proces s meat Variety meats Date Opened Open Trade Status Diff. Pre Post Notes Japan 1,391 2 50 .1 1,103 13 .5 224 .5 12/05 1,391 .2 Partial Ban A ll <21m Korea 814 5 291 .5 49 .0 8 .7 65 .2 9/06 449 .0 Parti al Ban 365 .5 Bone l ess <30m Hong Kong 90 8 19 .4 48 .4 521 22 .3 12/05 48 .4 Partial Ban 42 .3 Bone less <30m Taiwan 76 3 13 .5 55 .9 873 5 .9 4/ 05 55 .9 Partial Ban 20 .3 Bone less <30m ROW 1,482 6 1,357 .6 125 .0 Total 3,855 6 3,302 .3 553.3 Source: USMEF. September 2007. Figure 1 2 provides value loss share and illustrates the economic impact of the BSE on each segment of U.S. beef exports during the year 2006. As expected, fresh/chilled and frozen

PAGE 18

18 beef products represent the larges t share in exports revenues due mainly to premium prices paid by international consumers for U.S. beef cuts. Figure 1 2. Economic impact of the BSE b an across U.S. beef export products as January, 2006. Source: USMEF. September 2007. Today, U.S. beef ex ports primarily reflect demand for high quality fed beef, with most U.S. beef exports typically going to Mexico, Canada, and markets in Pacific Rim nations. The global U.S. beef exports are projected to increase slowly and recovery is assumed in the Japane se and South Korean export markets up to volumes similar to those before the first U.S. BSE case in December 2003. Beef trade flows among countries and world regions are determined largely by differences among countries in their resource base, their prefer ences for cuts, food safety issues, the extent and character of barriers to trade, and the industry structure. Future growth of beef trade depends on further liberalization of protectionist barriers, eradication of animal diseases, economic development, an d population growth. Beef Products $553,303 (57%) Live Cattle $31,929 (3%) Meat and Bone Meal $143,957 (15%) Pet Food $136,629 (14%) Fats and Greases $31,929 (3%) Tallow $47,033 (5%) Other Ruminant Products $25,755 (3%) Total Banned Amount = $ 970,495

PAGE 19

19 International U.S. Beef Promotions Purchase decisions are based on predictions of product performance. Consumers base their predictions in part on product cues and are accurate to the extent that they have properly learned the relatio nship between the cues and performance (Van Osselaer and Alba 2000). If consumers learn the relationship between product attributes and quality, they will differentiate among goods that possess different attributes and treat as close substitutes those that share the same attributes. During the past twenty years, the U.S. beef industry has experienced a slow transition from a traditional commodity selling perspective t o a more contemporary marketing driven sector. This evolution has encouraged the industry t o differentiate its products, review production processes, reorganize its value chain, and use a mix of price and non price marketing strategies in order to increase market share. Competition has become much more intense and the demands for red meat produc ts, in particular fresh, chilled, and frozen beef reveal the distinctive characteristics in product quality and production technology across exporting countries. Although only 12 % of the beef produced in the United States is exported, international market participation has an important impact on the industry because it raises the value of all products. It increases the number of customers which consequently boosts prices throug h a positive shift in aggregate demand and it allows packers to sell differentia ted products according to preferences in each international market, thus obtaining higher rents from those customers willing to pay global prices (Kinnucan et al. 1995). Advertising and promotion provide useful information necessary to facilitate purchasi ng decisions, and in some cases they are even used to change the underlying preference function for a particular product or service. Commodities in general, and specifically beef products, can be differentiated according to how consumers acquire informatio n and how this information affects

PAGE 20

20 Kong, and Taiwan, U.S. beef is highly appreciated because of its particular characteristics of taste, fat content, tenderness, an more than two decades (USMEF 2007). Studies before the BSE outbreak have shown that the most important characteristics for consumers of U.S. beef in these markets are that it tastes good, it lo oks fresh, does not have a lot of waste, and that it is USDA certified and inspected. Marketing oriented to recuperate market shares should reinforce these attributes, emphasizing the characteristics of the product in order to re build the positive percept ion about U.S. beef products. Promoting U.S. beef as a brand might be necessary to reach pre ban levels of consumption in these countries using product differentiation based on the attributes of the product rather than simply selling beef as a commodity. T he U.S. beef checkoff promotion programs have been an instrument for providing consumers with information about the attributes of U.S. beef. As in many promotion programs, the main objective of the beef checkoff program is to increase the demand for beef p roducts in the domestic market as well as internationally. The USMEF, the international arm of the U.S. beef board, has launched several marketing campaigns abroad to keep U.S. beef product s in front of consumers in markets that are still closed or to incr ease existing demand. Lately, efforts are aimed to overcome the trade barriers imposed by the EU on U.S. beef over the issue o f using hormones in cattle production, and age restrictions imposed by several countries in the Pacific Rim region. Problem Statem ent and Research Objective International beef trade has shifted from a domestic or sometimes regional commercial activity to a global network of buyers and sellers that continues to grow in complexity and importance. The international marketplace for beef products is influenced by several aspects including changes in consumer demand and preferences, national and international trade policies,

PAGE 21

21 macroeconomic characteristics of the country, and the increasing concern about the quality and safety of the food rea ching each household. Many studies have used trade models to estimate the impact of tariff and non trade barriers (NTB) policies on international commodity markets (e.g., Webb et al. 1989; Brester 1996; and Devadoss et al. 2006), where international grain embargoes, domestic price insulation policies, and food safety issues were the main reasons for trade disruption. These studies assumed that allocation of expenditure within a sector takes place in two stages corresponding to two levels of a utility funct ion. The upper level defines preferences based on the origin of the product differentiating domestically produced output and an aggregate of imported goods from all sources. Once expenditure is allocated to aggregate imports, the lower level defines prefer ences over different sources of supply. An empirical measurement of the import demands in each country is necessary to evaluate the impact of agricultural policies and to outline their implications. For this research, a Computable General Equilibrium mode l structure (CGE) based on previous research by Paul Armington (1969) is suggested. This specification is based on a theory of demand for products that differentiates them according to their kind and country of origin. The specification includes a one prod uct, eight region model in which elasticities of substitution for beef products from the United States, Australia, New Zealand, and Rest of the World (ROW) to markets such as Japan, Korea, Hong Kong, and Taiwan are calculated based on annual data for 1995 to 2006. Elasticities of substitution are known to be key behavioral parameters for trade policy analysis; hence, estimation and analysis of these parameters is necessary since it provides essential information to understand better the demand for beef and trade policies. Changes in trade policy affect the availability of a product in a particular market; and the size of those

PAGE 22

22 impacts largely depends on the magnitude of the elasticity. Specifically, the elasticity of substitution in the Armington model is a ssumed to be imperfect according to a single constant elasticity of substitution and homotheticity of preference. These elasticities will provide a benchmark for CGE models of beef trade providing information about market share decrease, the effect of tari ff and non tariff barriers (NTB) (e.g., beef import ban), and other factors affecting the import demand for U.S. products. Under the assumption of national product differentiation, the main objective of this study is to look into the international beef ma rket, analyze the impact of the BSE outbreak (December, 2003) on the import demand for U.S. beef products in selected markets, and determine the role of U.S. beef promotions on recover ing pre 2003 international market levels in terms of value and volume. Specific objectives for this research are: 1. Describe the condition of the beef industry in the United States and overseas. 2. Analyze the structure of the international beef market. 3. Develop the theoretical demand and export supply models for beef products in selected countries. Model specifications should reproduce demand differences across countries s, product quality and non tariff barriers The specifications should also reproduce the supply capacity of each e xporting nation and the responsiveness to changes in average prices. 4. Perform an empirical demand analysis on beef using econometric models in order to get the input parameters or elasticities required to estimate the impact of trade disruption or and the effectiveness of U.S. beef promotions in selected markets. 5. Suggest potential improvements for beef marketing overseas and provide trade policy recommendations. Scope The purpose of this research is to estimate empirically the import demand and su pply for beef products for some of the most important beef trade partners of the United States. The model suggested takes into account factors such as governmental non tariff barriers policies, relative

PAGE 23

23 prices and values, total market size of the industry, and expenditures on promotions and other marketing activities. Yearly export and import data collected from the World Trade Atlas (Global Trade Information Services, Inc.) during the 1995 to 2007 period is used in this research. The quantities and prices of beef products from the four producing regions to the four importing countries will be analyzed at the H.S. 0201 and 0202 levels 2 that d efine meat of bovine animals as fresh or chilled and frozen. Market shares for each region will be estimated within the construct of the model to allow for the estimation of consistent parameter estimates. The analysis will clearly illustrate how consumers in selected markets differentiate product s by their origin of production and how promotions can affect market share s for U.S. products. Figure 1 3. Beef trade p articipation of selected countries from 1980 to 2007. Source: USDA ERS. November 2007. 2 Harmonized System code of the product (HS) 34.8% 22.0% 0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0% 1980 1983 1986 1989 1992 1995 1998 2001 2004 2007 Years Percentage of global market share Percentage of World Exports (U.S., Australia, and New Zealand)

PAGE 24

24 Finally, it is important to mention at this point that the scope of this research is limited to four importing countries: Japan, Republic of Korea, Hong Kong, and Taiwan; and four exporting countries or regions: United States, Australia, New Zealand, and the Rest of the World (ROW). Using the data set obtained from the United States D epartment of Agriculture, For eign Agricul tural Service (USDA, FAS), Figure 1 3 shows that in 2007 those four importing countries accounted for 22% of all beef imports and the three exporting countries provide d almost 35% of all beef exports in that year. Together, these two groups of countries are considered among in this research. Research Methodology A comprehensive literature review of the international agricultural trade, i n particular the beef industry, and commodity promotion theory will be conducted to formulate hypotheses and evaluate empirical findings. Topics related to international livestock production, beef processing, retailing, and beef marketing management will b e discussed to provide a theoretical background and facilitate a further analysis of the resulting data. For this study, an empirical demand analysis is conducted on beef products in selected markets. A parametric analysis that includes a global trade mod el based on one outlined by Armington in 1969 will be used to evaluate changes in trade flows in general equilibrium models. Parameter estimates from empirical demand models (i.e., elasticities of substitution) are used to simulate exogenous trade impacts (i.e., BSE, NTB, etc.) and marketing efforts aimed to compensate for trade anomalies. Price elasticities measure the responsiveness of trade flows to price changes. Elasticities of substitution provide the cross price elasticity between products from diffe rent origin and measure the degree to which lower prices would provide some goods greater market share in a particular market. Since one of the objectives of this research is to

PAGE 25

25 measure the impact of restrictive policies on beef trade, the size of those im pacts largely depends on the magnitude of the elasticity (McDaniel and Balistreri 2002). The methodology suggested for this research distinguishes products by place of origin and uses a Constant Ratio of Elasticity of Substitution (CRES) functional form to bring prices, quantity, and other factors into the model, in order to measure trade allocation and demand preferences for beef products. In this model, total import demand for beef will be determined for each country and then it will be independently allo cated among competing sources of supply (Sparks and Ward 1992). However, this model represents a non linear system with a set of equations that are non linear in the parameters and their corresponding restrictions within each set of equations. Econometric procedures will be used to estimate product demands and market share equations to assure consistency results. The resul ting model and its estimates will be used to develop a sensitivity analysis to examine the effects of changes in selected factors on bee f trade and consumption. Armington Trade Model In the international food market, countries are assumed to behave competitively and commod ities like beef are usually seen as homogeneous goods. That is, similar products from different origins are considered perfect substitutes and the corresponding price ratios are constant in that particular market 3 However, reality shows that consumers tend to behave and adopt food products based on well defined personal preferences constructed upon perceptions of the attr ibutes of the product which, at the end, make competing products behave as imperfect substitutes. 3 The assumption of perfect substitution among products implies the presence of infinite elasticities of substitution.

PAGE 26

26 A model which recognizes that commodities may be imperfect substitutes was developed by Paul Armington in 1969 to analyze the world trade demand for machiner y and chemical products using Japan and France as differentiated sources of production. Moving away from the goods representing a group of products such as machine ry, meats, or appliances. Within a group category, products are not considered perfect substitutes; they are differentiated based on their region or country of production, but due to similarities in their core attributes, they belong to the same category. Consequently, this model suggests that one country's beef market is likely to be composed of demands for beef products originated in several distinct and independent regions. The strength of these individual product demand functions is largely determined b y the size of the market in the demanding country and the size of each country's beef market is affected by many variables with the most important including national incomes, average beef prices in the region, population levels, and the prices of substitut e goods. Armington proposed that in order to determine the overall demand for any product in any particular trade area and calculate market shares, some fundamental assumptions must be is independent of those demands for products in other goods markets (e.g., U.S. beef and Australian beef). Thus in any exporter nation, each industry produces only one product and this product is different from the product of any other country (i.e., separ ability assumption) (Lloyd and Zhang 2006). Second, the model assumes that as long as the prices of products within a category remain constant relative to each other, the shares of those products will not be affected by changes in the size of that market. This assumption implies that changes in prices of competing products affect product demands only indirectly through their influence on the total market size and that market shares are not

PAGE 27

27 affected by price fluctuations (i.e., homotheticity assumption) (Yan g 1993). What determines these initial shares in the Armington model specification is, however, of major interest for this study. Armington noticed that some issues of substitutability might arise as a consequence of the presence of large number of supplyi ng areas competing in the same market. Consequently, in order to make the model empirically feasible, additional assumptions were introduced in the specification to limit the substitutability and simplify product demand functions: (a) elasticities of subst itution between products competing in any market are constant (i.e., Constant Elasticity of Substitution (CES) assumption), therefore they do not depend on market shares; and (b) the elasticity of substitution between any two products competing in a partic ular market equals the elasticity of substitution between any other beef products in the same market. These assumptions yield a specific and very restrictive form for the relation between demand for a product, the size of the corresponding market, and rela tive prices; and the only price parameter in this function is the single CES coefficient in that market (Armington 1969). However, as argued in previous studies, while these assumptions are very useful in limiting the number of parameters in the model an d facilitating the estimation process, they are e xtremely restrictive and somewhat unrealis tic (Winters 1984; Alston et al. 1990). The separability assumption omits prices of substitutes which are likely to be correlated with the own price elasticity (i.e. positive cross price elasticities). When prices of substitutes are omitted, the own price parameter estimates are positively biased so that the own price elasticities would be underestimated (Green 2003). Although the homotheticity assumption simplifies the model the same proportion to all products. If an increase in the budget is realized, it would be allocated

PAGE 28

28 more to the more preferred product. In other w ords, expenditure elasticiti es may not be unitary (Hayes et al. 1990). The single CES assumption restricts responses of the import demand of each product to the price change relative to the price index for the good to be the same for all products (Weathe rspoon and Seale 1995). In addition, under the CES assumption the level of differentiation among products is only established by the place of production which implies the presence of a weak separability between different import sources that entails the con decision process viewed maybe as occurring in two stages (Varian 1992 ) As a result, in 1989 introduced a two step process to estimate the demand for the pro duct examined. In a first step, the demand/expenditure of a group is determined regardless of the source of origin. In a second step, given the assumption of independence, this demand/expenditure is allocated among the identified suppliers based on the pla ce of origin and/or the form of the product. This stage is a function of relative exporter prices, and it determines the market shares that each exporter will have in each importing region. Therefore, one of the major considerations of the Armington model is that the marginal rate of substitution between any two products of the same kind is independent of the consumption of other kinds of products (Alston et al. 1990; Blonigen and Wilson 1999). In reality, product differentiation is based not only on count ry of origin but also is based on differences in quality levels between products produced in different countries and the type of products consumed in a particular country (Reinert and Roland Host 1992 ) A more realistic and less restrictive approach was su ggested by Artus and Rhomberg in 1973 using a Constant Ratio of the Elasticity of Substitution (CRES) in which the elasticities of substitution between any pair

PAGE 29

29 of competing products might vary by a constant proportion, but the substitutability between eve ry product is not necessarily identical (Hanoch 1971 ) This framework would allow for those suppliers who produce unusually high or low quality beef and those whose products are distinguishable, but not highly differentiated (Sparks and Ward 1992). The the oretical groundwork for a model that deals with the differentiated commodities as suggested by Armington (1969) has generally been limited to using single equation estimation techniques (Art us and Rohmberg 1973). However, Winters (1984) estimated the deman d and supply of these exports using both single equation and systems techniques. However, most of these stud ies showed little recognition of external shocks or underlying structural changes that could produce changes in the perceived differentiability amon g the competing goods. This study proposes the use of a CRES functional form to estimate the parameters of a beef trade model which distinguishes products according to their country of production and measure s the impacts of exogenous policy shocks and gen eric promotions on import demands for beef products in selected U.S. markets. The choice of the CRES as the appropriate system estimation procedure is justified given that it allows a constant ratio of elasticity of substitution between any pair of competi ng beef products, without restricting the substitutability to be the same between every product. Hence, the CRES function will fully exploit the mathematical attributes of Armington's theoretical framework and will yield more consistent parameter estimates than would a single CES functional form. Data Set This study will examine world trade in beef products over several years and evaluate several databases to cross verify the quality of the information used in the analysis. Yearly data covering the period 1995 through 2006 were obtained from the United St ates Department of Agriculture, Foreign Agriculture S ervice using the World Trade Atlas program developed to

PAGE 30

30 collect U.S. and international export and import data on agricultural trade. In addition, select ed demographics, national economic indicators, and information on tariff and non tariff barriers were collected using the FAOSTAT database from the United Nations, the U.S. Department of Commerce, and the World Organization for Animal Health (OIE). For eac h of the selected international beef markets, the World Trade Atlas database records import shipments using the Harmonized System code of the product (HS) 4 at the 0200 level, the volume in metric tons, the CIF 5 value in US dollars, the country of origin, a nd a text description of the product. In addition, yearly information collected from country specific sources (i.e., U.S. embassies, U.N., and OIE regional/country offices) through field representatives is used to complement the original dataset. Missing d ata information is the result of lack of appropriate statistical records at country levels and in our case they represent a minimal proportion of the total data set. Finally, the total dataset for this research include s around 25,000 observations from whic h market share equations are estimated using more than 2,230 observations containing demographic variables. Outline of Chapters The importance of trade as the underlying force behind globalization has been presented those affecting international beef markets. Chapter 1 thus provides the background for this project. Chapter 2 discuss es the international beef industry and the characteristics of e ach import and exporting market during the period between 1995 trough 2007. This chapter reviews each beef supply industry, demand for beef in each importing nation, and international regulatio ns. 4 Governments use HS codes to classify trade into discrete categories for customs duties and for cataloging exports/imports. 5 CIF prices include cost, insurance, and freight

PAGE 31

31 Chapter 3 provides a review of the relevant literature on agricultural trade and promotions. distinguished by place of production including a full description of the conc eptual model and discussion of the relevant econometric issues of the model. In Chapter 5, the final econometric models are estimated and statistical, graphical and economical interpretations of the findings are presented. Chapter 6 presents simulations of the effects on the beef markets of the changes in exogenous variables. Finally, Chapter 7 presents the summary and conclusions of the study.

PAGE 32

32 CHAPTER 2 WORLD BEEF TRADE AND PRODUCTION Introduction The purpose of this chapter is to present an overview of t he world beef market based on descriptive statistics and trade data gathered from public and private international organizations. In particular, in this segment the objective is to describe trade flows between a particular set of countries that were previo usly defined as regions of interest. The discussion that follows includes a description of the data set with specifications on classification and characteristics of the traded beef products. In addition, country specific demographics will be presented acco mpanied with an analysis of relative prices, expenditures on beef products by country, volume traded, market shares, and values of export imports during the period 1995 to 2007. International Beef Trade Data For this research, the data set is based on in ternational trade statisti cs from four importing and three exporting countries who reported quantities, import export values, and average prices for beef products in their respective markets. All trade data are expressed in metric tons (MT) and values in t housands of U.S. dollars. The data set for world meat trade was obtained from The United States Department of Agriculture, the Foreign Agricultural Service, and the Economic Research Service using the World Trade Atlas Program. This data set was developed by Global Trade Information Services, Inc. (GTI 2007 ) to provide information on inte rnational merchandise trade to governments and corporations. Existing studies on international food trade suggest that agricultural world trade can be divided into region s using the criteria of geographic differences, the pattern of trade flows, and regions: South and North America, Asia which includes South and South East As ia, East Asia,

PAGE 33

33 the European Union (EU), Oceania, and the Rest of the World (ROW) which includes less significant beef producing regions/countries. In order to collect and analyze the information from this data set, beef products are classified using the H armonized Commodity Description and Coding System (HS) of tariff nomenclature, which represents an internationally standardized system of names and numbers for classifying traded products ( WCO 2008). According to this nomenclature, fresh/chilled and frozen red meats belong to the 02 group containing meat and edible meat offal. At this level, the code 02 is made up of all aggregate meat products and the objective of this study is to provide an analysis at a more specific level. Consequently, meat products of bovine animals are defined as part of the 0201 group if they are fresh or chilled; as part of the 0202 group if frozen; and as part of the 0203 group if they are meat of swine (fresh/chilled or frozen). The reasons for selecting these classification level s are mainly twofold. First, products within the 0201, 0202, and 0203 codes are the major U.S. meat export categories. Second, they provide an excellent base for applying the Armington specification for product differentiation. In particular, for this rese arch the 0203 code is mentioned only as a reference since the main objective is to focus on the impact of the BSE disease which only affects bovine animals. In addition, selected demographics, national economic indicators, and information on tariff and no n tariff barriers are included using the FAOSTAT dat a base from the United Nations, the U.S. Department of Commerce, and the World Organization for Animal Health (OIE). nd Research Board are included in this research to evaluate their impact on beef demand in selected foreign markets.

PAGE 34

34 This study deals with total trade data without regional bi lateral trade and re exports (i.e., Japan exports to Hong Kong or U.S. imports from Australia) to avoid the problem of double counting and demonstrate s the relative strength of each country in the international beef market. Therefore, exports of any given country to any partner country (or importing country) are defined as exports wi thout intra regional trade. The International Beef Market During the 13 year period under consideration, a number of significant changes have taken place in the world trade environment. One of the most important changes is the significant shift in the dem and for resources, in particular with respect to food and energy, from western economies to emerging Asian economies like India, China and many other nations on the Pacific Rim. These emerging economies are absorbing a large portion of the world supply of agricultural commodities helped by technological improvements in storage and transportation methods. For example, the average beef consumption in places like Hong Kong and Taiwan has increased almost 63% during the past 10 years while other developed weste rn nations have shown an opposite pattern during the same period. Beef exporting nations have shown a significant change in trade flows during the past decades. Before the 2003 BSE outbreak, the U.S. was considered among the top three beef exporting countr ies in the world with significant presence in Asia, Africa, and Central America. Today, Brazil is considered the largest exporter of red meats products followed by Australia, which clearly shows the economic impact of food safety issues and how countries a round the globe are adapting their policies to prevent future outbreaks. In order to see the impacts of these policy changes on trade patterns, this section shows beef trade flows between selected countries during the period 1995 to 2007 in terms of quanti ties traded, market shares, prices and expenditures, and consumption levels for the years before and after the BSE market restrictions policies took place.

PAGE 35

35 Volume, Share, and Growth of Global Production Tables 2 1, 2 2, and F igure 2 1 show the world total production by region, the share of each region, and the rate of growth for the years 1995 and 2007 in terms of quantities and percentages respectively. The world beef production totaled 48.53 millions of metric tons (M.MT) during the year 1995 and 54.48 M.MT during the year 2007, representing an increase of almost 11%. The Americas region concentrated more than one half of the world production with more than 50% of the share in 1995 and almost 53% in 2007, where the South American portion has shown the la rgest boost in production share with almost a 4% increase. In particularly, the North American region, which includes the U.S., Canada, and Mexico emerged as the largest producing region, while the South American region that includes Brazil, Argentina, Par aguay, etc., appears as the g region during the same years, Figure 2 1 Figure 2 1 T otal beef p roduction and s hares by r eg ion. Source: USDA FAS. January 2007.

PAGE 36

36 In 1995, the next largest share was for the European Union (EU) region with 18.5% followed by East Asia (10.3 %), Oceania (4.8%), Asia (2.6%), and the rest of the regions of the world with 13%. In 2007, however, East Asia (15.7%) surpassed the EU (14.7%) as the second largest producing region of the world followed by Oceania, Asia, and the rest of the world. With the exception of the EU and some small producing regions included in the ROW category, all other regions of the world show important advances in terms of quantity produced which guarantees a healthy supply and demonstrates the increasing importance of this commodity in Table 2 1. Compared world production levels and trade from 1995 to 2007 Source: FAS USDA. January 2008. Asia* represents values of South and Southeast Asia. Table 2 1 shows the imports to exports ratio 2007/1995, which is interpreted as a growth rate indicator for production le vels. Thus, production grew globally at a rate of 1.123 from 1995 to 2007 or in other words, the 2007 production level was 1.123 times the 1995 level. The Asian region shows the largest improvement in terms of production and that traditional beef producing regions, like North America and the EU, have been surpassed by competing regions, like South America and Oceania showing an important shift in the world beef supply. Using Figure 2 2 to support the discussion, production levels have increased at a rate of 1.2, 0.71, 0.33, and 0.25 times in Asia, East Asia, South America, and Oceania respectively, while the North American region, the EU, and the rest of the world (ROW) have shown small or negative increments during the same period. 1995 2007 2007/1995 Region 1000 MT % of World 1000 MT % of World Ratio S. America 9,9 75 0.206 13,235 0.243 1.327 N. America 14,363 0.296 15,514 0.285 1.080 Asia* 1,239 0.026 2,724 0.050 2.199 East Asia 5,003 0.103 8,574 0.157 1.714 European U 8,985 0.185 8,000 0.147 0.890 Oceania 2,347 0.048 2,921 0.054 1.245 ROW 6,626 0.137 3,521 0. 065 0.531 World 48,538 1.000 54,489 1.000 1.123

PAGE 37

37 Figure 2 2. Relative v ariation in growth rate of p roduction (2007/1 995). Source: FAS USDA. January 2008 Beef Trade Patterns This section presents a general description of each region in terms of trade, trends, and ratios between imports and exports volumes. Figure 2 3 and Tabl e 2 2 are used to compare levels of trade and explain the significance of each region in terms of imports and exports. The dataset shows that beef trade has expanded over time and that market shares between regions have changed. Rising demand for meat prod ucts in different parts of the world and increasing concerns about the safety of the beef supply chain have prompted nations to re define trade alliances in order to satisfy beef consumers and ensure the quality and safety of the products being traded. Usi ng trade statistics for the seven regions previously defined, it is possible to identify trade trends for the years before and after the BSE outbreak in December 2003.

PAGE 38

38 Figure 2 3. Pre and post BSE beef trade levels by region. Source: USDA FAS. February 2008. Exporting regions like North America, South America, Oceania, and Asia revealed significant levels of growth in their export markets, although in the case of the North America this trend drastically changed after 2003. Other regions such as Europe a nd the ROW presented an opposite trend, which is interpreted as a loss in market participation due to the negative impact of food safety issues affecting their supply chain, like in the case of t he BSE epidemic in the United Kingdom with al most 1,000 new c ases per week January 1993 (OIE 2007 ). Figure 2 3 presents the import and export quantities for years under consideration. During the year 1995, the major import regions were North America and East Asia followed by the European Union. These three regions i mported 1257, 1295, and 494 thousands of metric tons respectively representing 29.5% 1 30.4%, and 11.6% of the world imports respectively; during the year 2003, 2048 (39.6%), 1495 (28.9%), and 549 (10.6%) thousands of metric tons respectively; and in the y ear 1 Percentage share of world imports/ exports for beef products.

PAGE 39

39 2007 their import quantities were 2096 (37.3%), 1240 (22.0%), and 725 (12.9%) thousands of metric tons respectively. Together they comprised 71.5%, 79.1%, and 72.3% of the world imports for beef products, which once again illustrates the importance of these regions in term of beef trade. The quantities of world exports originated in South America, North America, and Oceania increased from 1995 to 2003 reflecting the growing demand for beef products in Asian nations, while European exports decreased due to food safety issues. During the 2003 to 2007 period, exports from South America dominated the world market, while exports from North America plummeted. In fact, the strongest region in terms of export growth was the South American Region. With a growth rate of more than 50% during the 1995 to 2003 period, South America exported approximately 28.9% of world totals; and during the 2003 to 2007 period this rate increased once again by almost 60% corresponding to a 44.2% of world totals. The North America re gion, while it increased quantities exported and world market shares from 1995 to 2003 by 31% and 19.7% respectively, in the next period decreased to 24.8% showing an absolute growth rate reduction of 9.2% in market share. Finally, Oceania also showed stea dy growth in terms of volume exported (1606, 1789, and 1965 thousands of metric tons respectively), although in terms of growth rates it showed a reduction during those years (29.4%, 26.1%, and 28.3% respectively ) Table 2 2 presents the import to export ratio for each region considered for this study. Thus, for a given year, a ratio with a value below 1 defines a p articular region as net importer and conversely, a ratio with a value above 1 describes a net exporting region for that year. Analyzing the obt ained ratios, it is clear that regardless of their production levels North America, East Asia, and, lately, the EU are considered net beef importing regions in terms of volume

PAGE 40

40 (quantity) traded, and that South America, Oceania, and Asia have maintained the ir characteristics of traditional beef exporter over the 13 year period considered for this study. Table 2 2. Ratio import to export quantities by r egions from 1995 to 2007 Source: FAS USDA. November 2007. Asia* represents values of South and Southeast Asia. In the case of the of the North American region, it is clear that after the 2003 BSE scare the trade ratio increased due to a significant reduction of the quantities exported by the United States and Canada, and to a strong domestic demand in the EU. Of particular significance is the situation in the East Asian market where the increasing import trend was disrupted and reduced by more than 50% after the BSE outbreak. On the other hand, after December of 2003, a positive ramification of this food safety issue can be perceived on the beef trade ratios from regions such as South America and Oceania characterized by their grass feed beef type products. Conse quently, it is important to describe the situation s in the North American region and the East Asian region, where the major U.S. beef markets are located ; and the Oceania region where competing beef products are produced in order to understand the dynamics of trade in this part of the world. Traditionally, t he North American region has been considered one of the largest beef producing regions of the world. The United States has been always been characterized a s a net exporter of beef in terms of export valu es, which before 2004 had generated a large trade surplus Year S. America N. America Asia* East Asia EU Oceania 1995 0.175 1.169 0.470 12.1 03 0.400 0.010 1996 0.270 1.082 0.107 11.819 0.385 0.010 1997 0.179 1.110 0.585 14.567 0.427 0.007 1998 0.190 1.220 0.386 14.893 0.503 0.007 1999 0.076 1.196 0.552 28.180 0.441 0.007 2000 0.085 1.238 0.355 32.979 0.647 0.008 2001 0.062 1.329 0.304 26 .396 0.684 0.009 2002 0.064 1.295 0.302 36.027 0.917 0.012 2003 0.041 1.307 0.294 41.528 1.253 0.010 2004 0.023 2.513 0.327 19.865 1.766 0.010 2005 0.020 2.241 0.222 14.842 2.810 0.010 2006 0.012 1.896 0.199 13.871 3.319 0.009 2007 0.010 1.784 0.221 12.525 4.143 0.009

PAGE 41

41 over the years However, U.S. beef imports have exceeded exports in terms of volume The United States ex ports high quality beef products and imports a large quantity of beef that is used to be comb ine d with domestic beef to make ground beef Much of the increase in beef imports has been due to competitively priced ground beef from Australia and New Zealand, the elimination of U.S. tariffs on beef from Canada, and the implementation of fresh beef quo tas for countries like Uruguay. Today, Australia, New Zealand, and Canada are the largest suppl iers of beef to the U.S. ( ERS 2007). Before December 2003, the East Asian region was the most important market f or the United States, importing 41.52 times more than it exported with most of the trade being concentrated between the United States, Japan, and the Republic of Korea. Today, beef imports from Australia and New Zealand have displaced the American beef presence in the Pacific Rim region although consume rs in these markets have always demonstrated a preference for U.S. beef (USMEF 2007). Lastly, the 2001 and 2002 cases in England and Belgium had a highly significant and negative impact on the EU beef industry causing EU exports to plummet. To summarize, based on the volume of beef traded, the fastest growing region s on the import side have been the United States, East Asia, and the European Union (EU), and on the export side, South America, Oceania, and Asia have shown the strongest growth on beef exports during the same period. Overall, trade restrictions due to decreased outbreaks have impacted global beef trade and traditional trading relationships. These circumstances have generated negative consequences for the United States and other beef suppliers b ut, in other cases, they created new market opportunities as in the case of Brazil and Argentina. In 2001, these two nations only accounted for 16% of the global beef trade, but today they account for more than 37% of the beef traded (FAS 2007).

PAGE 42

42 Consideri ng the recent political and economical changes worldwide, the supply advantages of each meat producer country and the increasing demand in different region of the planet, this study will focus its attention towards a beef trade model that includes four Asi an importing countries Republic of Korea (KR), Japan (JP), Republic of Taiwan (TW), and Hong Kong/ China (HK) and three exporting nations United States (US), Australia (AU), and New Zealand (NZ) and a region that aggregates the rest of the beef exp orting countries into a Rest of the Word category (ROW). These countries were selected according to the following criteria: the strength and manner of participation in the international meat market; the nature and scope of the BSE ban on their beef trade p atterns with the United States and its competitors; and the impact of international checkoff programs on selected Asian nations. The following sections of this chapter will describe the Pacific Rim Beef Market characteristics. This analysis includes statis tics generated from the dataset that will be used to evaluate and compare the selected nations, and to provide the background necessary for the study. Pacific Rim Demand for Beef Products Four Asian countries were selected according to the volume and valu e of beef products imported and expenditure levels on beef promotions by the exporting country. In this section each of the selected countries Japan, the Republic of South Korea, the Republic of Taiwan, and the Special Administrative Region of Hong Kong/C hina is discussed separately in terms of production, imports volume, and consumption levels. Figures 2 4 through 2 9 present these statistics for the years 1995 through 2007 which are used to explain market characteristics of each participating country and provide a background for the specification of the Armington model.

PAGE 43

43 Republic of South Korea growth before and after the BSE scare in 2003. During the period between 1995 and the first half of 1997, the strong e co nomic growth in Korea positively affected the demand for beef impacting prices and, consequently, stimulating domestic production (Figure 2 4). According to a report prepared by the USDA FAS in 1996, government re gulated import policies aimed to protect impact on domestic cattle prices forcing the government to soften import regulations and allow foreign beef to access the market at levels well above the established import quota level. The economic crisis between the fall of 1997 and the beginning of 1999 further stressed this market forcing imports and consumption to drop 44% and 11% respectively. The effect of Korea's economic crisis on do mestic cattle production revealed itself in inventory trends and slaughter levels showing a peak of almost 11% in 1998 In 2000, beef into a year long downturn. Beef co nsumption plunged at the beginning of the crisis and then gradually recovered during 2001 showing a significant consumer demand for beef given the delicate economic situation. During the following years, total beef consumption gr ew steadily mainly due to an improving economy that helped to boost the demand of food products in Korea. Nonetheless, during the year 2003 a sharp decrease in beef consumption due to BSE concerns affected the demand for imported beef products, in particul ar from the United States and Canada. Official estimates indicate that beef consumpti on and import levels dropped sharply early in 2004 due to food safety concerns, including BSE and E. coli (FAS 2004) which is clearly revealed in Figure 2 4.

PAGE 44

44 In recent ye ars, the demand for beef has been on the recovery showing rising consumption and import levels. Today, Australia, New Zealand, Brazil, and other beef exporting countries have displaced the United States as major beef supplier, although bilateral negotiatio ns are being conducted to reestablish the presence of American products in the Korean market. With respect to relativ e prices in the Korean market, Figure 2 5 illustrates a similar trend for American and world average prices for beef products during the p ast years. The increasing demand for American beef before the 2003 BSE announcement has pushed prices up, which is interpreted as a clear signal that consumers in Korea have a marked preference towards the unique attributes available only on U.S. beef prod ucts (i.e., fat content, tenderness, flavor, etc). Figure 2 4. Republic of Korea beef market indicators. Source: USDA FAS. November 2007. In the Korean beef market, prices for Australian products have shown a fluctuating trend with reduction of more tha n 42 % between 1999 and 2002. However, after the BSE outbreak in North America, Australian beef prices returned to previous levels showing an increase of more

PAGE 45

45 than 40%. These price fluctuations might be a consequence of the presence or absence of beef prod ucts from competing nations in the South Korean market. Despite the higher prices, beef products from New Zealand have shown an important participation in the Korean market, although during the past eight years a strong market competition has for ced its pr ices to drop more than beef import price. Figure 2 5. Average CIF prices for fresh beef products in Korea from 1999 to 2006. Source: USDA FAS. February 2008. Japan Beef m arkets in Japan have yet to return to level s reached before 1995, prior to the outbreaks of BSE in Europe and E. coli in Japan (Figure 2 6) Despite the dramatic effects the BSE scare and the E. coli epidemic have had on consumer beef purchases, total beef consum ption between 1996 and 2000 was clearly on the rise by more than 8.3%. However,

PAGE 46

46 Japa n's domestic beef production continue d its downward trend during the same period and imported beef continued to account for a greater share of Japanese consumption T he discovery of Bovine Spongiform Encephalopathy (BSE) in Japan in 2001 caused an immediate drop in consumption and reduced import levels during the year; ac cording to Japanese household consumption statistics, beef consumption for April and May (immediat ely after the BSE scare began) declined about 15% each month, while consumption of other meats declined only slightly. On the production side, almost ironically, the data showed an increase of nearly ivated by concerns that imported BSE infected beef may appear in the Japanese food supply. Figure 2 6. Japan beef market indicators. Source: USDA FAS. November 2007. After the 2003 BSE outbreak p ersistent concerns about the safety of imported beef forced the Japanese government to impose an import ban for all fresh and frozen beef from North America which deeply affected U.S. market participation since, prior to the ban, American beef

PAGE 47

47 accounted for one third of the beef consumption in Japan. Consequently, domestic production and imports from different sources, in particular Australia and New Zealand, have increased since then, but these sources of beef have been incapable of meeting the Japanese demand as show in Figure 2 5, resulting in a tight supply sit uation. Today, according to estimates from the USDA FAS, Japanese beef imports are expected to increase more than 10% between 2007 and 2008. The U.S. presence in the Japanese beef market has been almost insignificant compar ed to pre BSE levels due to trade restrictions that constrained American exports to cattle under 21 months old, in addition to high U.S. beef prices and consumer anxiety. Figure 2 7. Average CIF prices for fresh beef products in Japan from 1999 to 2006. Source: USDA FAS. February 2008. Less expensive Australian beef has been a major factor affecting U.S. market participation in the Ja panese market. As presented in Figure 2 7, American beef has been commercialized at higher prices in relation to Australian beef between 19% and 45% high er depending on the

PAGE 48

48 year considered. As in the Korean market, Japanese consumers have shown a strong preference towards American beef, which is translated into higher price levels. Following the same price trend, beef products from New Zealand have been priced above the average of U.S. and international price levels showing an important increase, in particular, after the December 2003 BSE scare. This circumstance has positively affected the demand for beef from grass feed cattle since cattle produced unde r this condition are less likely affected by BSE. Republic of Taiwan and Hong Kong Figure 2 8. Taiwan and Honk Kong beef market indicators. Source: USDA FAS. November 2007. Figure 2 8 includes data from the Republic of Taiwan and Hong Kong given that both countries reflect marked similarities in production, imports, and consumption trends. First, both

PAGE 49

49 countries do not have a significant beef livestock industry and depend heavily on beef imports to satisfy their domestic demand. The primary meat consume d in Taiwan is pork, where its consumption surpasses beef consumption by 17 fold. After a substantial decline in 1996 due to the BSE scare, beef consumption and imports rebounded significantly in 1997 showing an increase of more than 22%, as many consumer s substituted beef for pork following the March 1997 outbreak of foot and mouth disease (FMD) that devasta ted Taiwan's pig industry (FAS 2007). Beef imports increased again in 1999 by almost 11.5 % to 96,000 tons to match increasing levels of domestic cons umption. After 1999 and 2001, the substantial downward and upward trends on imported beef levels were the result of long downturn and the result of a growing economy that positively affecte d income levels leading to greater demand for imported beef products. During the year 2004, Figure 2 8 shows a sharp decrease in beef consumption 18.7 % mainly due to BSE concerns that affected the demand for imported beef products, in particular for t hose originating in the United States. In recent years, import and consumption levels returned to pre BSE levels, but the negative consequences on U.S. market shares are unquestionable since Australia and New Zealand are now the main suppliers of beef prod ucts In relation to Hong Kong, this special administrative region relies almost entirely on 109,000 animals during t he 1995 to 2007 period (FAOSTAT 2007). Accord ing to a report prepared by USDA FAS in 2004, Hong Kong is primarily a frozen beef market. T his preference is the result of

PAGE 50

50 have the time to purchase fresh meat daily, and the e xpansion of the food service industry (USMEF 2007). Figure 2 particularly stable even though beef consumption and imports were considerably down since early 1996 due ns over food safety issues. In March 1997, an outbreak of E. coli in Hong Kong's beef supply further discouraged consumers from buying beef though the E. coli bacteria were detected in ground beef originating from locally produced cattle. Demand fell almos t 14 % for all types of beef product s regardless of their origin ( FAS 2004). After these food safety in terms of imports and domestic consumption with 49 % and 42.1 % respectiv ely. The following F igures 2 9 and 2 10 show relative beef prices for Taiwan and Hong Kong. Of particular significance are the U.S. price levels, since they present a significant difference between pre and post BSE outbreak. American beef was considered a product of superior quality compared to competing products due to the fact that consumers were willing to pay a premium price. In fact, U.S. beef prices in Taiwan and Hong Kong were 15 % and 45 % more than the world average. Due to the 2003 BSE announce ment, beef prices increased rapidly as a consequence of the interruption in beef supply forcing traders to shift towards less expensive substitutes for U.S. beef the Australian and New Zealand grass feed beef to satisfy the increasing demand for import ed beef. In the case of Hong Kong, the market for U.S. beef responded quite differently after 2003 showing its highest prices during the years that followed the BSE announcements.

PAGE 51

51 Figure 2 9. Average CIF prices for fresh beef products in Taiwan from 199 9 to 2006. S ource: USDA FAS. February 2008. Figure 2 10. Average CIF prices f or fresh beef products in Hong Kong from 1999 to 2006. Source: USDA FAS. February 2008.

PAGE 52

52 International Beef Supply The present research is based on a trade model specification t hat follows the Armington model of product differentiation by country of origin using four exporting regions to formulate a beef trade model that compares beef products. Traditionally, the three major suppliers of beef products in the Pacific Rim region ha ve been the United States, Australia, and New Zealand. All other beef suppliers were aggregated and included in the model as part of the Rest of the World category with countries such as Brazil, Canada, Mexico, Argentina, and the EU community included in t his last group. In the following sections each country or region is discussed showing production capacity, consumption, imports (M) and exports (X) measured in metric tons (MT). United States As of December 2007, the United States is considered the third l argest exporter of beef products behind Brazil and Australia, and at the same time, it is also considered the largest importer of beef, specifically ground beef imports from Australia and New Zealand. According to official trade data on import and export v olumes, the U.S. is clearly a net beef i mporter. The ratios of imports to exports, during the last six years, support this statement in Table 2 3 Table 2 3. U.S. beef imports and exports volumes between years 2002 and 2007 (1000 MT) Source: USDA ERS. February 2008. Since 2 002, the United States has imported, on average, 3.41 times more than it has exported showing peak import levels right after the BSE outbreak. This represents the logical consequence of the ban on U.S. beef products overseas and the increasing domestic dem and in Year Imports (M) Expo rts (X) Ratio (M/X) %Production 2002 1459.66 1109.94 1.31 9.0 2003 1363.50 1142.15 1.19 9.6 2004 1668.77 208.65 7.99 1.9 2005 1632.48 316.15 5.16 2.8 2006 1399.33 518.91 2.69 4.4 2007 1384.36 649.09 2.13 5.9

PAGE 53

53 the United States during past years. The import bans caused U.S. beef exports to drop, and although some important markets re opened during 2004, export quantities for that year declined al beef exports or 2,026.88 million metric tons were commercialized in Japan, Mexico, South Korea, Canada, and Hong Kong markets during the years before the December 2003 BSE outbreak, but after that only Mexico and Canada remained among the major markets for U.S. beef, accounting for more than 65 % of total beef export volumes ( ERS 2007). A report prepared for the Kansas Department of Agriculture showed that in the very short run, cattle prices fell by about 16 % after the BSE announcement and that some in dicators predicted that U.S. domestic consumption could fall by 15 %. Nevertheless, in early 2004 U.S. domestic prices recovered showing that U.S. consumer demand and confidence had been minimally impacted (Coffey et al. 2005). Figure 2 10 illustrates the U.S. beef scenario in terms of total production, consumption, and exports of fresh/chilled and frozen beef products during the 1995 to 2007 period. United States domestic consumption increased by 8.5% over the period 1995 to 2007, while production slightl y decreased after 2004, by 3.3%, in part as a consequence of the loss of high quality beef markets overseas (USMEF 2007). Historically, the United States has exported about 10 % of its total production of fresh/chilled and frozen beef. World statistics on beef trade show that the United States had been considered the top beef exporter until the 2003 BSE scare, when Brazil and, in particular, Australia, took the lead as world leaders. Today in most of the previously considered U.S. markets, both Australia a nd New Zealand have taken an important lead while the United States is struggling to re gain market access. Current export volumes represent only one half of the 2003 monthly volumes due to trade restrictions imposed on U.S. beef by most of the major beef trade

PAGE 54

54 partners (i.e., restricted to beef from cattle under 21 months of age in Japan, under 30 months of age in Taiwan and Hong Kong, and a complete ban in Korea). Australia and New Zealand have benefited from the absence of American beef in Japan and Kore a, with 2006 exports increasing 17% over the 2003 volume (USMEF 2007). United States exports to traditional open markets such as Mexico, Canada, and Russia have increased by 33% in value and 17% in volume since 2003. The importance of the United States as an exporter has decreased since 2004, when only 1.86% of its production was sent overseas. Since then, that level has increased to 5.43% in 2007 with most of the markets located in North America and Eastern Europe. Figure 2 11. U.S. production, consumpt ion, and e xports of fresh/chilled and frozen b eef 1995 to 2007. Source: USDA FAS. February 2008. The year 2007 marked the fourth year of BSE related trade restrictions for U.S. beef products. However, in 2006 the World O rganization for Animal H ealth (OIE) declared the U.S.

PAGE 55

55 for U.S. beef in all countries. The OIE certification and the weak U.S. dollar have made U.S. beef exports more competitive overseas, giving American beef products a competitive advantage in relation to products from other beef exporting nations. As a result U.S. beef exports are forecast to climb over 19 % in 2008 to around 775,000 metric tons due to continued strong opportunities in the NAFT A countries (FAS 2007 ) Today, U.S beef maintains a strong presence in Canada and Mexico whi le slowly making inroads in Asian markets despite being impeded by a 20 month or younger age restriction in Japan and the current suspension of exports to Korea Australia rozen beef products have increased at a steady rate during the past thirteen years. Figure 2 12. Australia production, consumption, and e xports of fresh/c hi lled and f r ozen beef from 1995 to 2007. Source: USDA FAS. February 2008.

PAGE 56

56 The data set shows that Australian beef exports increased 23.4% and that domestic production and consumption increased by 15.7 and 24.8 % respectively over this period. These numbers reflect tha t Australia is a net beef exporter and that most of its production demand is located outside the country. Today Australia is the second largest exporter of beef products in the world, exporting almost 65% of its total production. Like the United States, Au stralia had focused its attention on markets around the Pacific Rim where it has a clear advantage over the Rest of the World beef exporting nations due to its proximity to most of the major markets in the region. S ince December 2003 Australia n beef has captured two thirds of the import market shares in the region as it filled the large market left by the U.S beef absence. T he current situation of ab sence or limited presence of U.S. beef in major Asian markets continues to present o pportunities for Austr alian products limited supplies due to a prolonged drought, limited range of cuts, and the increasing pressure from competitors (i.e., New Zealand) that have negatively exports are expected to decline by 3% during the next year as the United States regains markets in Asia and as Australia production declines (USMEF 2007). New Zealand Although a small producer of beef if compared to the United States and Australia, this nation has a significant importance as a world beef exporter due to its geographical location and its relatively large exports levels in relation to domestic consumption and production. New Zealand is a net beef export country, with around 79.3% of its pr oduction being exported overseas on average over the past 13 years and showing its highest peaks right after the 2003 BSE scare in North America with more than 81% of its production being exported. The average beef production during the period 1995 to 2007 was about 641,000 metric tons, showing a significant increase of more than 16% in production levels after 2002, while domestic

PAGE 57

57 consumption has remained stable within the same period representing 20.5% of its total production or 149,600 metric tons on aver age between 1995 and 2007. Figure 2 13. New Zealand production, consumption, and e xports of fresh/c hi lled and f rozen beef from 1995 to 2007. Source: USDA FAS. February 2008. As illustrated by Figure 2 13, during the year 2004 beef export results reflect ed soaring quantity levels to K orea and Japan as traders took advanta ge of the ban on U.S. beef. Consequently, expor ts to Japan and Korea increased 92 % and 99% respectively from January through August 2004 compared to the same period in 2003. D espite incre asing beef exports, New Zealand has orea's beef deficit due to its general inability to provide specific cuts and the quality of beef that have traditionally been supplied by the United States and Canada ( FAS 200 4).

PAGE 58

58 Rest of the World Brazil, Canada, Argentina, and other beef producing countries around the w orld have shown a solid growth i n their exports levels, in particular after the year 2003 with an increase of almost 28%. Economic growth in Asia, Eastern Euro pe, and nations from other parts of the developing world has boosted the demand for beef products which has also triggered global production. International trade has benefited from this economic bonanza as F igure 2 13 clearly illustrates a steady and posit ive trend of those indicators during the past 13 years. Thus, international beef trade, world consumption, and production levels have shown growth levels of more than 35%, 9.6%, and 12.6% respectively during the 1995 to 2007 period. Figure 2 14 Rest of the world production, consumption, and e xports of fresh/c hi lled and f rozen beef from 1995 to 2007. Source: USDA FAS. February 2008.

PAGE 59

59 Bovine Spongiform Encephalopathy (BSE) : Impact on Beef Trade nnouncement on major beef markets, the consequences for American beef exports and how this food safety issue affected patterns of trade. To facilitate the interpretation, each importing market is analyzed from the consumption and market share perspectives while country specific expenditures per capita on beef are used to understand BSE consequences in exporting nations. Figures 2 15 through 2 18 illustrate pre and post BSE per capita consumption of imported beef, as well as market share information for Japan, Korea, Taiwan, and Hong Kong. Figures 2 19 through 2 22 introduce the amount of U.S. dollars expended on imported fresh/chilled and frozen beef products in each of the four importing market with respect to a specif ic exporting country. Finally, F igu res 2 23 and 2 24 present a pre and post BSE trade volume matrix that represents trade patterns between each of the eight areas considered for this research. Consumption Levels and Market Shares According to the USDA FAS dataset used in this study, the R epublic of South Korea was considered the third largest market for U.S. beef products before the BSE announcement and accounted for roughly 19% of all U.S. beef exports on a volume basis and 22% on a value basis. In numbers, Korea imported more than $ 816 million worth of U.S. beef products during 2003, which represented a consumption level of 17.42 Kg (38.3 lb) per capita. Zealand, and the rest of the world (ROW) with 2 1%, 8%, and 2% of the imp ort market share respectively (F igure 2 the United States and its competitors were concentrated in the following categories according to their total volume: fr ozen boneless and bone in beef, fresh/chilled beef, and frozen edible offal.

PAGE 60

60 Figure 2 15. Korea imported beef consumption per capita and market shares. Source: USDA FAS. February 2008. After the December 2003 BSE scare in North America, U.S. beef export s to Korea were reduced to a little more than one million dollars during the year 2004. Consumption levels were also impacted as a result of this food safety issue, as presented in F igure 2 15, showing a decrease of more than 44% to 9.7 Kg per capita durin g 2007. Korea has the highest average price for beef products in the world, and the continued absence of U.S. beef has kept the domestic market undersupplied and prices high, resulting in a reduced overall per capita consumption. As expected, beef market shares have also changed as Australia and New Zealand captured the market portion lost by the United States. Hence, in 2007 Australia accounted for 71.7% of the import market followed by New Zealand, the U.S., and ROW with 22.2%, 4 %, and 2% respectively. Although the ban on U.S. beef was removed in September 2006, in practice,

PAGE 61

61 products are not yet moving into the market as Korea has been very strict in its enforcement of its boneless import requirements (USMEF 2007). Figure 2 16 summarizes pre and post B SE beef consumption and markets shares for the Japanese market. Prior to the BSE incident, Japan was the most important market for key U.S. beef products. In particular Japan represented the main market for products such as guts, tongues, intestines, and e dible offal, as well as the traditional fresh beef, and frozen boneless and bone in beef (USMEF 2007). Using the data available from the USDA FAS, the differences in volumes of U.S. beef exported to Japan between June 2003 and June 2005 indicate that the l oss of this particular market accounted for more than $967.3 million representing 46.8% in market share loss. Before December 2003, Australia accounted for 42% of the market, followed by the U.S., ROW, and New Zealand with 40%, 15%, and 3% of the import ma rket share respectively. Figure 2 16. Japan imported beef consumption per capita and market shares. Source: USDA FAS. February 2008

PAGE 62

62 The loss of confidence in the safety of beef products along the entire marketing chain and elevated levels of general co ncern for food safety have deeply damaged the position of American beef products in the Japanese market. U.S. beef exports to Japan have been limited to less than 8% of the 2003 monthly shipments since the marke t re opened on July, 2006 ( ERS 2007). In 2007 the United States accounted only for 6% of the market and Australia, the big winner, accounted for 82% of the import mark et while New Zealand and the ROW accounted for 8% and 4% respectively. Figure 2 16 also illustrates the changes in consumption within the Japanese market. Shortage of beef supply, unfavorable media reports, together with an unsatisfied Japanese preference for feedlot type beef produced in the U.S. accounted for the 18.5% overall decline in Japanese beef consumption during the period bet ween 2003 and 2007. Taiwan is mainly a pork meat producers. Therefore, it dependence on imported beef as major source of red meat is relatively small if compared to other Asian nations. This characteristic explains the reduced impact of the products. As shown in F igure 2 17 consumption levels dropped a little more than one kilogram per capita between 2003 and 2007. Taiwa n opened its market for U.S. beef products in February 2006, and by the end of that year total exports of U.S. beef to Taiwan increased 176% in volume and 142% in value over the previous year (USMEF 2007), which clearly shows the acceptance of American bee f among beef consumers in this country. Before the BSE scare, market conditions were relatively stable and Australia accounted for more than 37.4% of the beef market in Taiwan, followed by New Zealand, the United States, and the ROW with 31.3%, 25.3%, and 6.1% of the import market share respectively. At the end of 2007, the United States maintained a

PAGE 63

63 20% share of the Taiwan import market for fresh and frozen boneless beef, though bone in cuts and beef variety meat are still banned. As illustrated in Figure 2 17, Australia remains the largest player in the Taiwanese beef market followed by New Zealand with 26% and the ROW with 13%. Figure 2 17. Taiwan imported beef consumption per capita and market shares. Source: USDA FAS. February 2008 On the other hand, the situation in Hong Kong is quite different if compared to the rest of administrative region that depends almost entirely on China to satisfy it meat require ments. well above the rest of the countries in the region. In 2003, the United States was considered with 50% of the import market share followed by New Zealand and Australia with 12% and 5% of share respectively.

PAGE 64

64 Figure 2 18. Hong Kong imported beef consumption per capita and market shares. Source: USDA FAS. February 2008 economic indicators (per capita income growth and currency appreciation) contributed to a steady increase in the demand for imported beef products during the following years. In late 2005, Hong Kong resumed imports from t he U.S. of boneless beef products from cattle under 30 months of age only, while main China declared itself open for U.S. beef in 2006. In that year, Hong Kong imported 20 times more than the rest of China, which corresponded to 80,000 metric tons versus 4 ,800 metric tons respectively ( FAS 2007). This difference was also translated in terms of consumption levels of imported beef where China consumed only 5.4 kg per capita of imported beef, while Hong Kong consumed 20.6 Kg during 2007. In terms of market par ticipation, the ROW group has been the largest winner from the U.S. export ban absorbing 81% of the share

PAGE 65

65 from early 2004 through 2007. Th U.S market participation decreased to 7% while Australia and New Zealand kept a modest 6% of the market respectively Expenditures on Beef by Country of Origin In this section, expenditure levels on imported beef are used to analyze the position and relevance of each exporting nation. Figures 2 19 through 2 22 show related statistics in terms of U.S. dollars per capita for both the pre and post BSE announcements. Figure 2 19. Total expenditures on beef p roducts from the U.S. by country. Source: USDA FAS. February 2008. In the case of expen ditures on U.S. beef products, F igure 2 19 clearly illustrates that Japanese a nd Korean consumers have a manifest preference for grain feed U.S. beef products. Before the BSE announcements, these two countries were considered among the top three destinations for U.S. beef. During the year 2000, both Japan and Korea increased their e xpenditures on American beef by almost 14% and 43% due to the steady economic growth in

PAGE 66

66 Japan and trade barrier reductions in Korea. Although the 2001 BSE crisis in Japan affected domestic rather than imported beef, food safety concerns among Japanese cons umers had a dramatic impact on imports and expenditures on U.S. and Canadian as well. In Korea, changes in expenditures levels on U.S. beef products were related to the 2000 economic crisis that affected the country and erroneous government policies rather than related to food safety issues. The failure of these policies and the rebound of the economy helped U.S. exports and by the end of 2003 expenditures on U.S. beef were a t their historic maximum. After 2003 BSE announcements, the top U.S. beef markets w ere Taiwan and in some degree Hong Kong. As, of December 2007, these two countries plus the little trade with Japan represent expenditure levels of only $ 9.05 per capita down from the $ 36.90 level in 2003. Figures 2 20 and 2 21 illustrate a similar trend on expenditure levels on fresh/chilled and frozen beef from Australia and New Zealand, in particular during and after the year 2003. In 2004 expenditure levels on beef from these two countries increased more than 31% and in 2005 they increased once again by 27.6% and 11.4% respectively. However, during the next two years expenditures on New Zealand beef decreased first 5% and then 23.3%, while expenditures on Australian beef increased 5% in 2006 but decreased 15.8% in 2007. These f igures also illustrate th at Australia and New Zealand have different beef tr ade partners. In the case of Australia, Japan has been the major consumer per capita of its beef products. Before 2004, Taiwan was its second trade partner, but after the BSE scare Korea captured the statu partner. For New Zealand, Taiwan has been its major trade partner for the last ten years, as Kore a and Hong Kong interchanged second place until December 2003. Since then, Korea has beco me the second largest market for New Zea land beef due to the complete ban on beef imports from the United States, which forced Korean traders to locate alternative sources of beef.

PAGE 67

67 Figure 2 20 Total expenditures on beef products from Australia by country. Source: USDA FAS. February 2008. F igure 2 21. Total expenditures on beef p roducts from New Zealand by country. Source: USDA FAS. February 2008.

PAGE 68

68 Finally, F igure 2 22 shows expenditure levels of each importing nation on beef products from countries clustered in the Rest of the World (ROW) ca tegory. Until 1997 Hong Kong was part of the Commonwealth of England and all of it imports were made through the British authority. Since 1998, Hong Kong became part of China and most of it imports started to arrive from different places in the world, in p articular from Latin American countries, which after the 2003 BSE scare through 2007 remained the major beef supplier of Hong Kong in terms of expenditures per capita. The ROW region has been an important source of beef products for Japan during the years previous to the BSE announcement, but since then this region has lost importance for Japanese traders due mainly to the lack of enforcement of international food safety regulations and OIE standards. Figure 2 22. Total expenditures on beef p roducts from the ROW by country. Source: USDA FAS. February 2008.

PAGE 69

69 Pre and Post BSE Pattern of Trade Figures 2 23 and 2 24 present trade quantities by country and by trade partner nation for the years 2003 and 2007, which represent pre ach figure represents a coordinate system indicating three axes that represent trade quantities by importing and exporting country for the pre and post BSE announcements periods. Thus, the planes X, Y, and Z represent the demand, supply, and the quantities traded. Each market is identified with a particular co lor. As an example, looking at F igure 2 23, and the U.S. and Korea as beef trade partners, during the pre BSE year 2003 Korea imported 224,000 metric tons of fresh/chilled and frozen beef products from fresh/chilled and frozen beef imports and 46% of all U.S. exports to the region during that year. Following the same criteria to analyze trade relati ons among participant nations, F igures 2 23 and 2 24 are contrasted and discussed in the following pages. Recall that before the BSE scare the total supply of beef products in this category was about 1,031 ,000 metric tons and the United States was by far the l argest exporter to the region. F igure 2 23 reveals that more than 47% of all beef imported in this region originated from the United States and that Australia had a little more than 33% followed by the ROW and New Zealand with 13% and 7% respectively. In terms of value and volume traded, Japan was the main market for U.S. products but Australia was the leading beef supplier of this country during that of its total exports during that year, while exports to Japan represented 26%. In the case of Taiwan and Hong Kong, the figures show that they obtained a large percentage of their beef from the same set of suppliers, with Australia being the largest suppl ier to Taiwan and the ROW region in Hong Kong.

PAGE 70

70 Figure 2 23. Pre BSE ban trade scenarios for fresh/chilled and frozen beef in selected Asian nations. Source: USDA FAS. February 2008. During the following years, the BSE cris is in North America changed the way that beef products were traded not only in Asia but all around the world. Consumer concerns on the safety of the beef supply chain had a significant impact on the demand for grain feed beef products regardless of their o rigin. As a consequence, previous patterns of beef trade changed dramatically. Governments of almost all beef importing nations took drastic positions towards imported beef from North America, imposing a complete ban on a ll type s of exports from the United States and Canada, which have a direct impact on the total volume of beef available for consumption in the region. Comparing Figures 2 23 and 2 24 the negative shift on beef supply can be quickly observed. Total fresh/chilled and frozen beef supply was r educed from 1,031 million of metric

PAGE 71

71 tons (M.MT) in 2003 to 754 M.MT in 2007; the U.S. market shares decreased 40%, while Australian exports increased almost 38% during the same years. Figure 2 24. Post BSE ban trade scenari os for fresh/chilled and frozen beef in selected Asian nations. Source: USDA FAS. February 2008. Figure 2 24 illustrates that Australia, New Zealand, and other grass feed beef producing nations seized this circumstance to expand their market positions in t he region. Thus, in 2007 for fresh/chilled and frozen beef. Australia has largely profited from the absence of U.S. beef in Japan and Korea, as 2007 exports inc reased 18% over the 2003 period and as Australia has captured almost 94% of all beef exported by these countries. After the BSE ban, New Zealand became an important player in the region increasing its export levels from 69,000 metric tons to 98, 000 metric tons, which represented an increase of 29.5 % with respect to 2003 levels and

PAGE 72

72 having Japan and Korea as its main export destinations. Imports from the ROW region decreased 46% in all countries but in Hong Kong where traders concentrated their imports on a lternative sources, su ch as South America, resulting i n a significant 50% increase during the year 2007. Finally, although beef products from North America were totally ban ned shortly after the December 2003 BSE announcement, the United States managed to m aintain a small market participation, especially after 2005. Today U.S. beef exports are benefited by the strong economic growth of the Asian region as rising incomes lead to greater demand of high quality beef (USMEF 2007). For example, record U.S. beef exports of fresh/chilled and frozen beef to Taiwan during 2007 have surpassed 2003 volumes by 26%, for a total of 19,000 metric tons. In contrast, Japan, Korea, and Hong Kong today only represent 8% of the 2003 United States total annual volume of exports; while before the BSE scare they represented 97% of total U.S. beef exports. This significant difference is the result of trade restrictions that allowed only boneless beef and no variety meats under 21 months of age in Japan and under 30 months of age in the rest of the markets (USMEF 2007). As for fresh/chilled and frozen beef between December 2003 and the end of 2007, which represents a loss in value and qu antity traded of $1.627 billion and 431,000 metric tons respectively. According to USMEF and USDA FAS estimations, U.S. beef exports globally are expected to continue to increase, especially in Japan, Kore a, China, and Russia fueled by continued economi c growth, increasing incomes, and, more important, the OIE announcement that declares the United States access for U.S. beef in all countries.

PAGE 73

73 CHAPTER 3 LITERATURE REVIEW Introduction This chapter presents a review of previous research on areas related to agricultural trade, food safety issues and policies, commodity advertizing and promotion programs, and the Armington model and its applications. Since this study focuses on international t rade and marketing the discussion in this chapter will provide a foundation for evaluating some of the issues affecting market access efforts and import demand for beef products from the United States Australia, New Zealand, and the Rest of the World in markets such as South Korea, Japan, Taiwan, and Hong Kong International Beef Demand and Food Safety International beef consumption has increased considerably over the last three decades due to trade liberalization and more effective food safety policies. Since food is a necessity, consumers value the fact that their food is free of toxins, foreign material, and pathogens. Food safety concerns have increased as wealth has risen. Now that many consumers in the industrialized world have adequate quantities o f food, they (or their governments) can spend resources to ensure that their food is safer ( Jin et al. 2004). As of December 2007, the world has already been impacted by several food b orne disease outbreaks where BSE, Avian Influenza (AI), Salmonella, an d E. coli are among the most notorious and controversial. In the United States, the BSE disease remains a serious concern to the cattle and beef industry as traditional U.S. beef importers such as Korea and Japan have not yet fully opened their markets to American beef products. Coffey et al. (2005) argue that in the event of a widespread BSE occurrence there will be a large decline in domestic and foreign

PAGE 74

74 international beef markets. According to this report, the economic significance of animal disease outbreaks is also influenced by the degree of consumer response: fears that the disease can spread to humans can lead to sharp drops in consumption as was reported in Paci fic Rim nations. Similarly, Peterson and Chen (2005) e xplored the impact of BSE on the Japanese retail meat demand using a Rotterdam model that allowed beef product differentiation and found that effectively, the scare affected the demand for domestic beef (wagyu beef), domestic dairy beef, and imported beef from the United States and Australia. According to this research, Japanese retail meat demand underwent a two month transition period following the initial announcement and continued to adjust subsequen tly until it reached a new state within five months of the BSE announcement. Devadoss et al. (2006) employed a general equilibrium model to analyze the economic impact of the BSE outbreak on the U.S. cattle and beef related industries using different deman d scenarios in the domestic and international market. This study found that only if domestic demand declines significantly will the economic hardship in the U.S. industry be very large. Using a trade model that allows for imperfect substitution between goo ds, three different scenarios were analyzed in which the decline in foreign demand was in all cases 90 % and the decline in the U.S. market was 0%, 10%, and 25% respectively. Results showed that the impact of the BSE outbreak even in the worst case scenari o was not as damaging as it was in Canada, whose industry depends largely on the foreign market and exports prices dropped significantly. In the case of the United States since it only exports 10 % of its beef, reduction in foreign demand did not have a l ong term effect on the domestic industry in general, although cattle futures and beef cash prices declined 19% in the days following the BSE announcements.

PAGE 75

75 In 2007, Mutondo and Henneberry addressed the competiveness of U.S. meat products in Japan and Sout h Korea using a Restricted Source Differentiated Almost Ideal demand system (RSDAIDS). Economic (e.g., prices and expenditures) and non economic factors, (e.g., seasonality and outbreaks) estimate coe fficients were used to evaluate the demand for meat, and they showed that in the case the Japan the impact of the BSE outbreak had a larger and negative impact in terms of total beef consumption a nd the imports of U.S. product market shares, while in South Korea the impact only affected American beef products w hile total consumption remained at the same level. In the United States, the effects of the BSE events on feed and feeder cattle prices were analyzed by Marsh et al. (2008) using a structural econometric model that implies a system of simultaneous equatio ns to estimate relevant supply and demand elasticities at the feedlot and cow cal f levels of the marketing chain; and based on annual time series data from the USDA between 1970 and 2005. They found a strong correlation between trade disruptive policies (e .g., Minimum Risk Regional Rule MRR), market share sizes, and feed and feeder cattle prices. The results indicated that the demand for U.S. beef was affected to a much greater degree by the reactions of foreign governments to the BSE announcements than b y the reactions of U.S. households. F ood safety incidents in particular those related to meat products, have captured the attention of the media which have led to permanent changes in consumer perceptions about food safety and their food purchasing patte rns. In the case s of Europe and Asia, where the public outcry has been particularly strong, there have been changes in government regulations that have directly affected the demand for domestic and/or imported food products. For example, several studies ha the

PAGE 76

76 United Kingdom, Burton and Young (1996) investigated the impact of BSE on the demand for beef and other meats using an AIDS model which included indices of media coverage of BSE They found that the influence of negative press had significant effects on the allocation of consumer expenditure s among the meats. A short run impact was identified that accounted for a large portion of the perceptible drop in the market share of beef i n the early 1990s. Results also showed that t here also appears to be a significant long run impact of B SE, which by the end of 2003 had reduced the beef market share by 4.5%. Kinnucan et al. (1997) investigated the impact of health information and generic advertizing on U.S. meat demand using a Rotterdam specification and concluded that this demand is affected by prices, expenditures, and health information, but the effect of generic advertizing is less clear and not robust. Health information elasticities in general were larger in absolute value than price elasticities, which suggest that small percentage changes in health information have larger impacts on meat consumption than equivalently small percentage changes in relative prices. In 1998, Latouche, Rainelli, and Vermersch conducted a survey using a contingent valuation method to analyze consumer behavior in France after the BSE outbreak in the United Kingdom Their survey revealed that the BSE disease raises the problem of loss of public confidence in addition to the fact that beef consumers expected greater transparency from the industry and that they would accept paying for a system that guarantees the safety of the beef products consumed (w illingness to pay). Similarly, Verbeke and Ward (2001) in vestigated fresh meat consumption in Belgium during the period from 1995 through 1998 using an AIDS model which incorporated a media index to measure the BSE impact. They found that health safety scares have a strong and

PAGE 77

77 negative impact on beef demand, mai nly because of the negative publicity from the press and in particular the TV coverage of the issue. They results also showed that as consequence of the negative media, pork consumption increased significantly. Adda (2002) investigated the effects of past consumption of risky goods on current consumption patterns in France, using the BSE crisis as a natural experiment. He found that new health information interacts with prior exposure to risks. Consumers with intermediate levels of past consumption decreas ed their demand for beef and sought higher quality products, while low and high stock consumers did not alter their behavior after the crisis. In a number of experiments and surveys, consumers have indicated that they would be willing to pay more for food with lower risks of disease Hence, previous studies have found that U.S. consumers were willing to pay a premium of 15 % to 30 % per meal to reduce their risk of becoming ill from their meal and that consumers are willing to pay a premium for reduced pesti cide residues in produce (Buzby 2003). In a previous study, Jin and Koo (2003) suggested a three way approach to the BSE effect consumers or producers welfare and evaluating the economic consequences of the outbreaks. A non parametric approach was used by them to show that there has been an ongoing structural systematic ally moved away from beef to its substitutes. Ishida et al. (2006) investigated the impacts of the BSE and Bird Flu on the demand for meat products in Japan using the Almost Ideal demand system. A s expected, they found evidence that the appearance of BSE a nd Bird Flu cases in Japan deeply affected the structu ral demand for beef and chicken and have a positive impact on the demand for their closest substitutes (e.g., pork and fish products). This research

PAGE 78

78 showed that in the Japanese market, Bird Flu concerns although BSE fears increased consumer demand for chicken products. Empirical results also showed that both impacts do not persist over time and that in the short run the impact dissipates depending on the characte ristics of the disease (e.g., incubation period, cure rate, and infection risk) and the ability of the government to efficiently respond to an outbreak. Trade and Beef Markets Issues Global food trade will likely increase due to expected increases in inc ome levels around the world improved transportation networks, and growing populations requiring greater and safer quantities of food In addition, the creation of the General Agreement on Tariff and Trade (GATT) has integrated global markets into an overa ll framework that affects not only the volume of goods traded but also the diversity of products flowing across nations. As a result, this ongoing process of trade liberalization has raised consumers concerns about the safety of food products (Bureau et a l. 2002). The USDA ERS published a report (Regmi 2001) that studied structural changes in international trade and consumption as a result of food safety incidents. It argued that following the resolution of the problem that caused a major international inc ident, consumer perceptions may be slow to change, and these perceptions have a lasting influence on food demand and global trade. Consumers value a safe food supply and recently food safety scares like the BSE or problem or the E. coli outbreak have raised awareness about food safety issues. Food safety regulations and the perception of risk are different among countries. Even if the food safety risks are t he same across countries, countries may perceive and handle these risks differently which can lead to persistent trade frictions and even reduce food trade (Buzby 2003) Although little disruption to trade has occurred for food safety reasons (considering the total

PAGE 79

79 volume of food trade), trade issues related to food safety are wide ranging. These issues and crises challenge policymakers and industries to both protect domestic food supplies and nurture international markets. Meanwhile, consumers in develope d countries are demanding safer food which lately has become an important issue in public policy both domestically and internationally. The safety of food is a credence characteristic which gives rise to a particular form of market failure involving eithe r asymmetric information and moral hazard or symmetric imperfect information (McLaren 2006) Risk reduction measures and quality certification programs can not only anticipate food safety crises, but can better position exporter s in emerging overseas marke ts ( ERS 2007). Beef trade represents a complex pattern of business and regulations that have dramatically changed during the past years as a consequence of raising trade barriers aimed to increase food safety and protect domestic production. As result of t hese circumstances, trade between the United States and its major trading partners is deeply distorted by a wide variety of foreign measures. These distortions include a growing disparity in the respective regulatory regimes between trade partners, particu larly with respect to bovine spongiform encephalopathy (BSE), and a diverse array of tariff and subsidy policies that influence global beef trade flows. In the past, different studies have approached the circumstances affecting the flow of products b etween regions and countries. Economic theory has explained trade based on comparative and absolute advantage assumptions, resource allocation and factor endowments, and government policies that affect the degree to which trade is conducted among nations. Basically, international trade allows countries to specialize in the production of those goods (commodities or products) in which they have a comparative advantage, thus increasing domestic welfare in terms of production and capital returns (Bhuiyan 1993). At the same time,

PAGE 80

80 industries and governments are constantly seeking and en forcing the policies necessary to participate in the international market while protecting local markets from foreign competition. Many studies from different countries have focuse d their attention on beef trade and the economic impact of the policies designed to address food safety issues. For example, in 2003 Poulin and Boame from the Canadian Business and Trade Statistics Department focused their efforts to measure the economic i mpact of the May 2003 BSE announcement on cattle production, domestic prices and beef exports to United States and Mexico, since Canada exported more than 92 % of its production to these countries. As a consequence of this outbreak, beef exports fell virt ually to zero during the following months affecting entire supply chain from the breeders to the meat packing plants. The immediate loss in terms of trade represented almost $ 1 billion, with most of this loss concentrated in the fresh beef U.S. market whe re 30% of the beef imported came from Canada. Other impacts were found at the domestic level where the collapse of exports triggered an increase in beef supplies and a fall in prices that generated a 50% drop in cattle prices, although consumers did not se e a significant drop in the price of beef at the retail price which clearly shows a price asymmetry effect of the embargo. Differences in what food products countries want and what they will accept in imported food ultimately affect patterns of food demand and global trade, and complicate the development of workable trade rules that are acceptable to different trading partners (Buzby, 2003) The effect of trade barriers on consumer demand for traded goods of different quality was approached by Bureau et al. (2005) under the assumption that s pecific tariffs a nd tariff rate quotas (TRQ) have an e ffect in the composition of imported goods due to a variation of the price ratios between products of different quality. The study used a standard Constant Elasticity o f Substitution (CES) function to deal with differentiation, where the parameters represented

PAGE 81

81 preferences for the different qualities in beef product imports data from Mercosur countries t h e study shows that TRQ s have a significant impact on the average quality of exports demonstrating that under specific tariff reduction low quality beef (Brazil) is maximized, while high quality beef (Argentina) tr ade is maximized with large TRQ s. The research concludes that the use of tariff dispersion policies impose a proportionally higher protection on low unit value products, and have a quality upgrading effect on imports. A cut in these tariffs may therefore change the composition of imports towards lower unit value product s. Coffey et al. (2005) analyzed the BSE incident on the wholesale sector of the U.S. beef industry. Specifically, the research estimated economic losses to the U.S. beef industry caused by increased regulatory costs and reduced exports to Japan and South Korea. These losses were estimated using excess demand and excess supply relationships coupled with price elasticity estimates previously reported in other studies. Changes in the excess demand and supply functions caused by the BSE cases were used to cal culate changes in producer and consumer surplus. International Beef Promotions Agricultural commodities are considered less differentiated goods and are generally advertised through some type of cooperative effort that includes several sectors of the supp ly chain such as producers, processors, and exporters (Forker and Ward 1993). Under this cooperative scenario, adverti s ing and promotion of commodit ies is known as generic advertis ing and it was defined by Forker and Ward as the promotion of a nearly homog eneous product to disseminate information about its underlying characteristics to existing and potential consumers for the purpose of strengthening demand for the commodity. Competition in the international agricultural market has become much more intense, triggering innovation and transformation inside the industry. As a result of this, a non segmented

PAGE 82

82 sophisticated industry. In this contest, commodity advertising is used to influence market shares and, where necessary, to increase the volume of information about product attributes demanded by consumers (Ward and Lambert 1993). Agents, along the food supply chain, have learned how to differentiate and increase the value of their products (e.g., live animals and meat products) in order to build demand and increase financial profits for the firm. Advertising and promotions, which often highlight production and marketing practices, are examples of the methods that comp anies and individuals have utilized to set their product apart in order to meet the needs and desires of specific consumer segments (Allen and Pierson 1993 ). The fresh beef market is a good e xample of this practice where exporting countries and their agent s have learned to differentiate their beef cuts using particular attributes as central message in their marketing campaigns (e.g., In markets suc e Car consumer confidence in U.S. beef using a range of advertising and consumer events. According to the USMEF, the objective of the campaign was to counter the negative emotions generated by e U.S. industry. The campaign highlighted the concerns of the entire U.S. beef industry in delivering safe and quality beef products for consumers and retailers, and provided updated information on BSE prevention systems being implemented in U.S. plants ( USMEF, 2007). Commodities in general, and in our case beef products, can be differentiated according to search goods consumers have an active participation searching out the product attributes which can be observed at the time of purchase

PAGE 83

83 (e.g., price, color, size). Alternatively, experienced goods are those where the attributes can be observed at the time of consumptio n (e.g., taste, tenderness, fat content) and influence credence goods include attributes that are not explicitly observed in the product (e.g., safety: antibiotic and pesticide residues; heal consumer credibility of the claims made about the product is influenced by a high level of confidence i n the effectiveness of the systems in charge of creating and monitor ing the product (Codron, Sterns and Reardon 2003). Purchase decisions are based on predictions of product performance. Consumers base their predictions in part on product cues and are accurate to the extent that they have properly learned the relationship between the cues and performance. If consumers learn the relationship between product attributes and quality, they will differentiate among products that possess different attributes and treat as commodities those brands that share the same attributes. Onc e the predictive rule is learned, it may be applied to any existing or new product that possesses the same set of attributes (Van Osselaer and Alba 2000). Advertising and promotion provide this information facilitating purchasing decisions, and in some ca ses they even change the underlying preference function for a particular product or service. Depending on the characteristics of the product, potential buyers, and the scope and beneficiaries of the marketing campaign, there are two types of advertising an d promotion of commodities: generic advertising, a cooperative effort among producers designed to collectively increase the primary demand (i.e., the size of the pie) of a product by advertising the attributes and characteristics of the product, without in fluencing the market sha re of any producer (e.g., Beef check off program, Florida citrus, Washington apples); and brand advertising designed and

PAGE 84

84 product from other suppliers in order to increase the market share of the brand within the same industry (Ward 1997). For example, previous market studies conducted by the USMEF (2007) on beef demand in Japan have shown that the most important characteristics of food products were that they taste good and be guaranteed safe to eat, while the most important characteristi cs of beef products were that they look fresh, not have a lot of waste, be certified as USDA inspected, and be free of chemical residues and foodborne h azards. New marketing programs sponsored by the USMEF adding value, and imp roving presentation in order to build a positive perception about branded products. In the case of U.S. beef exports, producers and trade agents have been able to differentiate American products from other sources (e.g., Australia and New Zealand) due to continued efforts Kong. If safety is a concern, experimentation is not an option and consumers look for assurance about the safety. Quality assurance through industr y seals and government inspection may solve some of the problem. Even with safe products, some attributes cannot be judged through experience. These credence attributes must be identified through means other than consumption, and branded beef products may capture shares of a market through a consistent message about one or more credence attributes (Ward and Lambert 1993). For example, brands emphasizing U.S. produced beef may capture some loyalty even with the generally homogeneous nature of the product gro up. Thus, specific beef products within a generally common product category may

PAGE 85

85 growth in product differentiation within the beef group. In 2008, Kinnucan and Myr land disputed the concept that country of origin or 'brand' advertising should be more profitable than generic advertising in that it enhances product differentiation a nd reduces free riding. They argued that unlike generic advertising, brand advertising d ecreases the demand for competing imports and lowers their prices when supplies are upward sloping. In addition to inviting retaliation, the decline in the prices of competing products erodes the price of the advertised product through second round or 'mar ket feedback' effects. Using the U.S. beef promotion program in Japan they showed that generic advertizing is more profitable in most cases and that the gross benefits are distributed across exporters in proportion to the expenditure elasticities. Herrmann Thompson, and Krischik (2002) examined the impact of generic promotion s on Bavarian beef demand during the BSE outbreak in Europe and empirically evaluated the potential off setting effects of advertising and promotion (positive) and heightened food sa fety awareness (negative). Previous studies have treated advertising as a slope shifter that alters the price elasticity of demand For this research, they introduced an information variable as an intercept shifter. This new variable consisted of generic p romotion as well as BSE information that allowed a direct comparison of their respective effects on the demand for "Bavarian" beef. Model estimates showed that advertising expenditures increased (4.6%) consumer demand for Bavarian beef and the benefit cost ratio of the program was well above unity. Despite this success of the program, the negative demand effects of food safety concerns as measured b y BSE information were stronger (6.9%). Hence, generic promotion could only compensate partly the

PAGE 86

86 inward shift in per capita beef demand induced by both BSE information and longer term consumer preference changes away from beef (15.2%) preferences for meat (e.g., beef and pork) and fi sh, announcement. They used a switching AIDS model and a sample period that covered January 1994 through May 1998. Preference shifts due to the BSE scare reduced expenditure shares for beef, minced meat and meat products by 2.5, 3.3 and 7.9 percentage points respectively. There were offsetting gains in t he shares of pork, prepared beef, and fish. They concluded that changing preferences over the whole period reduced beefs share by 4.9 percentage points and increased those o f poultry, prepared meat, and fish by 4.1, 4.9 and 5.2 percentage points respectively. Cong, Kaiser, and Tomek (1998) evaluated the effectiveness of the U S government non price promotion programs on U S exports of red meat to the Pacific Rim: Hong Kong South Korea, Singapore, and Taiwan using pooled time series and cross sectional data. An estimated import demand equation was used for in sample simulations to address the question of what would have happened if the current promotion expenditures had bee n reallocated among the four countries. The calculated rates of return to promotion investment were computed and estimates showed that a 50% increase in expenditures increased returns by $47 on average. Also, they found that South Korea is the only market in which meat promotion appears to be effective. Generic advertising describes promotional activity in markets that are comprised of homogenous, highly substitutable products. Often, such products are not branded, so consumers are not inclined to choose o ne variety over another. However, in the case of the promotion of U.S. beef products, it may imply higher product differentiation and represent an important

PAGE 87

87 transition from the generic idea of selling beef as a commodity to a campaign involving market segm entation and target marketing of specific consumer segments. According to Phillip Kotler (2002) market segmentation may be described as an assumption that all consumers are unique and the needs of individuals may not be satisfied with a mass marketing app roach. Similarly, target marketing may be defined as a market segment profiled by its demographic, economic, and psychographic characteristics so that marketing opportunities may be evaluated. The precise nd attributes, as well as the demographic and non demographic variables influencing consumer selection and decision making process, would be useful in developing marketing and merchandising strategies. Chakravarti and Janiszewski in 2004 examined the infl uence of generic advertising on primary demand and brand preferences. They concluded firstly that generic advertising can increase or decrease the perceived differentiation among competing branded products and, thus, influence consumers choice. Second, inc reases in differentiation occur because generic advertising increases or decreases the weight consumers place on differentiating or non differentiating attributes. Generic advertisements that discussed a differentiating attribute decreased access to infor mation about the non differentiating attribute, which resulted in an increase in the importance of the differentiating attribute and increased price responsiveness. The importance of U.S. beef promotion programs within the four Asian nations considered in this study differs considerably and likely has evolved over time representing major implications for the beef industry and it s marketing strategies. Brands may segment the market and may or may not grow total demand for the product category. Such progra ms are often mandatory, where producers or exporters of a promoted good (i.e., beef) must contribute to the designed to avoid free riders In 2005,

PAGE 88

88 Carwell estimates that for every $500 million spent each year on U S agricultural promotion programs, the average cost to farmers is U S $1000. The Armington Trade Model Using economic models to evaluate changes in agricultural trade policy generally requires the conversion of policy changes into price effects. The Armington m odel use s these price shifts to determine how policy is expected to affect output, employment, trade flows, economic welfare, and other variables of interest (Jung 2004) The direction and magnitude of a trade policy change on individual variab les depends on the size of the shock as well as the behavioral relationships present in the economy. When evaluating policy shifts within an economic model, these behavioral relationships largely take the form of elasticities reflecting price responsivenes s of one set of variables to a change in a second set ( McDaniel and Balisteri 2002). Prices of goods produced in different countries do not typically move together. This behavior was first pointed out by Armington (1969). Ever since, it has become a stand ard practice among empirical trade researchers to treat goods produced in different countries differently and to assume a constant elasticity of substitution among them. Such elasticity for example, the elasticity of substitution between the basket of U. S. goods and that of Australian goods is referred to as the Armington elasticity (Lloyd and Zhang 2006) As a result, a key relationship for model analysis is the degree of substitution between goods according to their origin. In general, knowledge of el asticities is important for policy considerations. Changes in The size of these impacts will largely depend on the magnitude of elasticities. In 1990, Duffy, W ohlgenat, and Richardson studied the price elasticities of export demand for U.S. cotton and estimated feedback effects of U.S. prices on competing countries. T he usual Armington framework was extended to account for finite elasticities of export supply fr om U.S.

PAGE 89

89 competitors. The resulting "total" elasticities were more realistic for evaluating the effects of U.S. policy changes. Accounting for feedback effects of the U.S. price on other countries' cotton prices had a significant impact on the estimated exp ort demand elasticity. They concluded that for commodities such as cotton, for which the United States accounts for a large share of total world trade, the conceptually relevant elasticity for policy analysis is the total export demand elasticity because i t takes into account the feedback effects of the U.S. price on other countries' prices. Solomon and Kinnucan (1993) estimated the effects of government subsidized export promotion on the demand for U.S. cotton in the Pacific Rim (e.g., Japan, South Korea, Taiwan, Thailand, Hong Kong, and the Philippines) using an Armington type trade model. The results of this study showed a significant relationship between promotion expenditures and U.S. market share in four of the six countries examined. One of the two co untries exhibiting a non significant effect had very low promotion expenditures (Taiwan), suggesting that a minimal level of funding may be necessary to achieve a market response. The results also showed that export promotions have a carryover period lasti ng beyond one year, contrarily to what previous studies showed. They concluded that non price promotion is a viable policy instrument for increasing agricultural export and that the impact of export promotion on domestic producers and taxpayers still is un clear. Shiells and Reinert (1993) disaggregated U.S. imports into those from the NAFTA members and those from the rest of the world (ROW). Using quarterly data over 1980 1988, they obtained estimates for 128 mining and manufacturing sectors. Elasticities w ere estimate d using three specifications: (1 ) generalized least squares estimation technique, based on a Co bb Douglas price aggregator; (2 ) maximum likelihood estimation using a CES price aggregator; and, (3 ) a

PAGE 90

90 simultaneous equation estimator that uses a C obb Douglas price aggregator and employs a distributed lag model. Shiells and Reinert found the estimates to be relatively insensitive across the three alternative estimation procedures. Satyanarayana and Johnson (1998) empirically estimated the impact of export credit guarantee programs on U.S. wheat exports to major international recipients. An Armington demand specification was used to estimate U.S. shares of the wheat import market in six countries including the values of the credit guarantees as an add itional explanatory variable and used as a utility shifter proxy. An optimization framework was formulated to address the allocation of credits guarantees to optimize U.S. exports revenue. Results indicated that export credit guarantees raised the U.S. sha re in all six markets and based on the optimal allocation of the credit resources a revenue increase of $200 million should be expected if compared to previous years. Marginal effects showed that countries such as Brazil and South Korea are less responsive to credit guarantees increase, while market shares in Mexico and Tunisia will quickly respond to any increase in credit guarantees. Kapuscinski and Warr (1999) estimated the elasticities of substitution between imported and domestically produced forms of over 30 commodities in the Philippines that were intended for use in a 50 sector computable general equilibrium (CGE) model for the Philippine economy. Changes on tariff applied to imports were also included in the model to determine the resource or expen diture allocation between goods from different origins. For this research, an OLS based Armington specification was contrasted with other models to overcome the problem of incomplete adjustments in the marke t prices. Partial Adjustment Models (PAM) and Err or Correction M odels (ECM) were estimated. The PAM model attempts to measure economi c cost in response to quantity adjustments due to price changes. The ECM model emphasizes the time

PAGE 91

91 series characteristics of variables and postulated that non stationary va riables may form a stationary relationship in the long run that is called co integrating relationship. Results showed that dynamic models in estimation of Armington elasticities are preferred and, in this particular research, the ECM provided the most adeq uate characterization of the process of substitution between imports and domestic production. Gallaway, McDaniel, and Rivera (2003 ) developed a disaggregated set of Armington elasticity estimates. The authors consider explicitly the long run aspect that is applicable to applied partial and general equilibrium modeling. They provide estimates for 311 industries at the 4 digit SIC level over the period 1989 to 1995. Significant long run estimates range from 0.53 to 4.83. Long run estimates are up to five time s as large as short run estimates, and on average twice as large as the short run estimates. This is important since long run estimates are more appropriate for most trade policy analysis than short run estimates. Saito (2004) used two versions of Armingt on specifications (one for studies using multilateral trade data, and the other for those using bilateral trade data) to identify discrepancies between elasticities obtained from multilateral trade data and those obtained from bilateral trade data especial ly when trade consist largely of intermediate inputs. The author argued that the growth of outsourcing and associated changes in the composition of intermediate inputs trade may not be captured in multilateral trade data and hence may result in a bias in t he estimates obtained from these data. The Armington specification certainly attracts most of the critics on the CGE approach. In fact this specification entails two main constraints: first, it imposes an unique and constant substitution elasticity betwee n all pairs of goods from the different countries (including domestic goods), while there are good reasons to consider the case of changing elasticity of subst itution

PAGE 92

92 among some goods and that substitution elasticities may evolve with quan tities (Hillberry et al. 2006) ; and second, this specification does not acknowledge zero trade flows (Fmnia and Gohin 2007). In order to tackle thes starting econometric point is the methodology explained by Golan, Perloff and Shen (2001) who estimate by Generalized Maximum Entropy (GME) an AI demand system for the Mexican meat consumption allowing zero purchases by individual households. As they demonstrate, this GME approach is really useful in order to introduce (in) equality con straints on the parameters and thus is well suited to the analysis of corner solutions. In summary, the present study of the world beef trade will simultaneously estimate market and product demands as well as the market shares for each of the competing bee f suppliers. independently allocated among competing sources of supply or beef products. The use of a systems estimation technique will yield parameters that are less biased than if the simple Armington Constant Elasticity of Substitution (CES) were used. Instead, a CRESH specification based on previous work by Sato (1967), Hanoch (1971), and Sparks and Ward (1992) will be applied to calculate the impact of non tra de barriers (NTB) based on food safety issues. Analysis of the demand parameters will likely indicate the future pattern of trade between the nations considered. Results will be used to understand the forces driving the international market for beef produc ts and analyze the implication of the policies applied to prevent the dissemination of food born diseases.

PAGE 93

93 CHAPTER 4 EMPIRIRICAL TRADE MO DEL FOR BEEF PRODUCT S Introduction In this chap ter the conceptual framework is to estimate the demand for fresh/chil led and frozen beef products in South Korea, Japan, Taiwan, and Hong Kong from the United States, Australia, New Zealand and the rest of the world. Specifically, this study estimates the im pacts of economic variables ( prices and expenditures) and non econo mic variables (animal disease outbreaks and beef promotions) on product demands and market shares for U.S. beef products compared with beef from other sources. Using an Armington specification, the model suggested for this study is based on the premise tha t changes in trade flows can be distinguished not only by their kind but also according to their origin or place of production Beef Trade Schematic Representation Before turning to the specific nature of the substitutability among beef products, it is us eful to employ a generalized trade system that will be adopted for the present study. Accordingly, the schematic representation in F igure 4.1 displays the causal relationships between prices and quantities of beef products (fresh and frozen), two market re gions (supply a nd demand markets) consisting of a set of four exporting countries and four importing countries respectivel y, which is set forth for a one way trade. Figure 4.1 shows the United States, Australia, New Zealand, and the Rest of the World as su ppliers of beef products ( ) with prices ( ). The total supply of beef products is constrained to equal total demand in the importing markets for beef products ( ), which in this trade system is formed by South Korea, Japan, Taiwan, and Hong Kong. The average pr ice paid for beef in a given importing country is ( ) and is defined in Equation 4.7 Thus, the objective of this model is to obtain product demands and market shares in each importing market for beef products differentiated by the country of origin

PAGE 94

94 Figure 4 1. Schematic representation of the Armington trade system for beef products in selected countries of the Pacific Rim region. Prices are the crucial link to allocate products and t heir impact on product demands and market shares are presented in E quation s 4 17 and 4 19 respectively and are considered exogenous variables. Within this framework, there are several relationships operating to equilibrate product demand and supply. The r esulting econometric model represents a non linear system and parameters restrictions within each set of equations. Econometric procedures will be used to estimate product demands equations and assure the consistency and un biased Australia New Zealand Rest of the World United States Total supply for beef products Total demand for beef products Hong Kong (4) Taiwan (3) Japan (2) South Korea (1) ( Product Demand ) (Mkt. Shares) ( Product Demand ) (Mkt. Shares) ( Product Demand ) (Mkt. Shares) ( Product Demand ) (Mkt. Shares)

PAGE 95

95 characteristics of the re sults. Model estimates are used to develop a sensitivity analysis and examine the impacts of trade policies on beef consumption in selected Asian markets. Adapting the Armington Model to Beef Trade The Armington model is used to estimate demand elasticitie s between products from different sources and then used to simulate the effect of exogenous demand shocks from normal marketing efforts, from food safety scares such as BSE, and from trade restrictions. This approach is flexible and provides cross price el asticities between imports from different sources using estimates of the aggregate price elasticity of demand for imports (Jung 2004). Furthermore, demand elasticities as well as estimated coefficients of non economic variables can be used to formulate eff ective policies targeted towards expanding sales and market shares for U.S. beef. Price elasticities measure the responsiveness of trade flows to price changes. Elasticities of substitution provide the cross price elasticity between products from different origin countries and me asure the degree to which price changes could influence market share among the importing countries to a specific importing country. Since one of the objectives of this research is to measure the impact of restrictive policies on bee f trade, the size of those impacts largely depends on the magnitude of the elas ticity (McDaniel and Balistreri 2002). The Armington specification presents a two stage budgeting procedure in which consumers allocated their total expenditures in a sequence o f steps. In the initial step, a consumer or buyer decides how much of a particular product defined as to buy ( Equation 4 1). Next, given the total amount demanded, the buyers decide how mu ch to import from each country (Equation 4 2). Using i to represent the four demand regions and j to represent the four supply regions, the product import demand functions for beef for each r egion can be represented as in Equation 4 1 where the dot denotes the summation over all beef imports. This general demand function assumes that market demands for beef products are functions of the average price in the

PAGE 96

96 region income, population levels, and other demand drivers consistent with the underlying theory. Consequently, the following models specification is presented : (4 1) where is the total demand for all beef products in region i is the average market prices of beef products in region i is the gross domestic product in regio n i which represents national income, Pop i is the population of region i and captures other import demand references, promotions, etc). Fo total demands are derived given budget co nstraints, utility maximization, and price of competing products (See Appendix A). Product demands ( ) are functions of the market demand ( Equation 4 1) and the price ratio between the prices of the beef products in the market as shown in E quation 4 2 : (4 2) where is the price in the market for beef produced in j which includes the costs of tariff and preferent ial treatments (Sparks and Ward 1992), and is the price of all imported beef consumed in country i 1 Equation 4 3 shows that products are distinguished by place of production and product demands in a particular market are functions of market size and the ratio between product price a nd average price in that market. Thus, market shares equations can be derived to represent the percentage of total beef imports by a specific market and show that an improvement through pricing str ategies should yield an increasing share and vice ve rsa (Ar mington 1969): (4 3) 1 P rices include the costs of tar iff and preferential treatments and are expr essed in U.S. dollars to avoid problems with fluctuations in exchange rates. For convenience, p rices are deflated in importing countries by the national index to the base year 1995.

PAGE 97

97 Within this framework, there are several relationships operating to equilibrate the demand and supply of beef products and their most general functional representations are discussed below Since this model assumes that import demand must equal supply we have an equilibrium condition from which t wo restrictions can be obtained: and (4 4) Differences in import prices among demandi ng (importing) countries are driven by asymmetries in cost of insurance and freight, quality of products, market structures, non tariff barriers to trade, and information. Non tariff barriers (or NTB) are mechanisms that impede the flow of international go ods by imposing unilateral and arbitrary restraints to normal trade. The most common NTB are import quotas, exchange controls, subsidies, boycotts or bans, technical barriers, and voluntary restraints among others 2 A tariff is a tax levied by the governm ent in country i on goods imported into that country from country j (or import duty). The main objective of a tariff rate is to make country i products more competitive by increasing the price at which the goods from country j are sold. The effect of tarif f rates on prices of products from country j in market country i is quantified as: (4 5) where is the market price of the products, is the cost of tariff expressed in percentage terms, and represents import prices (CIF) that are a function of FOB prices and a prox y variable (V) created to capture the variation in cost over time : (4 6) Then, the average price paid for beef in a given importing country is defined as follows: 2 A product produced and consumed domestically does not incur in cost associated with shipping and barriers to entry. Consequently, this price is assumed to be equal to the average of all export prices in a particular market (Sparks and Ward 1992).

PAGE 98

98 (4 7) At this junction it is very important to reiterate that prices included in the data set collected for this study are in fact product prices in U.S. dollars for each beef product in each of the four markets and that these prices already i nclude all tariff le ved by the importing country. In the case of a model with a large set of products being trade d most of the equations already presented would not have practical solutions due to the large number of parameters in the model. A way to s implify this equation is to introduce the assump tions that (a) import demand is separable among import sources, (b) elasticities of substitution between all pairs of products within a group and within a market are constant (i.e. Constant Elasticity of Sub stitution CES), and (c) the elasticity of substitution between any two products competing in a market is the same as that between any other pair of products competin g in the same market (Armington 1969). Equation 4 8 represents the Armington specificatio n that imposes the CES functional form on the product demand function ( Equation 4 3) implying the separability among product sources (Alston et al. 1990). In terms of utility function specification, these assumptions are equivalent to the specification th at the s 3 is a constant elasticity of substitution (CES) having the form: (4 8) where is a share parameter, is the single CES in the market, is the demand function for goods (i.e., beef products) in country i and is the import demand in the market for a product produced in the j U ) into E quation 4 3 the necessary and suffi cient conditions are that (a) the Marginal Rate of Substitution (MRS) between any pair of products competing in the market is independent of demand for any other product(s) in that market which is known as the assumption of independence, and (b) t he 3 See Appendix A

PAGE 99

99 quantity index representing the aggregative character of the func tion is linear, homogeneous, and shows a CES. 4 Although the homotheticity assumption simplifies the model specification and the separability assumption reduces multicollinearity pr oblems, the Armington assumptions of separability and homotheticity are strong restrictions for practical applications. In the first assumption, if a good is differentiated, changes in consumer's budget might not be reflected in the same proportion to all products. If an increase in the budget is realized, it would be allocated more to the more preferred product. In other words, expenditure elasticities may not be unitary (Hanoch 1971). In the case of the second assumption, if prices of substitutes are omit ted, the own price parameter estimates are positively biased so that the own price elasticitie s would be underestimated ( Jung 2004 ). Finally, in the case of the CES, this assumption restricts responses of the import demand of each product to the price chan ge relative to the price index for the good to be the same for all products. In order to specify the equations of the system in more specific functional forms, assumptions must be made regarding product substitutability. As discussed earlier, the assumptio n of a constant and equal elasticity of substitution (CES) between all products in a particular market may be overly restrictive. However, restrictions on the elasticities of substitution which force them to vary by a common, constant proportion, yet allow ing for differences in substitutability between products, are more reasonable and embedded in the CRES function. Since these restrictions are placed on product substitutability, they reduce the number of parameters to be estimated, and yet retain some flex ibility. The imposition of the CRES 4 The form is regarded as a quantum index of all elements that belong to the market. For convenience, demand function.

PAGE 100

100 technical relationship determines the functional nature of the product demands and market shares, while the other relationships in the model remain unaffected by the use of the CRES function (Sparks and Ward 1992). The C onstant Ratio Elasticity of Substitution An important aspect of the CES assumption is that it implies the separability between occurring in two stages (Varian 199 2; Webb et al. 1989) In the initial stage, the importer decides how much of the good is going to be imported ( Equation 4 1); and then, in the second stage ( Equation 4 2), given the total amount imported, the importer decides how much product to i mport from each country (D avis and Kruse 1993). The single CES restricts responses of the import demand of each product to the price change relative to the price index for the good to be the same for all products (Winters 1984; Weatherspoon and S eale 1995). Therefore, the issue of substitutability must be addressed in order to formulate a more specific system. Previous studies applied to agricultural trade analysis have raised questions about several properties of the Armington model. Studies by Alston et al. 1990; Moschini et al. 1994; Seale et al. 1992; Weatherspoon and Seale 1995; and Winters 1984 all showed the biased character of the Armington specification when it is based on the CES functional form. Furthermore, Yang and Koo 1993 suggested a less restrictive set of assumptions on demand relationships than those of the Armington model in order to avoid inconsistent parameter estimations. In order to bring prices, quantities, and other factors into the model, this study suggests a functional form: one product, many suppliers demand function using a more flexible form, the Constant Ratio of Elasticity of Subst itution (CRES) (Mukerji 1963, Hanoch 1971). That is, its Elasticity of Substitution (ES) varies along a mapping of isoquants (i.e., all i nput bundles that

PAGE 101

101 produces exactly y units of outputs) and differs between factors (Varian 1992). The choice of the CRES rather than the CES is justified because it allows the estimation of parameters without restricting the substitutability to be the same between every product, and consequently fully exploiting the attributes of the Armington's theoretical framework (Sparks and Ward 1992). The CRES technical relationship determines the functional nature of the product demands and market shares from compet ing supply regions. For this study, we assume that the CRES function permits variations of the elasticities of substitution among competing beef products in a market i to vary by a constant proportion, thus allowing for differences in substitutability betw een products of different origin within a market. This assumption increases the flexibility of the model by reducing the number of parameters to be estimated. Following earlier specifications used by Mukerji (1963), Hanoch (1971), and in particular Sparks and Ward (1992), the derivation of the product demand functional form from the CRES technical relationship is developed as follows: (4 9) where the is the product import demand for beef in region i in period t and is the j The is the share or distribution parameter and is a function of the exogenous variables affecting the system (e.g., market structure) while the parameters and are substitution parameters that are functions of variables like generic advertizing, country origin, and information. The degree of substitutability between imported beef from different sources of supply is c aptured by the CRES. Consequently the higher the value of this parameter, the closer the degree of substitution. In other words, a high value of this parameter means that beef products from all sources are considered by consumers to be virtually identical Conversely, a low value

PAGE 102

102 of the parameter means that the two products are dissimilar or, that they are weak substitutes (Kapuscinski and Warr 1999). The imposition of the CRES technical relationship on the markets clearly shows that all elasticities of s ubstitution ( ) vary proportionately with the common factor of proportional change 5 That is, the elasticity of substitution and the value share of the beef product in the beef market are represented as (4 10) where (4 11) These elasticities play a significant role in trade modeling, especially when analyzing the impact o f trade policies. For example, when a tariff applied to beef imports is increased to protect the domestic beef production in a country such as South Korea, this change automatically raises the domestic price of the imported beef. Nevertheless, the effect t hat this change in the tariff has on the price of the domestically produced beef is what determines its domestic resource allocation ef fects (Armington 1969). In the special case of specific tariff applied to U.S. beef products competing in Asian markets, if all competing beef products are perfect substitutes, then the price of beef from the United States will necessarily change (increase) by the same proportion as that of the other suppliers change (decrease). However, if the goods are imperfect substitute s, prices may not change by the same proportion. Thus, the impact that changes in trade policy have on the structure of the markets (e.g., market shares) basically depends on the degree of substitutability between imported beef, and this is what the model that is suggested for this study captures (Kapuscinski and Warr 1999). 5 Previous studies by Mukerji (1963), Gorman (1965), Sato (19 67), and in particular the paper by Hanoc h (1971) suggested that all could vary proportionately, with a common factor being defined as: where is a weighted average of with the factor shares as weight.

PAGE 103

103 The and coefficients have particularly important meanings in the CRES function. First, the shows that each country has an initial share of the import markets and those shares can differ for a fixed level of import demand. Clearly, does not just happen. It must depend on many conditions such as non tariff restrictions, previous trade history, infrastructure, food safety concerns, preferential treatment, etc. Second, the produ ct category also differs across countries and shares change as the imports change. Both and reflect those differences and like the they do not just happen They reflect true quality and other attribute differences among e xporting countries as well as perceptions about a particular product from country j in country i attributes. Anything that enhances this knowledge also impacts the (e.g., generic advertizing, product attributes). Later in the chapter, more specific forms for capturing the levels of both and are developed showing that both coefficients may change over time. Given t he CRES market demand function, Equation 4 9 the following optimization steps will define the product demand function for beef products. The first order condition for optimum product mix implies (4 12) where is the price level in the market and (4 13) Th us, the partial derivatives in Equation 4 13 depend only on ratios of quantities of products deman ded in the market, and these ratios in turn depend only on ratios of the product prices which mean that market shares must depend only on relative prices of the pr oducts in the market (Armington 1969). The fulfillment of this condition determines the

PAGE 104

104 optimum quantities of beef imported by country i Substituting Equation 4 13 into 4 12 can be written as : (4 14) Rearranging terms in Equation 4 14 the market clearing condition for which, the Marginal Rate of Substitution (MRS) between competing products are equal to the price ratios (Varian, 1992) can be written as: (4 15) Further substituting Equation 4 15 the following expressions are obtained : (4 16) then, (4 17) At this point, given Equation 4 9 and using the previous equation it can be show n the functio nal forms for product demands are : (4 18) where the function clearly shows the relation between the CRES and the market clearing condition that the MRS between compet ing products are equal to their price ratios. Re arranging the functi 18 (4 19) then the corresponding market share equations can be defined as

PAGE 105

105 (4 20) Equations 4 18 and 4 20 specify the functional forms of the product demand and market share equations imposed by the CRES technic al relationsh ip on the markets and hence non linear estimation techniques must be used to quantify the parameters (Sparks and Ward 1992) 6 Using Equation 4 18, the following form is obtained : or (4 21) If and are in turn functions of events and changes, then clearly the product demand is highly non linear and must be solved using a non linear technique. Since the objective of the m odel is to estimate coefficients of demand, the fi r st estimate is of the differing and substituting country intercept and slope coefficients using exponential an d logistic function and the results are incorporated into Equation 4 21 to obtai n the corresponding estimates of the model Since the double log functional form is used to estimate each country product demand, the final coefficients were the estimated ela sticities of the corresponding variables. Therefore, for convenience and to incorporate the impact of promotions, food safety and other factors into the analy sis, the final product demands Equation 4 21 can be written as follows: 6 See Chiang 1987 for detailed explanation of the log function properties.

PAGE 106

106 (4 22) where the first component, is a constant of integration that represents the effects of distribution and differentiating coefficients on market shares; the second component, is the estimated value of the product demand relative price elasticity and is given by ; the third component, is the estimate value of the product demand market size elasticities; and the last componen t, is a white noise error of the estimable model and allows for error in the adjustment of the inputs to their utility maxim ization level. Thus, comparing Equations 4 18 and 4 21 the following identities are obtained: (4 23) and (4 24) (4 25) Equation 4 22 a nd its corresponding identities clearly show that the parameters a nd are non linear, non constant and may change over time; that is, these parameters are some measure of the coefficients of the cross function (Sparks and Ward 1992). The adju stment process captured with the s e relation s are the most revealing par t of the model, since it shows whether or not the parameter is moving in a particular direction representing the changes in beef demand. Thus the system for beef trade as specified for econometric estimation will obtain elasticities of substitution to anal yze the impact of the parameter coefficients included in the CRES function. Recalling th e previously defined Equations 4 7 and 4 22 the price elasticity can be obtain ed substituting them back into Equation 4 22 as follows: (4 26)

PAGE 107

107 Taking derivatives with respect to and dividing the system by the following relationsh ip is obtained : (4 27) Finally, su bstituting Equation 4 11 into Equation 4 27 the price elasticity can be written as: Share value ) (4 28) In summary, Armington (1969) proposed that the elasticities of substitution among suppliers are constant and equal for any given pair in each market and the corresponding price ratios are constant over the time. However, the liter ature examined for this study suggests a different approach implying that the elasticity of substitution will vary proportionally to maintain a fixed ratio, but they are allowed to be different between any two pairs of products competing in the same market (Artus and Rohmberg 1973). These functional forms known as CRES technical relationships are considered less restrictive than the CES approach in terms of the degree of product substitutability in addition to a reduction in the number of parameters to be e stimated and allow some flexibility in the model. Analysis of Trade Variations In the Armington (1969) framework, trade flows are explained by a behavioral function that includes the imperfect substitutability of the good supplied by competing exporters vi a the that the reasons for and are not fully explained and exploited. In this context, this

PAGE 108

108 study introduces a CRES technical r elationship to analyze trade anomalies or exogenous period 1995 trough 2007. At thi s junction, it is clear that three components of the CRES technical relationship (4 9), and determine the functional nature of beef product demands and market shares. The parameter represents a non constant (i.e., non among imported goods that is a funct ion of variables such as generic advertizing (G) country origin (O) and the parameter variables indicating changes (C) Likewise, the parameter represents a function of v ariables (S) can be included (e.g., beef import ban due to BSE) and market size (M). Thus, a major consideration at this point is that both the differentiating coefficients and the share distribution coefficients are influenced by exogenous conditions. Consequently and can be represented in a functional form as: (4 29) and (4 30) (4 31) The parameter coefficients would be expected to cause shifts in product demand and in the existing state of knowledge about the product, real attributes of the good, current levels of consumption, and the quality and intensity of and That is, these coefficients could lead to a reduction or increase in the elasticities of demand, along with absolute shifts on the

PAGE 109

109 demand for beef products. Further, without much l oss in generality the previous Equations 4 29, 4 30, and 4 31 can be also written as: (4 32) an d (4 33) (4 34) assuming that and ; and that the (Z) variable represents a set of dummies and variables representing promotions and the type of beef In order to take into account the parameter, this study suggests the use of a logistic regression to capture the impact of each of the factors affecting and an exponential form to estimate the effect of the variables affecting In both cases, a set of dummy v ariables are introduced in the model to understand fully the effect of each parameter. For the sake of clarification and without distracting the attention from the main model specification, the logistic function can be written as and (4 35 ) where represents each of the parameters and ; and the "input" ( ) indicates a regression model containing the respective subset of variables and dummy variables repres enting the market and product characteristics. The logistic function is useful because it can take as an input, any value from negative infinity to positive infinity, whereas the output is confined to values between 0 and 1. The variable ( ) represen ts the exposure to some set of trade factors, while represents the likelihood of a particular parameter, given that set of factors. The exponential function is used because in this case is always positive (a required assumption for parameters), although it gets arbitrarily close to zero. Thus, the parameter and can be written as:

PAGE 110

110 (4 36 ) and (4 37 ) (4 38)

PAGE 111

111 where i, j =1,2,3,4 ; are dummies representing each importing country (equal to 1 if belongs to 0 otherwise); and are dummies representing each exporting country in the market or c ountry (Gujarati, 2003). Dummy variable devices classify data into mutually exclusive categories. Each variable is defined in such a way that if the qualitative variable has m categories, it introduces only (m 1) as a dummy variable. One approach for de aling with dummy variables is to drop one of the variables in each category such as dropping the first year in the case of time trend variables, one of the beef types (i.e., frozen), or one of the importing or exporting countries (i.e., Rest of the World) The category in which no dummy variable is assigned is known as the base and all comparisons are made in relation to it. The intercept value represents the mean value of the base category. The coefficients attached to the dummy variables are known as the differential intercept coefficients because they tell by how much the value of the intercept that receives the value of 1 differs from the intercept coefficient of the base category (Gujarati, 2003). Since Equations 4 29, 4 30 and 4 31 are multiplicative they can be expressed as follows: (4 39) and (4 40) (4 41) At this point, it is easy to combine Equations 4 18 or 4 20 w ith Equations 4 3 8, 4 39, and 4 40 to fully estimate the impact of beef promotions on Pacific Rim markets. However, before doing that we must substitute identities 4 24 and 4 33 into Equation 4 42 to obtain the link necessary to evaluate the effect of promotions and other non economic indicators into the product demand functions The suggested link can be expressed as: (4 42)

PAGE 112

112 Finally, substituting Equations 4 32, 4 34 and 4 42 into Equation 4 21 is written as (4 43) The use of a non linear regression model to estimate the parameters of the model will yield consistent estimates in the market share equations for the years 1997 to 2007 for the beef exporti ng countries considered in the study. These parameters will measure the strength of the influence of food safety, non trade barriers, and other factors on the demand for differentiated products, and measure the substitutability between beef products within each Pacific Rim market considered. Given the increa sed competition and the disease driven restrictions imposed on U.S. meats by its traditional importers, understanding and differentiating the importance of economic and non economic factors is crucial i n determining the changes in demand for U.S. beef. For instance, if the government of the Republic of South Korea were to impose or increase its tariffs on imported beef in order to protect its domestic beef production, the impact on its major beef supplie rs could be estimated. The model can also assess the impact of economic crisis on the international beef markets. In particular, in the case of 1997 Asian financial crises that heavily affected country and household incomes and weakened the demand for impo rted beef, the model can measure the effect on products originated in the United States and determine the extent of the damage in terms of market share lost. In short, any of the independent variables of the model could be shocked and their corresponding i mpact measured. In terms of U.S. beef exports, this can yield important insights into the consequences of policy measures aimed to prevent food born diseases and impact of commodity advertizing campaigns designed to communicate the quality and safety chara cteristics of the product.

PAGE 113

113 CHAPTER 5 EMPIRICAL RESULTS Introduction This chapter presents results of the model introduced in Chapter 4. The product demand equation (WQQ) for beef products in the Pacific Rim region includes a relative price component, a variable indicating the domestic demand for beef in each importing country, a set of parameters indicating market share differentiation and the substitutability among products, and the assumption that defines a constant rate elasticity of substitution i n the demand for differentiated beef products. The initial estimation of the beef trade model was carried out using a non linear single equation regression w h ere the objective function is the sum squared residuals and the resulting regression coefficients are the proposed changes in the parameters (Hall and Cummins 2005). Data Set The t rade data used in this study were originally organized in terms of quantities and export values by country and trade partner fresh/chilled and frozen beef (HS codes 0201 and 0202) during the period 1995 to 2007. There were some missing data points on the world trade data, which complicate the econometric estimation of the model. The majority of the missing values was concentrated in the earlier years (1995 and 1996) and could not be confirmed by other data source s This circumstance was more evident in countries such as Taiwan and Hong Kong, in particular, for beef products from New Zealand and Australia. Therefore, the decision was made to use only data from 1997 to 2007 in th e creation of the final data set. Since missing data will result in the entire observation in which the missing data are contained b eing thrown out and with only 343 observations, missing prices and quantities for the year 1998 were calculated.

PAGE 114

114 Import pri ces were regressed on world product prices for all of the years of the data and the estimated relationship was then used to estimate an import product price for a year when it was miss ing. Finally, missing values in the original data set were approximated through linear interpolation between existing quantities (Chiang 1997). Manipulations were carried out on these data sets, after which they were merged by year and by country to create the final data set. These manipulations resulted in a data set that sho ws space as well as time dimensions, which are typical characteristics of panel data sets where th e same cross sectional unit (such as a country) is surveyed over time. Thus, import quantities are expressed in thousands of metric tons and values in thousan ds of U.S. dollars per metric ton. Both quantities and values are used to derive import prices. Given the nature of the data, prices are treated as exogenous variables and already include the corresponding tariff schedules for beef products originated outs ide the country. For any given year, the data on beef trade represent a cross sectional sample. Then, in the case of fresh beef for any given year there are 16 observations and 11 time series observations ; thus there is in all (16x11) = 176 (pooled) obser vations on fresh beef (Gujarati 2003). Thus, combining cross sectional and time series data into a pooling data set give s more degrees of freedom, reduces or eliminates the issue of multicollinearity among variables and minimizes the bias that might resul t from aggregation. The rest of the data consist of country specific statistics such as GDP, CPI, population levels, total food imports, and country domestic beef production and consumption levels that were obtained from the UN and FAO data bank and arran ged in a country by country basis. Product Demand Equation The product demand equation is calculated assuming that there is a product differentiation among beef products and that the rate of substitutability is no constant across

PAGE 115

115 country corresponding product demands can be written in its simple form as follows: (5 1) Each of the theta ( was estimated separately and is some function of and as set forth in the previous chapter. Once estimates of and are known, then the a re determined. Equation 5 1 is estimated with a pooled time series cross sectional estimator and is double log form. After their calculation, these param eters were incorporated into Equation 5 1 to obtain the corresponding product demands. Note that for a given set of one can easily show the corresponding elasticities. The resulting model had one equation for each country and type of beef product (Fresh/Chilled or Frozen). The objective is to compare the results from each country and draw conclusions regarding th e behavior of their paramete rs in terms of their values, si g n s, significance, and consistency with the economic theory. The reminder of this chapter will discuss the result of the non linear single equation used to estimate the parameters of the product d emand equations. All model estimates were obtained using TSP software, as well as the statistics indicating the performance of the model. The magnitudes of the estimated parameters or elasticities will be discussed and contrasted with the economic theory a s they indicate the responsiveness of the dependent variable (product demand) to the explanatory variables. In general, the results of the estimation are good in that they make economic sense and generally correspond with what one would expect given the tr ading situation described in earlier chapters. Model Results The following section describes the parameter estimates for the product demands functions ( ) or as defined in the TSP programming, the WQQ dependent variable. As presented in

PAGE 116

116 Chapter 4, Equation 4 18 product demands include a set of product and market differentiation parameters, domestic demands ( ), and a price ratio between ave rage domestic prices ( and import prices ( This set of parameters is considered the base of the model and define s the elasticities of demand among products competing in a specific market. Therefore, using (4 38) and the corresponding d ummy variables specified in Equations 4 35 through 4 37 the model calculates their coefficients, including those for beef promotions and type of beef, and their corresponding T values and supportive statistics as illustrated in Table 5 1. The t values are compared at 1.96 for a two tail 95 % confi dence level, which means that any value greater than |1.96| will show evidence against the null hypothesis that the parameter is equal to zero and hence conclude that the parameter is statistically significant at the 95% confidence level. Since these are estimates from logistic and exponential regressions, one cannot immediately infer that the coefficients indicate a dimensional change in the d ependent variable. Conclusions i n this regard are presented later in th e simulation chapter. In general, the statistics of the model correspond to what one would expect given the annual data. The Durbin Watson statistic is close to two, suggesting little or no serial correlation in the data. With respect to the R 2 value, it i s reasonably high for the type of model ( hig hly non linear regression) and data (p anel) and implies that about 76 % of the variation in the (log of) product demands parameters is explained by the (log of) relative sub parameters, type of beef and commodity promotions. Therefore, from a purely statistical viewpoint, the estimated regression fits the data quite well (Greene 2003). Table 5 1 illustrates three sections representing the parameter estimates for the and components of the CRES function applied to the product demand Equation 4 17

PAGE 117

117 The top portion of the table represents estimates for the sub parameters for K ( related to the parameters; the estimates in the second portion of the table belong to the sub parameters G ( and are components of the parameters; and the estimates in the last portion of the table are the sub parameters H ( ) related to the parameters. Note that in Table 5 1 the last category of the coefficients within any and dummy class is not shown as well as the first category within a dummy class As explained in Chapter 4 one of the variables of each category was dropped to facilitat e the estimation of each parameter and avoid issues of perfect multicollinearity (Gujarati, 2003). The signs of the parameter estimates for the statistically significant variables are theoretically consistent. However, in terms of the direction of the imp act ( positive or negative signs), they cannot b e fully evaluated at this stage since they are components (sub parameters) of the product demand parameters (Equation 4 21). Their impa (Equation 4 22) will be analyzed in Chapter 6 where the final simulations of the model s are presented. Nevertheless, F igures 5 1 through 5 4 are introduced in this section to illustrate the behavior of the and product demand parameters across time. In order to determin e whether there has been a significant change in the variables affecting substitution and market participation, the G, K, and H parameters and sub parameters are analyzed using their corresponding t values to detect significant changes with respect to the mean. The majority of the c oefficients presented in Table 5 1 are statistically significant when compared to their mean value at the 95 % level of confidence. However, there are some exceptions of particular relevance that might require some consideration.

PAGE 118

118 Table 5 1 Parameter e stimates and t values Parameter Description Estimate t value Beef demand in Hong Kong (HK) 4.573 0.471 Beef demand in South Korea (KR) 4.633 0.482 Beef d emand in Japan (JP) 0.740 1.423 Demand for Rest of World products 3.693 0.375 Demand for U.S. products 1.189 4.522 Demand for Australian products (AU) 0.990 3.778 Demand for N. Zealand products (NZ) 1.028 4.185 Demand for U.S. products in S. Korea 5.160 0.530 Type of beef products (Frozen) 0.330 2.371 Commodity promotion (1: yes, 0:no) 0.621 0.744 ROW market share (1997) 1.610 4.575 ROW mar ket share (1998) 0.667 1.920 ROW market share (1999) 0.254 0.759 ROW market share (2000) 0.260 0.771 ROW market share (2001) 0.332 0.970 ROW market share (2002) 0.623 1.771 ROW market share (2003) 0.746 2.081 ROW market share (2004) 1.981 4.761 ROW market share (2005) 1.021 2.915 ROW market share (2006) 1.190 3.217 ROW market share (2007) 2.033 4.850 U.S. market share (19 97) 1.717 3.094 U.S. market share (1998) 0.484 0.636 U.S. market share (1999) 0.092 0.122 U.S. market share (2000) 0.065 0.086 U.S. market share (2001) 0.141 0.185 U.S. market share (2002) 0.387 0. 511 U.S. market share (2003) 0.485 0.640 U.S. market share (2004) 0.639 0.869 U.S. market share (2005) 5.262 5.956 U.S. market share (2006) 3.899 4.878 U.S. market share (2007) 0.662 0.899 AU market share (1998) 1.566 2.768 AU market share (1999) 1.122 2.058 AU market share (2000) 0.861 1.572 AU market share (2001) 1.155 2.037 AU market share (2002) 1.359 2.335 AU market shar e (2003) 1.388 2.368 AU market share (2004) 3.073 4.778 AU market share (2005) 2.109 3.440 AU market share (2006) 2.288 3.736 AU market share (2007) 3.091 4.711 NZ market share (1998) 0.801 1.49 1 NZ market share (1999) 0.299 0.570 NZ market share (2000) 0.214 0.416 NZ market share (2001) 0.028 0.054 NZ market share (2002) 0.122 0.239 NZ market share (2003) 0.454 0.863 NZ mark et share (2004) 2.239 3.842 NZ market share (2005) 1.188 2.202 NZ market share (2006) 1.313 2.410 NZ market share (2007) 2.107 3.580 NOB 343 D W 1.678 R 2 0.763

PAGE 119

119 In the case of both the differen tial intercept and dummy coefficients are statistically insignificant for all parameters ; however, the signs of the parameters for Korea and Japan are negative indicating an inverse relationship between these estimates and the parameter With res pect to the estimates for U.S. products in South Korea, the results show negative and insignificant values, while all exporting countries show positive and significant estimates. In the particular case of the United States, the estimates are posi tive and very significant w hen comparing with the mean but at this point it does not indicate either the size or importance of the observed variable or it s impact on the elasticities. However, these results could indicate a tendency for beef importing coun tries to become somewhat more responsive to the of imported beef. Dummy variables for Japan, Taiwan, and Hong Kong were initially included in the model to see whether these countries had different impacts on and parameters. However, these variables were all statistically insignificant and showed clea r symptoms of multicollinearity; therefore they were not reported. The type of beef, defined by the parameter DF1, is statisticall y significant but again the effect o f this variable will be discussed in Chapter 6 However, it is possible to assume that the form of the product (fresh or frozen) has a substantial in direct impact on which should be translated in the elastici ties between fresh and frozen beef products. On the other hand, estimates for beef promotions ( GGI ) show a negative and insignificant value with respect to the mean (first year), which somehow refutes initial expectations of the effect of this variable on the demand for differentiated b eef products. The can be interpreted as influencing the share that the exporter j has in the ith import market when equals T heoretically, must be positive, should

PAGE 120

120 one and each estimate should be in the (0, 1) i nterval (Armington 1969). However, when their impact across time on U.S. market shares is illustrated, the range is above one in the first year (Figures 5 1 through 5 3). This circumstance is the result of including the te rm in the model specification, which allows the market share parameter for beef imported from the United States to have a higher initial value. At this stage it is important to mention that the BSE effect is nested on the time trend as show n in the last portion of Table 5 1 and that the significance of these parameters is relatively more important in the evaluation than the sign of the dummies. In the case of the base used to compare the estimates is the Rest of the World. With this in mind, it is clear that U.S market shares have been affected by th e BSE scare when comparing the t values after the BSE announcements All sub parameters sho w insignificant values with respect to the mean, except for the first year and the post BSE coefficients. Also, the effect of the BSE announcements in North America shows a significant effect on the Australian and New Zealan d market participation compared to previous years. The figures below illustrate the significance of the estimates in Table 5 1 and their impact on the parameters and across time. One of the most revealing aspect of these figures is the fact that in each import market the parameter values of are constant across beef suppliers and time. That is, all values for the S outh K orean beef market are about 52%; for the Japanese market these values are about 2%, and for Taiwan and Hong Kong the values are 1% respectively. In the particular case of the parameters there are two point to be considered: first, it is suitable to think that regardless of the origin of the beef product and how much is available for consumption, markets demands show constant rate of substitution between beef and all other goods in that market and that consumers substitute beef for other goods at a proportional rate across time; and second, they

PAGE 121

121 have a measurable impact on the elasticity parameters only in the case of the South Korean beef market. Figure 5 1 South K orea n product demand parameters for be ef imports (US AU NZ) In the case of the coefficients, it is clear that only in the case of the South Korean Market there is some variation across time, although the range of this variation is only about 2 p ercentage points. For example, F igu re 5 1 shows values for the United States, Australia, and New Zealand. In the case of U.S. beef, these values vary between 56% and 58%; between 61% and 63% in the case of Australian beef; and between 60% and 62% in the case of beef from New Zeal and. In the rest of the markets the fluctuation of the coefficien ts is almost insignificant (+/ 1%), which is clearly shown in Figures 5 2 and 5 3

PAGE 122

122 Figure 5 2 Japan product demand parameters for beef imports (US AU NZ) With respect to the parameters, all figures show a clear fluctuation across time and across beef suppliers. Comparing the coefficient results for the United States, Australia, and New Zealand, the minimum and maximum value f or each of these suppliers were 0. 2% (2005 to 2006) and 111% (1997 ) for U.S. beef products; 20% (1997) and 60% (2005) for Australian beef; and 12% (2002) and 26% (2004) for New Zealand beef. In addition, although it is impossible at this point to identify the full impact of these coefficients, it is evident that the BSE announcements in 2003 have an important effect on the trend.

PAGE 123

123 Figure 5 3 Taiwan and Hong Kong product demand parameters for beef imports (US AU NZ) The BSE largest impact is, undoubtedly on U.S. beef products showing a decline in market participation of 57% across markets between the years 2003 and 2005. Overall, the trend show s a reduction of 83% in U.S. beef market participation between 1997 and 2007. Figures 5 2 and 5 3 also show the other side of the BSE scare on these Asian Markets. In the case of beef products from Australia and New Zealand, the overall upward trend on the coefficients is very clear, showing an incre ase of almost 38% in the case of Australia and 15% in the case of New Zealand between 1997 and 2007. In particular, after the BSE announcements the increment on the trend was almost 22% and 14% respectively. This upward trend reflects, in

PAGE 124

124 theory, the growth in market shares of these two suppliers during the last years but the real magnitude of their impact will be disclosed in the simulation chapter. Table 5 2 presents summary statistics for and in all markets during the 10 year period of this research. Although, there is a difference in significance between the minimum and maximum values, numerically they show the variation of each parameter across time. Thus, for the parameter the range of variation is abou t 63.6 percentage points, the variation between the parameters is about 51 percentage points, and in the case of the parameter the range of variation is approximately 111 percentage points. Table 5 2 Summar y s tatistics product d emands p arameters Parameter Mean Std.dev. Minimum Maximum 0.142 0.219 0.010 0.646 (1.162) (4.261) 0.163 0.266 0.005 0.515 (0.158) (23.749) 0.353 0.298 0.002 1.113 (0.485) (4.17 7) As stated before, it is impossible to immediately infer the impact and direction of the parameter co efficients shown in Tables 5 1 and 5 2 Their analysis will require a chapter alone (Chapter 6), with simulations showing the magnitude and impact of t hese parameters on product demands, market shares, relative prices, and elasticities of substitution among beef products. The set of figures introduced in C hapter 6 reproduce the results discussed before and place them under different scenarios in order to draw the conclusions of the study.

PAGE 125

125 CHAPTER 6 SIMULATION ANALYSIS Introduction The previous chapter concentrated on a general discussion of the econometric estimates obtained using the Armington model and the CRES assumption about the elasticity of s ubstitution. In this chapter, a more detailed analysis of the significance of the different parameters and sub parameters estimates from Chapter 5 will be presented. Thus, the objective is to simulate the model specified in Chapter 4 in order to evaluate t he effect of changes in relative prices, quantities demanded across time, and commodity promotions on product demands across four beef markets in the Pacific Rim region. Under different price, promotion, and trade flow scena rios, this section will compare the product demands and market shares relative to variations in the mean or average price or quantity. The simulation process involves a set of preliminary steps that were conducted using the TSP program, where a ll the variables of the model are set to t he mean level (all dummy variables are set to zero) in order to compare the changes in the dependent variable when one of the indepen dent variables varies, and all other input parameters are held constant at their mean value. The following step is then to predict product, market share demands, and elasticities using calculated relative prices, commodity promotions, and quantities at their mean values. When calculating these values for each of the variables, their respective means were included and multiplie d then by an adjusted parameter set to one for simulation purposes (i.e. ). The simulations measured the strength of the relationship between the independent and dependent variables. Tables 6 1 throu gh 6 4 illustrate the base values estimated for each beef exporter in each beef market.

PAGE 126

126 Table 6 1. South Korea : estimated base values Country of Origin Product Demands Fresh Frozen Relative Price + Elasticity Fresh Frozen United States 129.03 44.17 1.3416 1.68 0.78 Australia 8.75 4.46 0.9455 2.34 2.03 New Zealand 6.63 3.56 1.0200 2.33 2.01 In Millions of Metric Tons. + U.S. dollars per Metric Tons. Table 6 2. Japan : estimated base values Country of Origin Product Demand s Fresh Frozen Relative Price + Elasticity Fresh Frozen United States 74.96 53.41 1.3416 0.65 0.75 Australia 23.19 16.54 0.9455 0.93 0.95 New Zealand 20.66 14.74 1.0200 0.93 0.95 In Millions of Metric Tons. + U.S. dollars per Metric Tons. Table 6 3. Taiwa n: estimated base values Country of Origin Product Demands Fresh Frozen Relative Price + Elasticity Fresh Frozen United States 18.80 13.43 1.3416 0.21 0.44 Australia 5.80 4.15 0.9455 0.83 0.88 New Zealand 5.17 3.70 1.0200 0.84 0.89 In Millions of Metric Tons. + U.S. dollars per Metric Tons. Table 6 4. Hong Kong : estimated base values Country of Origin Product Demands Fresh Frozen Relative Price + Elasticity Fresh Fr ozen United States 14.76 10.55 1.3416 0.21 0.44 Australia 4.55 3.26 0.9455 0.83 0.88 New Zealand 4.06 2.91 1.0200 0.84 0.89 Millions of Metric Tons. + U.S. dollars per Metric Tons Product demand measure s beef products. Recall that the models used for this study distinguishes products by their region of production and that they are a function of the size of its market demand and of product prices. Keeping this concept in mind, the set of tables above show that on average, the total fresh beef market in all four importing countries was 1.8 times bigger that the frozen market (316.36 M.MT vs. 174.9 M.MT) and that the largest demand for fresh products was in South Korea with more that 45% of the market, follow ed by Japan, Taiwan, and Hong Kong with 37.5%, 9.4% and 7.3% of the total fresh product demand. On the other hand, during the 1997 to 2007 period the largest demand for

PAGE 127

127 frozen beef products was in Japan with more than 48% of the market, followed by South K orea, Taiwan, and Hong Kong with 29.8%, 12.1%, and 9.5% of the frozen product demand. On the s upply side, the United States was by far the largest supplier of fresh and frozen products with 75% and 69% of the market respectively, while Australia provided 1 3.3% and 16.2% respectively and New Zealand 11.5% and 14.2%. The importance of these results from the U.S. point of view is that in order to re gain market access, private (consumer level) and public (government level) confidence need to be improved. Empi rical results seem to indicate that trade regulations based on food safety concerns, not price, drive the presence of U.S. beef in these four markets. At the government to government level, negotiations and inspections of the beef supply chain have been d one to re assure that U.S. beef is produced to the highest levels of safety and quality. With respect to private efforts, they are mainly focus ed o n promoting U.S. beef in these markets using several types of marketing campaigns that reinforce the attribut es of quality, characteristic U.S. beef taste, and the safety of the products. Surprisingly, the results from this study show beef promotions parameters that are statistically ins ignificant and with positive si g n (see Table 5 1), which can be credited to the characteristics of the data used. Because the null hypothesis of no relationship between product demand and promotions cannot be rejected for any of these countries, theoretically their product demands would be unaffected by changes in the l evel of pro motions. Given the lack of statistical significance simulation s for beef promotions will be not discussed since they provide unreliable results. Simulations for the parameters in Table 5 1 are considered in the following sections of the chapter. A compar ison will be made to their mean values as relative prices and quantities

PAGE 128

128 demanded change ac ross time and country, and the corresponding impacts of these variables are graphically presented for their analysis. It is important to mention that some variables may be statistically significant but numerically unimportant when predicting probabilities. Total Product Demand Trend Estimated trade responses across time for the United States, Australia, and New Zealand are pres ented in F igures 6 1 through 6 6 in term s of beef product demands in each of the importing countries across time. Product demands were estimated using the methodology introduced in Chapter 4, which includes relative prices, domestic demands for beef in general, and the CRES parameters The reper cussion of the BSE announcements and the rapid increase of diagnoses across the European Union (EU) during the 1997 to 1998 period had an important impact on beef trade around the world. In particular, in the case of feed lot beef producing countries such as the United Stat es, the BSE scare in Europe has had an important and negative effect o n their beef exports. That is, F igures 6 1 and 6 2 show a significant decrease i n the demand for this type of beef in all Asian countries selected for this study. Estim ates for the 1997 to 1998 period show that product demands for U.S. fresh beef in South Korea declined more than 34%, while in Japan, Taiwan, and Hong Kong the reduction was around 17%. With respect to frozen beef and as expected, the South Korean market s howed the largest negative impact (29%) while in the other markets the impacts were about 16%. After a considerable p eriod of stability, the 2003 BSE announcements in North America had a massive negative impact on U.S. beef products to the point that duri ng the 2004 to 2006 period every U.S. export market of any significance closed their border to U.S. beef and cattle (USMEF 2007). As shown in Figure 6 1 U.S. product demands in all four market s fell rapidly to virtually zero as the result of a complete ba n on U.S. beef exports based on animal age restrictions and the

PAGE 129

129 type of beef (see Table 1 2 for a full description of the restrictions). However, during the year 2006, a gradual re opening of trade was translated in to an increased product demand for U.S. b eef products across countries, in particular in Japan where consumers have shown a mar ked preference for imported, g rain feed, beef from the United States. Figure 6 1 Estimated f resh U.S beef product demands across time in se lected Asian markets. As of September 2007, the increase in U.S. product demand represented 32% of the pre BSE levels ; and market prediction by the USDA, Foreign Agricultural Service and the USMEF show ed a positive trend in the volume of beef exported as well as in the value of export to Japan for the incoming years. In the case of South Korea, the situation is far more complex and driven by policy rather than marke t demand. As previously mentioned South Korea is the largest importer/consumer of beef in the region and even though th e complete ban on U.S. beef has been lifted by the

PAGE 130

130 government, the strong opposition among politician and radical consumer groups has affected the normal flow of U.S. beef to this country. As F igure 6 1 illustrates, the 2007 pr oduct demand levels are only 8 % of the pre BSE level for both fresh and frozen beef products. Figure 6 2 Estimated f rozen U.S beef product demands across time in selected Asian markets. Despite the difference in the quantity of U.S. beef products consumed, the markets in Taiwan and Hong Kong show a quite similar behavior in terms of the results obtained in this study. In both cases, the estimates for these markets show the same reactions to the BSE announcements, showing a di fference of 67.3% in product demands between the pre and post BSE announcements. Considering that these two countries share the same culture these results are not surprising and are much closer to what was expected not only in the case of the United Stat es but also with respect to product s from Australia and New Zealand.

PAGE 131

131 Figures 6 3 through 6 6 show the other side of the BSE impact, as the demand for U.S. beef products disappeared the demand for beef originated in Australia and New Zealand shows an upwa rd tr end during the same ten year period. Figure 6 3 Estimated f resh Australian beef product demands across time in selected Asian markets Since 1998, South Korea has the largest market for Australian beef in the region, show ing a 9 fold increase with respect to1997, as exports to the other countries of the region also increased more than 1.5 fold during the same period. During the next years and before the 2003 BSE scare in North America, Australian product demand estimates s howed some upwards and downwards trends, in particular during the 2000 economi c crisis in Asia. As expected, Figures 6 3 and 6 4 illustrate that the BSE announcements had a positive and significant impact on Australian beef demand in th e region. Only in S outh Korea, the BSE scare

PAGE 132

132 repr esented an increase of more than 200% and in Japan, it corresponded to an almost 60% increment in fresh beef product demand during the period 2003 to 2004. Figure 6 4 Estimated f rozen Australian beef product demands across time in selected Asian markets. In the case of frozen beef products Figure 6 4 shows that Japan is the larger dest ination for Australian products; however estimates for the South Korean market show that during the ten year peri od of this study product demand for frozen beef has increased almost 106% while in the Japanese market that increase was about 34%. The difference between pre and post BSE announcements in both South Korea and Japan markets is remarkable; in the case of Australian frozen beef there was an increment of 140% and 52% in the South Korean and Japanese markets respectively, while in the case of Taiwan and Hong Kong that growth was about 50%.

PAGE 133

133 Figure 6 5 Estimated f resh New Zealand beef product demands across time in selected Asian markets. The demand for beef products that originated in New Zealand increased, after the 2003 BSE a ppearance in North America. As Figures 6 5 and 6 6 show, the strong pre BSE presence of U.S. beef in the marke ts of the region was reflected i n low demand levels for New Zealand beef products between 1998 and 2002. In particular, the South Korean demand for beef products that originated in this country reached its lowest levels since 2002 when fresh and fro zen product demands bar ely passed one M.MT. However duri ng the 2003 to 2004 period, F igure 6 5 illustrates that in Japan, Taiwan and Hong Kong the demand for New Zealand beef products almost doubled while in South Korea it increased a lmost four times duri ng the one year period.

PAGE 134

134 Figure 6 6 Estimated f rozen New Zealand beef product demands across time in selected Asian markets. In all figures, it is very clear that as the United States started to re gain market access in the r egion during the year 2007, while Australian and New Zealand product demands started to decline in all four markets. According to the estimates of the model, the demand for U.S. beef products has its largest increase in Japan, 33% (fresh), and 32% (frozen) while it s lowest increase was in South Korea with only 8% (fresh) and 8% (frozen) with respect to the 2003 levels. M eanwhile, New Zealand product demands showed the largest decline in 2007 with a range between 37% (South Korea) and 17% (all other markets ) in the case of fresh products and 32% and 17% in the case of frozen products. Australia, however, did not show such a large decline during 2007 and in both cases the decline was between 8% (South Korea) and 3.5% (all other markets).

PAGE 135

135 The most appealing e xplanation for this change is the size of the players in this particular market. The United States is one of the largest players in the global beef market and, at any point, its presence or absence in a particular market will affect the dynamics of the mar ket. Beef consumers in all four markets have shown preference toward American beef products, in promotion campaigns should have reinforced the quality and safety attributes of all type s of beef products that originated in the Unites States by increasing the information and knowledge levels of consumers. Product Demand Simulations To analyze the effects of changes in the relative price of beef on product demands b efore a nd after the BSE announcements, beef prices from all three suppliers w ere allowed to shift between 10 % below and above the mean price, of each exporting country holding all other beef prices at the mean. The simulation results for pr oduct demands a r e presented in F igures 6 7 through 6 10 and include the total amount of beef products demanded in each beef market during the years 1998, 2002, and 2006, faced with a change in the prices of each beef supplier. As demand is a function of the relative size of its market demand an d of its product prices ratios, assuming that is exogenous. The corresponding price elasticities are included to illustrate how changes in prices lead to changes in the quantity of product demanded. Note that for all markets the price ratio of beef products relative to the average (mean) across the four importing markets were $1.34 per Metric Ton (MT) for U.S. beef, $0.94 per MT for Australian beef, and $1.02 per MT for New Zealand beef. With the exception of U.S. prices, quantities, and elasticities for the post BSE year of 2006, in general the figures below show that as the price of beef from one exporter goes up, the quantity of beef products demanded goes

PAGE 136

136 down, which is expected for beef products in Asian market s However, there is a difference between the estimated elasticities for the South Korean beef market and the elasticities for the other beef markets in the way of the type of the elasticity. That is, the price elasticities for beef products in Japan, Taiwan, and Hong Kong are classified as almost unit elastic, meaning that as the price increases the quantity of beef product demanded decreases in the same proportion, while in the case of South Korea the price elasticities ar e classified as elastic or in other words, an increase in beef price is met by a more than proportionate quantity decrease (Nicholson 2002). In the next sections of this chapter a detailed discussion of each of the beef markets in terms of price and quanti ty changes, and the price elasticity of demand will be presented. Figure 6 7 South Korea beef market demands and price elasticities in 1998, 2002, and 2006 Figure 6 7 includes the distribution of beef product demands among e ach beef exporter country to the South Korean market, faced with a change in prices during the years 1998, 2002,

PAGE 137

137 and 2008. The figure shows the impact of the complete ban on U.S. beef products that is translated into a large and positive difference in Aust ralian product demands between the pre BSE years and the year 2006. This difference represents more than 150% incr ease in product demands. I n the case of New Zealand there w as a decline of 3% in product demand between 1998 and 2006 As the relative prices of exported beef increase from 10 % below the mean or average price to 10 % above the mean prices there was an identical decrease across time in the quantity of beef product demanded of 28 .8 % in the case of U.S. beef, 41 .9 % in the case of Australian beef, and 38 .9 % in the case of New Zealand beef. Figure 6 8 Japan beef market demands and price elasticities in 1998, 2002, and 2006. The product demand responses to variation in the mean prices are more extreme in the case of Aus tralian beef, where in all cases a 10% reduction in the price of beef o riginated in this

PAGE 138

138 country led to an increase of almost 25 % in the quantity demanded. These results are anticipated as the estimated relative price elasticities have values greater than two, as in the case of elastic price elasticity of demands. The Japanese market was the second destination in terms of market size for beef products that originated in any of the exporting beef countries. In relative terms, this market represents 76% of th e South Korean market when compared to the total product demands across time. The United States was clearly the major player in this market until the BSE announcements in 2003. Figure 6 8 shows, once again, how the impact of the complete ban on U.S. beef p roducts affected the scenario of the Japanese beef market. While U.S. beef disappeared from the market in 2006, Australian and New Zealand beef products expanded their presence one and a half times more and almost two times more than in the pre BSE years. In Japan, p roduct demand variations with respect to changes in the relative price show a behavior similar to the South Korean market. That is, as the price increase d the quantity of beef products demanded declined but in this case in almost identical prop ortional magnitudes. Therefore, as beef prices increase from 10 % below the mean or average price to 10 % above the mean prices there is an identical decrease across time in the qu antity of beef product demanded, of 14% in the case of U.S. beef, 19% in the c ase of Australian beef, and 18% in the case of New Zealand beef. Since price elasticity estimates show values very close to one, these results are plausible as they indicate a unit elastic price elasticity of demand. With the exception of the year 2006, si milar unit elastic demand curves were obtained for all three beef exporting countries across time. Figure s 6 9 and 6 10 show the differences in market size between the beef markets in Taiwan and Hong Kong and the previous markets. In terms of product dema nds, the combined sizes of these two markets represent less than one h alf of the Japanese market and one third of

PAGE 139

139 the South Korean Market. That is, the amount of products demanded during the three years considered was 90,000 MT in Taiwan, while in Hong Kon g this amount was about 70,000 MT. I n addition, relative price elasticity estimates for Taiwan and Hong Kong show unit elastic values, thus as beef prices increase from 10 % below the mean or average price to 10 % above the mean prices there is an identic al decrease across time in the quantity of beef product demanded of 14% in the case of U.S. beef, 19% in the case of Australian beef, and 18% in the case of New Zealand beef. Figure 6 9 Taiwan beef market demands and price elasticities in 1998, 2002, and 2006. In summary, these simulation results correspond to the descriptive statistics which were shown in Chapter 2 to be the largest import and export markets. Results show that product demand responses to variation in the av erage market prices were more extreme after the 2003 BSE announcements, in particular in the larger markets, such as South Korea and Japan.

PAGE 140

140 Considering product demand levels for beef products that originated in Australia and New Zealand between 2002 and 20 06, the figures above clearly illustrate this trend. In all these scenarios it is clear that the absence of U.S. beef product s has helped the expansion of Australian products elsewhere, in particular in South Korea were consumers have set a strong position against beef imports from North America. Figure 6 10 Hong Kong beef market demands and price elasticities in 1998, 2002, and 2006. This set of scenarios yields useful information to policy makers and beef in dustry representa tives attempting to understand the impact of restrictive trade policies affecting the normal flow of agricultural commodities and that are based on arbit rary food safety concerns as expressed in several occasions by World Organisation for Animal Health (OI E) 1 1 Based on the Terrestrial Animal Health Code this international regulatory agency specifies all health measures to be used by the veterinary authorities of importing and exporting countries to avoid the transfer of agents pathogenic for animals or humans, while avoiding unjustified sanitary barriers The OIE officially announced the U.S. as a

PAGE 141

141 Pre and Post BSE Market Distribution Simulations In this section, market demands are analyzed according to the percentage of participation of each of the three exporters in the four importing markets. In order to understand better the impact of the B SE announcements in 2003, the following figures will present pre and post BSE results for fresh and frozen beef products. In F igure s 6 11 and 6 12, the reader can easily see the results of the simulations and fully compare the impact of the BSE on the rel ative market participation during the pre BSE period (1997 to 2003) and after the BSE announcements (2004 to 2007). Figure 6 11 Pre BSE market distribution of fresh and frozen beef products. Us ing relative demand adjustments across time, beef market shares were simulated from 20% below the mean or average quantity demanded from each beef exporter to 20% above the mean quantity. In order to illustrate the variations across countries, Table 6 5 pr esents the

PAGE 142

142 corresponding market shares values above, at, and below the mean, while F igures 6 11 and 6 12 illustrate the values at the mean levels. Table 6 5. Estimated variations in market demand distribution United States A ustralia New Zealand PRE POST PRE POST PRE POST Fresh Frozen Fresh Frozen Fresh Frozen Fresh Frozen Fresh Frozen Fresh Frozen S.K. +20 0.697 0.273 0.001 0.001 0.432 0.175 0.883 0.461 0.031 0.019 0.086 0.046 0 0.714 0.269 0 .001 0.001 0.456 0.176 0.895 0.463 0.032 0.020 0.091 0.046 20 0.725 0.266 0.001 0.001 0.471 0.453 0.916 0.420 0.033 0.061 0.093 0.042 JP +20 0.202 0.143 0.357 0.001 0.155 0.110 0.246 0.174 0.054 0.038 0.082 0.058 0 0.201 0.143 0.445 0 .001 0.155 0.110 0.245 0.174 0.054 0.038 0.082 0.058 20 0.201 0.049 0.511 0.047 0.154 0.174 0.245 0.168 0.054 0.067 0.081 0.056 TW +20 0.449 0.321 0.089 0.003 0.345 0.246 0.546 0.390 0.120 0.086 0.182 0.130 0 0.449 0.321 0.112 0.003 0.34 5 0.246 0.546 0.389 0.120 0.086 0.182 0.130 20 0.449 0.108 0.128 0.105 0.345 0.388 0.546 0.374 0.120 0.149 0.182 0.124 H.K. +20 0.449 0.321 0.070 0.003 0.345 0.246 0.546 0.390 0.120 0.086 0.182 0.130 0 0.449 0.321 0.088 0.003 0.345 0.246 0.546 0.390 0.120 0.086 0.182 0.130 20 0.449 0.109 0.101 0.105 0.345 0.388 0.546 0.375 0.120 0.150 0.182 0.124 L ooking at the simulation results for the United States before 2003, it is clear that the U nited States was the leader in perc entage of product demanded in all four importing markets and that the share of fresh beef was larger than the share for frozen products. That is, in South Korea the difference between U.S. beef and products originated in Australia was 25.8% and 9.3% for fr esh and frozen products respectively; in Japan the difference in product demand shares was 4.6% and 3.3 % and in comparison to New Zealand products the difference between market shares was 68.2% for fresh and 24.9% frozen; and in Taiwan and Hong Kong this difference was 10.4% and 7.5% in each market. Figure 6 11 illustrates that in 2003 New Zealand beef, in all markets, has the smallest shares in terms of product demands representing a combined share (fresh and frozen products) of only 5.2% in South Korea and 9.2% in Japan, while in Taiwan and Hong Kong the combined share is 20.5% in each market.

PAGE 143

143 Changing the perspective of the analysis, results show that there was a significant difference between the types of product demanded. In the case of U.S. product shares, the difference between fresh and frozen beef were 44.5% in South Korea, 5.8% in Japan, 12.8% in Taiwan, and 12.8% in Hong Kong. These numbers show how markets differ in term of preferences towards U.S. beef products, in particular when comparing th e South Korea and the Japanese market. In the first market, fresh U.S. beef product demanded are preferred, while in Japan the figure shows that there is almost no difference in preferences between fresh and frozen U.S. beef products. With respect to produ cts originated in Australia, in South Korea this difference is 28.1% in favor of fresh products, which reinforce s the idea that consumers in this market have a marked preference for imported fresh products. Figure 6 12 Post BSE market distribution of fresh and frozen beef products. After the 2003 BSE scare the beef market changed radically for all beef exporters. Figure 6 12 shows that the complete ban on beef originated in North America eliminat ed all U.S.

PAGE 144

144 market shares in the region. In all four markets, the BSE scare represented a total U.S. beef export value loss of $2.4 billion during the 2003 to 2004 period (USMEF 2007). Since then, Australia has dominated the fresh and frozen beef market of the region, while New Zealand has increased it s presence in all beef markets o n a much smaller scale. For example, in South Korea, Australia n fresh beef products represent 89.5% of the shares, while New Zealand is second with only 8.1% of the product dema nd shares. With respect to frozen products, the shares for Australian beef are about 47% and 5% in the case of New Zealand. In relative terms, these numbers represent an increase in product demands of 49% (fresh) and 61% (frozen) for Australia; and 64% (fr esh) and 57% (frozen) in the case of New Zealand products. In the Japanese beef market, Australian shares are quite less compared to the South Korea market showing only 24.5% of the total fresh market and 17.4% of the tot al frozen market. Comparing pre a n d post BSE shares, this number represent s a relative increase of 36% i n product demand shares in both the fresh and frozen market. In the case of New Zealand, the results show market share levels of 8.1% (fresh) and 5.8% (frozen), which represent a slight increase with respect to pre BSE product demand levels. Less expensive Australian and New Zealand beef has helped these two countries to compete in the Japanese market but strong consumer preferences towards grain feed production type of beef have helped beef products coming from other parts of the world (i.e., South America) to establish a significant presence in Japan. In addition, reports from the United States Department of Agriculture have shown a slow but positive increase in the demand for U.S. beef products since 2007, which has become a main focus of interest for U.S. producers and exporters due to the fact that Japan was the most important market for U.S. bee f products in general before December 2003.

PAGE 145

145 In Taiwan and Hong Kong, the post BSE markets show similar shares for Australian beef, 54.5% for fresh products and 38.5% for frozen products, which represent an increase with respect to pre BSE levels of 20.1% and 14.3% respectively. In the case of New Zealand beef, Taiwan and Hong Kong also show si milar market results. That is, New Zealand products represent 18.1% of shares for fresh beef and 13.1% of the shares for frozen beef in these two markets, which correspond to 6.2% and 4.4% increase in fresh and frozen market shares with respect to pre BSE levels. This chapter concentrated o n a more detailed interpretation of the coefficients estimates presented in Chapter 5. Simulations were conducted for each of the four importing beef markets on the functional relationship presented in Chapter 4, except for simulations representing the Rest of the World. That is, the pa rameters estimated with the non linear procedure were used to simulate the reactions of the dependent variable to changes in the explanatory variables, holding all other variable constant ( ceteris paribus ) Three types of simulations were carried out on the product demand equations. First, the product demands were analyzed across time and the effect of the BSE announcements measured; second, the average prices were varied from 10 % below the mean or average price to 10% above the mean prices (1997); and third, product demand shares were simulated when the average quantity demanded varied from 20% below the mean or average price to 20% above the mean prices (1997). The simulation results for t he beef trade model proposed in Chapter 4 make economic sense and yield insights into the trends observed in Chapter 2. This speak s very well of the model; it captures the real world trends occurring in beef trade during the 1997 to 2007 period.

PAGE 146

146 CHAPTER 7 SUMMARY AND CONCLUSI ONS This dissertation focuses on the economic impact of food safety issues, in particular the Bovine Spongiform Encephalopathy (BSE), on four of the most important international beef markets and its repercussion on U.S. beef exports. P rivate and governmental efforts to re gain market access and promote the attributes of U.S. beef products in terms of quality and safety have captured the attention of this study looking at the efficacy of these campaigns. This chapter serves as a brief su mmary of the findings of this research, where conclusion and implications are discussed. To review briefly the dissertation, Chapter s 1 and 2 examine the international beef market and discuss the importance of agricultural trade and beef promotions The li terature review is presented in Chapter 3, where previous research on beef demand, food safety, trade, and promotions are discussed, as well as previous applications of the Armington model The theoretical basis of the model is presented and developed in C hapter 4, where the specific CRES functional representation, along with a non linear trade model, is defined to measure the variations on product demand across time. In Chapter 5 the results of the econometric estimation are presented and discussed. Figur es of the actual data observed along w ith tables with the parameter coefficients predicted by the model and the statistics indicating the performance of the model are included. Chapter 6 presents the simulation results and their implications for beef trade Simulations were carried out on product demand and market share functional rel ationships for each of the beef importing markets. All product demands calculated using the estimated parameters from Chapter 5 and pecifications capture the general direction taken by the actual data presented in Chapter 2. In addition to that, Chapter 5 and 6 include a number of

PAGE 147

147 figures that plot predicted parameter values and the corresponding simulations showing that the model capt ures the major trends and turning points in the trade of beef products in the Pacific Rim region. in which beef products are distinguished by place of production. This theoretical s pecification assumes that beef products produced in different countries are considered not homogeneous products and therefore perfect substitutes, implying that thei r elasticity of substitution is not infinite. The CES function introduced by Armington in t he derivation of the p roduct demand equation, assumes that products competing in a market have constant and equ al elasticities of substitution. This research, however, approaches the elasticity of substitution in a more flexible way The degree of substitu tability between domestic and imported sources of supply for beef product s is captured by the CRES. The resulting functional relationships impose that all products competing in a market have elasticities of substitution that vary by a constant ratio, allow ing products differentiated by their place of production to have varying elasticities of substitution among products within a category. Several types of data sets were needed in order to estimate the beef trade model and a number of difficulties were enco untered in obtaining reliable information, in particular, on international beef promotions in South Korea, Japan, Taiwan, and Hong Kong due to the confidential character of the information. In order to include this information in this study, several source s of data were consulted, including official foreign government Web sites, meetings with USDA representatives from the Foreign Agricultural Service, and consultations with members of the U.S. beef board. The obtained information was used to create a global data

PAGE 148

148 set on promotions expenditures, which turn out to be one the most important limitation s in this study due to the lack of applicability at the country level. Using a combination of logistic and exponential functions the and parameters were obtained and applied to the product demand functional form specified in Chapter 4 The resulting econometric model represents a non linear system that shows e stimates and supportive statistics indi cating that this trade system fits wel l the trade data ( R square = 0.763) and explain the economics behind the international beef trade in the region. The resulting Durbin Watson statistic values indicate little serial correlation among the parameters of the model. With the exception of the ec estimates resemble the economic trends of the beef trade in the region. price and the size o f the beef market in a particular country, any changes in market demands also affect product demands. Thus, for each market, two set of simulations were conducted on the product demand s where the trade volumes from each of the exporting countries was simu lated over time (1997 to 2007) relative prices were varied from 10% below to 10% above of their base or mean price ( US: $1.34 per MT; AU: $0.94 per MT; and NZ: $1.02 per MT) and market demands were varied from 20% below to 20% above of their base or aver age quantity demanded in each market (KR: 99.58 MT; JP: 286.55 MT; TW: 32.1 9 MT; HK: 25.26 MT). All product demands show a negative slope in terms of price variations, that is as the average or mean price increases, the quantity of beef products demanded d ecreases. For the simulations where the average price was varied, the product demands for South Korea, Japan, Taiwan, and Hong Kong reflect a negative relationship between price and quantity demanded as expected by the econo mic theory.

PAGE 149

149 Considering the pre and post BSE periods, the results show an extreme change in market composition in particular when considering the c ase of Australia that doubles its market shares between 2002 and 2006. In general, price elasticities of the product demands are such that t hose for the South Korean market are more elastic than those for Japan, Taiwan, and Hong Kong. These results clearly indicate the volatility of the South Korean market, while the rest of the markets are likely to remain relatively constant. In the case of beef promotion s, the simulation results show little or no effects on product demands. Contrarily to what was expected, this research could not find a link between pro motions and increasing levels of the demand for beef products. A s promotion expenditures i ncreased or decreased from th e mean products demands and market shares show very small changes indicating that the po sitive parameter coefficient is no different than zero Despite marketing efforts aimed to promote and expand the presence of U.S. beef pr oducts in the region, the negative demand effects of food safety concerns as measured by the embedded BSE effect across time trend has been the single factor driving all beef import demands in these four countries giving little room for an effective and s ignificant impact of the commodity promotion programs aimed to compensate the inward shift in U.S. beef demand induced by both BSE information and longer term consumer preference changes. Despite the negative press and the reduced presence in Japan, Taiwa n and Hong Kong, U.S. beef is considered by consumers in these countries to be the high quality product while Australian beef is considered to be the low quality product. However, with food safety being one of the main drivers of the beef import demand, a high quality characteristic may not be enough to overcome the feelings towards the production process and the potential of food borne diseases that are related to intensive grain feed produced beef.

PAGE 150

150 Future research will require more detailed data on bee f promotion expenditures including data from competing countries such as Australia and New Zealand in order to fully analyze the relation between promotions and beef demand. In addition, the own nature of an international commodity promotion program should be re considered. Since the basic idea of such programs is to expand the total demand for the commodity and not the demand for products from a particular the problem of free riders. These countries get the benefits of increasing beef product demands without an active participation in the promotion of their products. That could be one of the explanations for the lack of significance of the promotion parameters in this research. Future research must be able to disaggregate total international expenditures on beef promotions to fully address their impact in a particular market. Previous research on international commodity promotions conducted by Kinnucan and Zhen g (2004) and Piggot t et al. (1996) suggested that market demands are insensitive to changes in advertising expenditure, and that unless the beef exporter country has a surprising degree of market power in a particular international market, the promotional campaigns may not have been profitable, which clearly reflect the promotion results in this research, in particular in the case of U.S. beef exports after the BSE announcements. In all four markets considered in this study c urrent market restrictions r equ ire that all imported beef be produced from animals between 30 and 20 months of age or younger at the time of slaughter and the ability of U.S. beef exporters to provide specifi c beef cuts could be used as a m arketing tool by turning around these restricti ons into competitive advantages. The highly integrated and efficient U.S. beef supply chain can easily satisfy the preferences of these consumers in terms of age and type of beef cut in particular in sophisticated markets such as

PAGE 151

151 Japan and South Korea. Imp orted beef from the United States has a competitive advantage in these beef markets as well as in the rest of the markets of the region, because U.S. beef has generally been viewed by consumers as being a higher quality product, in terms of taste and fat c ontent, than beef imported from other sources. In conclusion, as illustrate d in Figures 6 11 and 6 12 the BSE announcement in December 2003 has definitively alter ed the landscape and patterns of international trade in beef products in the Pacific Rim reg ion. The reactions of governments through a total ban on U.S. beef products increased and shifted the demand for grass feed produced beef from Australia and New Zealand as feedlot produced beef from the United States disappeared from the markets. T he compl ete ban on U.S. beef products represented a total economic loss of about $ 2.4 billion between the periods 2003 to 2004 According to USDA estimat es, the U.S. might be able to reach pre BSE levels in terms of value of beef exports to the region but, it wil l never reach pre BSE levels in term of export volume due to the significant reduction in market shares with respect to other beef exporters, in particular Australia. Based on the results obtained in this research, the relevance of promotions among the U. S. beef products can be evaluated from the stand point of the perceived importance that consumers even added values such as specific type of beef cuts. The perceived characteristics of the product imply some degree of security and reliability which could potentially enhance the total demand for that particular product. A major challenge for U.S. beef exporters is to satisfy the increasing demand for more con venient products, as in the case of beef consumers in Hong Kong, and value added cuts consistent with the well defined preferences in markets such as Japan and South Korea.

PAGE 152

152 APPENDIX A CO NSTANT ELASTICITY OF SUBSTITUTION (CES) The Constant Elasticity of S ubstitution (CES) function was first introduced by Arrow, Chenery, Minhas, and Solow in 1961 using a two factor (capital and labor) production function specification 1 As its name suggests, the CES describe a particular output level given a constant combin ation of inputs, which allows for greater flexibility since the value of the elasticity of substitution can be chosen. Varian (1992) defined the CES production function as a general case that comprises several other well known production functions as speci al cases, depending of the value of the parameter (e.g., Linear production function = 1 Perfect substitutes, Cobb Douglas production function and Leontief production function = Perfect complements ). Thus, the higher the value of the elasticity, the closer is the degree of substitution and vice versa. In consumer theory, the same functional form can be found acting as an aggregator function combining two or more types of consumption into an aggregate quantity and showing constant e laticities of substitution. In 1969, Paul Armington used this aggregator function to explain trade among countries and suggested a model that evaluates changes in trade policy by 1 This paper derives the CE S as following: which is assumed to be then we can say that thus, Let the marginal product of labor (MPL) be and since a monotonic relation between y and w is observed we have : Differentiating this last term we obtain a differential equation (dfe): Assuming a linear relationship between the logarithms of and w, (i.e., ) the dfe becomes: After taking the antilog, solving for and integrating the partial fractions we obtain: or where we set and for convenience. To be precise, represents the Elasticity of Substitution (ES). Solving this last expression for and then y we have: Setting and we finally obtain a mathematical expression that defines all functions exhibiting a CES for all values of K/L: or written out in Production function form, where is an efficiency or shift parameter, is a distribution or share parameter, is the substitution parameter or function exponent, which permits the parameter of elasticity to range from

PAGE 153

153 allowing the conversion of these changes into price effects. The Armington sp ecification assumes that imports originating from different sources are imperfect substitutes of each other, that the demand functions show constant elasticities of substitution (CES), and the elasticity of substitution between any two products competing i n a market is the same as that between any other pair of products competing in the same market. As in Preckel, Cranfield, and Hertel (2005) and Fmnia and Gohin (2007), the most convenient way of representing the explicit form of the Armington CES produc t demand function is: (A 1) where u denotes utility and denotes a vector of differentiated goods by their origin, and and are the distribution and substitution parame ters of the CES utility function. In order to simplify the interpretation, Armington (1969) "collapsed" the Utility function assuming that any given quantity index function, 2 has the generalized CES form as previously defined 3 Suppressing the argum ents of u and rearranging Equation A 1 leads to the following implicit expression of the same relationship: A 2) Since the utility is an ordinal measure, t here is no consequence to normalize the parameters by imposing their sum equal to one ( ). Thus, based on the theory of the utility maximizing consumer, where any given quantity of a good is to be obtained at least money cost, we know that the marginal rate of substitution (MRS) between two goods must equal the ratios of their prices (Varian 1992): 2 is assumed to be linear and homogeneous, thus all demands for products are the same (Armington 1969). 3 The CES technical relationship is defined as:

PAGE 154

154 i =1, 2, (A 3) Solving Equation A 3 for (A 4) Using this function, Equation A 2 can be expressed as a relation between and and their cor responding prices. Rearranging terms, (A 5) from which it follows that the elasticity of substitution between and any other product competin g in the same market is equal to the constant which denotes the elasticity of substitution (Varian 1992) 4 Thus, Equation A 4 can be expressed as follows: n (A 6) where defines the degree of substitution among products. For example, at the limiting cases if the products are perfect complements or if the products are perfect substitutes (Armingto n 1969). Substi tuting Equation A 2 in to Equation A 6 and writing in terms of we have: (A 7) 4 The elasticity of substitution (ES) between and is:

PAGE 155

155 Then, solving Equation A 7 for (A 8) Using Equations A 2 and A 3 and rearranging terms we have, (A 9) Th en, substituting from Equation A 7 (A 10) Therefore, (A 11) Finally, substituting Equation A 11 into Equation A 8 the Armington product demand function is obtained assuming a Constant Elasticity of Substitution in each market : (A 12) As A rmington (1969) suggested, the Equation A 12 can be used for different purposes; for example written as or (A 13) (A 14) where market demand is expressed as the dependent varia ble. As is clear from Equation A 14 value shares are constant if (Armington 1969). Taking logari thm of Equation A 14 we have:

PAGE 156

156 (A 15) where denotes the market share of imports from source j No expenditure term in the right hand side in Equation A 15 imp expenditure does not affect the market share (Yang and Koo 1993). As a result, all expenditure elasticities within a group are equal and unitary and import market shares change only in respon se to relative price changes.

PAGE 157

157 APPENDIX B TSP PROGRAM: ARMINGTON SPECIFICATION AND SIMMULATIONS OPTIONS MEMORY=200 RESID; TITLE 'OSCAR FERRARA BEEF TRADE DATA IN THE ASIA MARKETS'; ARM_SIM#6.TSP THIS NEW DATABASE NOW HAS ALL THE DATA STACKED BY WIMP,WEXP ,WTYPE; IN 'C: \ ZSTUDENT \ OFERRARA \ TSPPRG \ BF_ARM#3'; IN 'C: \ ZSTUDENT \ OFERRARA \ TSPPRG \ BF_ARM#3'; DBLIST 'C: \ ZSTUDENT \ OFERRARA \ TSPPRG \ BF_ARM#3'; OC(REPLACE) YEAR 'YEARS'; OC(REPLACE) CNTY 'COUNTRY 1=KOREA, 2 =JAPAN, 3=TAIWAN, 4=HONG KONG'; OC(REPLACE) POP 'POPULATIONS IN MILLIONS'; OC(REPLACE) GDP 'GROSS DOMESTIC PRODUCT %'; OC(REPLACE) CPI 'CONSUMER PRICE INDEX'; OC(REPLACE) TotImp 'TOTAL IMPORTS INTO COUNTRY'; OC(REPLACE) FoodImp 'TOT AL FOOD IMPORTS INTO COUNTRY'; OC(REPLACE) BeefProd 'BEEF PRODUCTION IN COUNTRY'; OC(REPLACE) BeefCons 'BEEF CONSUMPTION IN COUNTRY'; OC(REPLACE) V1_WL 'VALUE TO 1 FROM WORLD ($1000) ALL FRESH BEEF'; OC(REPLACE) Q1_WL 'QUANTITY TO 1 FROM WORLD (KG) ALL FRESH BEEF'; OC(REPLACE) V1_US 'VALUE TO 1 FROM UNITED STATES ($1000) ALL FRESH BEEF'; OC(REPLACE) Q1_US 'QUANTITY TO 1 FROM UNITED STATES (KG) ALL FRESH BEEF'; OC(REPLACE) V1_AU 'VALUE TO 1 FROM AUSTRALIA ($1000) AL L FRESH BEEF'; OC(REPLACE) Q1_AU 'QUANTITY TO 1 FROM AUSTRALIA (KG) ALL FRESH BEEF'; OC(REPLACE) V1_NZ 'VALUE TO 1 FROM NEW ZEALAND ($1000) ALL FRESH BEEF'; OC(REPLACE) Q1_NZ 'QUANTITY TO 1 FROM NEW ZEALAND (KG) ALL FRESH BEEF'; OC(REPLAC E) V2_WL 'VALUE TO 2 FROM WORLD ($1000) ALL FROZEN BEEF'; OC(REPLACE) Q2_WL 'QUANTITY TO 2 FROM WORLD (KG) ALL FROZEN BEEF'; OC(REPLACE) V2_US 'VALUE TO 2 FROM UNITED STATES ($1000) ALL FROZEN BEEF'; OC(REPLACE) Q2_US 'QUANTITY TO 2 FROM UNITED STATES (KG) ALL FROZEN BEEF'; OC(REPLACE) V2_AU 'VALUE TO 2 FROM AUSTRALIA ($1000) ALL FROZEN BEEF'; OC(REPLACE) Q2_AU 'QUANTITY TO 2 FROM AUSTRALIA (KG) ALL FROZEN BEEF'; OC(REPLACE) V2_NZ 'VALUE TO 2 FROM NEW ZEALAND ($ 1000) ALL FROZEN BEEF'; OC(REPLACE) Q2_NZ 'QUANTITY TO 2 FROM NEW ZEALAND (KG) ALL FROZEN BEEF'; OC(REPLACE) V2_ROW 'VALUE TO 2 FROM OTHER EXPORTERS ($1000) ALL FROZEN BEEF'; OC(REPLACE) Q2_ROW 'QUANTITY TO 2 FROM OTHER EXPORTERS (KG) A LL FROZEN BEEF'; OC(REPLACE) V1_ROW 'VALUE TO 1 REST OF THE WORLD ($1000) ALL FRESH BEEF'; OC(REPLACE) Q1_ROW 'QUANTITY TO 1 REST OF THE WORLD(KG) ALL FRESH BEEF'; OC(REPLACE) P1_WL 'PRICE TO 1 FROM WORLD ($1000) ALL FRESH BEEF'; OC(REPLAC E) P1_US 'PRICE TO 1 FROM UNITED STATES ($1000) ALL FRESH BEEF'; OC(REPLACE) P1_AU 'PRICE TO 1 FROM AUSTRALIA ($1000) ALL FRESH BEEF'; OC(REPLACE) P1_NZ 'PRICE TO 1 FROM NEW ZEALAND ($1000) ALL FRESH BEEF'; OC(REPLACE) P2_WL 'PR ICE TO 2 FROM WORLD ($1000) ALL FROZEN BEEF'; OC(REPLACE) P2_US 'PRICE TO 2 FROM UNITED STATES ($1000) ALL FROZEN BEEF'; OC(REPLACE) P2_AU 'PRICE TO 2 FROM AUSTRALIA ($1000) ALL FROZEN BEEF'; OC(REPLACE) P2_NZ 'PRICE TO 2 FROM NE W ZEALAND ($1000) ALL FROZEN BEEF'; OC(REPLACE) P2_ROW 'PRICE TO 2 FROM OTHER EXPORTERS ($1000) ALL FROZEN BEEF'; OC(REPLACE) P1_ROW 'PRICE TO 1 REST OF THE WORLD ($1000) ALL FRESH BEEF'; PROM RESEARCH CON_INFO IND_INFO FOREIGN PRO_INFO EVAL DEVELOP PRG_EXP USDA ADMIN TOT_EXP;

PAGE 158

158 LIST WVARW WYEAR WCNTY WPOP WGDP WCPI WTotImp WFoodImp WBeefProd WBeefCons WV1_WL WP1_WL WQ1_WL WV1_US WP1_US WQ1_US WV1_AU WP1_AU WQ1_AU WV1_NZ WP1_NZ WQ1_NZ WV1_ROW WP1_ROW WQ1_ROW WV2_WL WP2_WL WQ 2_WL WV2_US WP2_WL WQ2_US WV2_AU WP2_AU WQ2_AU WV2_NZ WP2_NZ WQ2_NZ WV2_ROW WP2_ROW WQ2_ROW WPROM WRESEARCH WCON_INFO WIND_INFO WFOREIGN WPRO_INFO WEVAL WDEVELOP WPRG_EXP WUSDA WADMIN WTOT_EXP WIMP WEXP WTYPE WTT; HIST(DISCRETE) WCNTY WYEAR; ====== =============================================; CORRECTING THE PROBLEM WITH ZERO PRICES WHEN ; TRADE DID NOT OCCUR ; ===================================================; DOT(CHAR=F) 1 2; ? TYPE FRESH/FROZEN; DOT(CHAR=X) US AU NZ ROW; SELECT WQ.F_.X>0; OLSQ WP.F_.X C WP.F_WL; SELECT WQ.F_.X=0; WP.F_.X = @COEF(1) + @COEF(2)* WP.F_WL; SELECT 1; ENDDOT; ENDDOT; ===================================================; MAKING ALL MISSING QUANTITIES EQUAL ZERO ; ======== ===========================================; DOT(CHAR=F) 1 2; ? TYPE FRESH/FROZEN; DOT(CHAR=X) WL US AU NZ ROW; ? MMM=MISS(WQ.F_.X); ? SELECT MMM=1; ? MMM2=WQ.F_.X; ? WQ.F_.X=0; ? SELECT 1; WQ.F_.X = W Q.F_.X/1000; ? ALL QUANTITIES ARE NOW IN MILLIONS M ETRIC TONS (1000KG); ENDDOT;ENDDOT; =============================================================; CREATING THE MARKET SHARES Xij/Xi. & RELATIVE PRICES Pij/Pi.; =============================================================; DOT(CHAR=F) 1 2; WQ.F_ALL = W Q.F_US + WQ.F_AU + WQ.F_NZ + WQ.F_ROW ; WV.F_ALL = WV.F_US + WV.F_AU + WV.F_NZ + WV.F_ROW ; WP.F_ALL = WV.F_ALL / WQ.F_ALL; DOT(CHAR=X) US AU NZ ROW; WSH.F_.X= WQ.F_.X / WQ.F_ALL; ? EXPORTER SHARE OF COUNTRY i MARKET; WRP.F_.X= WP.F_.X / WP.F_ALL; ? EXPOR TERS PRICE RELATIVE TO THE AVERAGE; ENDDOT; ENDDOT; =============================================================; CREATING THE Q, V, P, AND SHARES IN GENERAL ACCORDING TO ; WIMP, WEXP, WTYPE ; ================ =============================================; DOT(VALUE=L,CHAR=F) 1 2; DOT(INDEX=K,CHAR=X) US AU NZ ROW;

PAGE 159

159 SELECT WEXP=K & WTYPE=L; WQALL=WQ.F_ ALL; ? THIS IS THE TOTAL QUANTITY EXPORTED TO WIMP BY TYPE; WQQ = WQ.F_. X; ? THIS IS THE QUANTITY (MIL MTONS) EX PORTED FROM WEXP TO WIMP BY TYPE; WVV = WV.F_. X; THIS IS THE VALUE ($1000) EXPORTED FROM WEXP TO WIMP BY TYPE; WSH = WSH.F_.X; ? THIS IS THE MARKET SHARE OF WEXPij EXPORTED TO WIMP BY TYPE; WRP = WRP.F_.X; ? THIS IS THE RELATIVE PRICE OF WEXPij RELATIVE TO THE AVERAGE PRICE TO WIMP BY TYPE; SELECT 1; ENDDOT; ENDDOT; DOT(VALUE=L,CHAR=F) 1 2 3 4; DOT(INDEX=K,CHAR=X) 1 2 3 4; XF = (WIMP=L & WEXP=K); MSD(WEIGHT=XF,NOPRINT) WRP; PRINT L K @MEAN; ENDDOT; ENDDOT; WFOR1 = EXP(WFOREIGN/1000); WFOR2 = EXP(WFOREIGN/10000); WFOR3 = EXP(WFOREIGN/1000000); MSD WFOR1 WFOR2 WFOR3; SET CV1 = @MEAN(1) / @STDDEV(1); SET CV2 = @MEAN(2) / @STDDEV(2); SET CV3 = @MEAN(3) / @STDDEV(3); PRINT CV1 CV2 CV3; HIST(DISCRETE) WEXP WIMP; ========================== ===================================; CREATING WIMP AND WEXP DUMMY VARIABLES ; =============================================================; DOT(VALUE=J) 1 2 3 4; DI.=(WIMP=J); ? DUMMY VARIABLES FOR THE IMPORTING COUNTRIES; DX.=(WEXP =J); ? DUMMY VARIABLES FOR THE EXPORTING COUNTRIES; ENDDOT; DOT(CHAR=I,VALUE=J) 1 2 3 4; DOT(CHAR=X,VALUE=K) 1 2 3 4; D.I.X=(WIMP=J & WEXP=K); ENDDOT; ENDDOT; DOT(VALUE=J) 1 2 ; DF.=(WTYPE=J); ? DUMMY VARIABLES FOR THE FRESH VERSUS FROZE N; ENDDOT; =============================================================; TIME SPECIFIC VARIABLES SUCH AS BSE ; =============================================================; SET YRBSE=2003; BSE=(WYEAR>=YRB SE); T1=0; T1=(WYEAR 1994)*(WYEAR<=YRBSE) + (WYEAR>YRBSE)*T1( 1); T2=BSE*(WYEAR YRBSE);

PAGE 160

160 SELECT WYEAR>1996; DUMMY(PREFIX=DYR) WYEAR; DOT DYR1 DYR11; MSD(WEIGHT=.) WYEAR; ENDDOT; DOT(CHAR=X) 1 2 3 4; DOT(CHAR=#) 1 11; DYR.#_.X=DYR.# DX.X; ENDDOT; ENDDOT; SELECT 1; AI= 1.00; AIJ=1.00; BIJ=1.00; TH0=1.00; TH1=1.00; TH2=1.00; LWQQ = LOG((WQQ+.001)/1000); LSHWQQ = LOG( (WQQ+.001)/ WQALL ); =============================================================; THE NEW ARMINGTON MODEL WITH DATA STACKED USING DUMMIES ; TO IDENTIFY WIMP AND WEXP ; =============================================================; ?FRML EQ1 LOG(WQQ + .001) [ TH0 + TH1*( LOG(WRP) ) + TH2*LOG(WQALL) ]; FRML EQ1 LWQQ = [ TH0 + TH1*( LOG(WRP) ) + TH2*LOG(WQALL/1000) ]; ?FRML EQ1 LSHWQQ = [ TH0 + TH1*( LOG(WRP) ) + (TH2 1)*LOG(WQALL/1000) ]; ? THERE CAN BE SEVERAL ALTERNATIVE SPECIFICATION FOR AI AIJ AND BIJ; ? 0 < Aij < 1.0 USING A LOGISTIC FUNCTION; IDENT EQ2 AIJ = 1 / [1+ EXP[G0 + GX1*DX1 + GX 2*DX2 + GX3*DX3 + GI1*DI1 + GF1*DF1 + GG1*(WFOREIGN/10000) ] ] ;? + GI3*DI3 + GI2*DI2; ? 0 < AI < 1.0 USING A LOGISTIC FUNCTION; IDENT EQ7 AI = 1/[ 1 + EXP[ K0 + KI1*DI1 + KI2*DI2 ]]; ? + KI3*DI3; ? Bij > 0; USING AN EXPONENTIAL FUNCTION; IDENT EQ 3 BIJ = [EXP[ [H0 + HYR2*DYR2 + HYR3*DYR3+ HYR4*DYR4 + HYR5*DYR5 + HYR6*DYR6 + HYR7*DYR7 + HYR8*DYR8 + HYR9*DYR9 + HYR10*DYR10 + HYR11*DYR11] + [HYR1X1 + HYR2X1*DYR2 + HYR3X1*DYR3+ HYR4X1*DYR4 + HYR5X1*DYR5 + HYR6X1*DYR6 + HYR7X1 *DYR7 + HYR8X1*DYR8 + HYR9X1*DYR9 + HYR10X1*DYR10 + HYR11X1*DYR11]*DX1 + [HYR2X2*DYR2 + HYR3X2*DYR3+ HYR4X2*DYR4 + HYR5X2*DYR5 + HYR6X2*DYR6 + HYR7X2*DYR7 + HYR8X2*DYR8 + HYR9X2*DYR9 + HYR10X2*DYR10 + HYR11X2*DYR1 1]*DX2 + [HYR2X3*DYR2 + HYR3X3*DYR3+ HYR4X3*DYR4 + HYR5X3*DYR5 + HYR6X3*DYR6 + HYR7X3*DYR7 + HYR8X3*DYR8 + HYR9X3*DYR9 + HYR10X3*DYR10 + HYR11X3*DYR11]*DX3 ] ]; ? + (HI1*DI1) + (HI2*DI2) + (HI3*DI3) ;? *(DX1=0 & D X2=0 & DX3=0) IDENT EQ4 TH0 = [1/(AIJ 1)]*[ LOG(AI) LOG(AIJ) LOG(BIJ) ]; IDENT EQ5 TH1 = [1/(AIJ 1)]; IDENT EQ6 TH2 = [ 1/( AIJ 1) ]*(AI 1); EQSUB EQ4 EQ2 EQ3 EQ7;

PAGE 161

161 EQSUB EQ5 EQ2; EQSUB EQ6 EQ2 EQ7; EQSUB EQ1 EQ4 EQ5 EQ6; PARAM K 0 .75 KI1 .75 KI2 .75 KI3 .75; PARAM G0 .75 GX1 .75 GX2 .75 GX3 .75 GF1 .75 GG1 .75 GI1 .75 ; ? GI1 .75 GI3 .75 GI2 .050; PARAM H0 .75 ; ? HI1 .75 ; ? HI2 .75; ? HI3 .75 ; PARAM HYR2 .75 HYR3 .75 HYR4 .75 HYR5 .75 HYR6 .75 HYR7 .75 HYR8 .75 HY R9 .75 HYR10 .75 HYR11 .75 HYR1X1 .75 HYR2X1 .75 HYR3X1 .75 HYR4X1 .75 HYR5X1 .75 HYR6X1 .75 HYR7X1 .75 HYR8X1 .75 HYR9X1 .75 HYR10X1 .75 HYR11X1 .75 HYR2X2 .75 HYR3X2 .75 HYR4X2 .75 HYR5X2 .75 HYR6X2 .75 HYR7X2 .75 HYR8X2 .75 HYR9X2 .75 HYR10X2 .75 H YR11X2 .75 HYR2X3 .75 HYR3X3 .75 HYR4X3 .75 HYR5X3 .75 HYR6X3 .75 HYR7X3 .75 HYR8X3 .75 HYR9X3 .75 HYR10X3 .75 HYR11X3 .75; SELECT (WYEAR>1996) & (MISS(WQQ)=0) & (MISS(WQALL)=0) & (MISS(WRP)=0) & (WRP>.50); TREND TTT; CORR DI1 DI2 DI3; LSQ(HETER O) EQ1; PRINT @COEF; MAT NR = NROW(@COEF); PRINT NR; ? SHOW ALL; ANALYZ EQ4 EQ5 EQ6 EQ2 EQ3 EQ7; MSD WQQ WQALL; SET KK0 = @COEF(1); SET KKI1 = @COEF(2); SET KKI2 = @COEF(3); SET GG0 = @COEF(4); SET GGX1 = @COEF(5); SET GGX2 = @COEF(6); SET GGX3 = @COEF(7); SET GGI1 = @COEF(8); SET GGF1 = @COEF(9); SET GGG1 = @COEF(10); SET HH0 = @COEF(11); SET HHYR2 = @COEF(12); SET HHYR3 = @COEF(13); SET HHYR4 = @COEF(14); SET HHYR5 = @COEF(15); SET HHYR6 = @COEF(16); SET HHYR7 = @COEF(17); SET HHYR8 = @COEF(18); SET HHYR9 = @COEF(19); SET HHYR10 = @COEF(20); SET HHYR11 = @COEF(21); SET HHYR1X1 = @COEF(22); SET HHYR2X1 = @COEF(23); SET HHYR3X1 = @COEF(24); SET HHYR4X1 = @COEF(25); SET HHYR5X1 = @COEF(26); SET HHYR6X1 = @COEF(27); SET HHYR7X1 = @COEF(28); SET HHYR8X1 = @CO EF(29);

PAGE 162

162 SET HHYR9X1 = @COEF(30); SET HHYR10X1 = @COEF(31); SET HHYR11X1 = @COEF(32); SET HHYR2X2 = @COEF(33); SET HHYR3X2 = @COEF(34); SET HHYR4X2 = @COEF(35); SET HHYR5X2 = @COEF(36); SET HHYR6X2 = @COEF(37); SET HHYR7X2 = @COEF(38); SET HHYR8X2 = @COEF(39); SET HHYR9X2 = @COEF(40); SET HHYR10X2 = @COEF(41); SET HHYR11X2 = @COEF(42); SET HHYR2X3 = @COEF(43); SET HHYR3X3 = @COEF(44); SET HHYR4X3 = @COEF(45); SET HHYR5X3 = @COEF(46); SET HHYR6X3 = @COEF(47); SET HHYR7X3 = @COEF(48); SET HHYR8X3 = @COEF(49); SET HHYR9X3 = @COEF(50); SET HHYR10X3 = @COEF(51); SET HHYR11X3 = @COEF(52); AAID= 1 /[ 1 + EXP[ KK0 + KKI1*DI1 + KKI2*DI2 ]]; AAIJ = 1 / [1+ EXP[GG0 + GGI1*DI1 + GGX1*DX1 + GGX2*DX2 + GGX3*DX3 + GGF1*DF1 + GGG1*(WFOREIGN/10000)] ] ; BBIJ = [EXP[HH0 + [HHYR2*DYR2 + HHYR3*DYR3+ HHYR4*DYR4 + HHYR5*D YR5 + HHYR6*DYR6 + HHYR7*DYR7 + HHYR8*DYR8 + HHYR9*DYR9 + HHYR10*DYR10 + HHYR11*DYR11] + [HHYR1X1 + HHYR2X1*DYR2 + HHYR3X1*DYR3+ HHYR4X1*DYR4 + HHYR5X1*DYR5 + HHYR6X1*DYR6 + HHYR7X1*DYR7 + HHYR8X1*DYR8 + HHYR9X1*DYR9 + HHYR10X1*DYR10 + H HYR11X1*DYR11 ]*DX1 + [HHYR2X2*DYR2 + HHYR3X2*DYR3+ HHYR4X2*DYR4 + HHYR5X2*DYR5 + HHYR6X2*DYR6 + HHYR7X2*DYR7 + HHYR8X2*DYR8 + HHYR9X2*DYR9 + HHYR10X2*DYR10 + HHYR11X2*DYR11]*DX2 + [HHYR2X3*DYR2 + HH YR3X3*DYR3+ HHYR4X3*DYR4 + HHYR5X3*DYR5 + HHYR6X3*DYR6 + HHYR7X3*DYR7 + HHYR8X3*DYR8 + HHYR9X3*DYR9 + HHYR10X3*DYR10 + HHYR11X3*DYR11]*DX3 ] ]; TT0= (1/(AAIJ 1) )*( LOG(AAID) LOG(AAIJ) LOG(BBIJ) ); TT1 = ( 1/(AAIJ 1) ); TT2 = ( 1/(AA IJ 1) )*(AAID 1); SET I=0; ========================================================================; RESETTING THE SIMULATION VARIABLES BASE IS FRESH BEEF ; ========================================================================; DOT( CHAR=%) SDI1 SDI2 SDI3 SDX1 SDX2 SDX3 SDF1 SWFOREIGN SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; SET .=0; SET SDF1=1; ENDDOT; MSD(PRINT) WRP WFOREIGN; SET MEANWRP=1.0; SET MEANWFOREIGN=@MEAN(2); LL=(DI1=1); MSD(WEIGHT=LL,NOPRINT) WQALL; SET MEANWQALL1=@MEAN;

PAGE 163

163 LL=(DI2=1); MSD(WEIGHT=LL,NOPRINT) WQALL; SET MEANWQALL2=@MEAN; LL=(DI3=1); MSD(WEIGHT=LL,NOPRINT) WQALL; SET MEANWQALL3=@MEAN; LL=(DI4=1); MSD(WEIGHT=LL,NOPRINT) WQALL; SET MEANWQALL4=@MEAN; LL=(DX1=1); MSD(WEIGHT=LL,PRIN T) WRP; SET MEANWRP1=@MEAN; LL=(DX2=1); MSD(WEIGHT=LL,PRINT) WRP; SET MEANWRP2=@MEAN; LL=(DX3=1); MSD(WEIGHT=LL,PRINT) WRP; SET MEANWRP3=@MEAN; LL=(DX4=1); MSD(WEIGHT=LL,PRINT) WRP; SET MEANWRP4=@MEAN; PRINT MEANWRP1 MEANWRP2 MEANWRP3 MEANWRP4 ; S ET ADJ_WRP=1.0; SET ADJ_WQALL=1.0; SET ADJ_WFOREIGN=1.0; SET SWRP= [ (MEANWRP1 (1 ADJ_WRP))*SDX1 + (MEANWRP2 (1 ADJ_WRP))*SDX2 + (MEANWRP3 (1 ADJ_WRP))*SDX3 + (MEANWRP4 (1 ADJ_WRP))*(SDX1=0 & SDX2=0 & SDX3=0) ]; SET SWFOREIGN=MEANWFOREIGN*ADJ_WFOREIGN; SET SWQALL=[ MEANWQALL1*SDI1*ADJ_WQALL + MEANWQALL2*SDI2*ADJ_WQALL + MEANWQALL3*SDI3*ADJ_WQALL + MEANWQALL4*(SDI1=0 & SDI2=0 & SDI3=0)*ADJ_WQALL ]/1000; PROC ZRESETZ; DOT(CHAR=%) SDI1 SDI2 SDI3 SDX1 SDX2 SDX3 SDF1 SWFOREIGN SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; SET .=0; SET SDF1=1; ENDDOT; ? NOTE THAT WE HAVE SET THE DEFAULT TO FRESH WHERE SDF1=1; MSD(NOPRINT) WRP WFOREIGN; SET MEANWRP=1.0; SET MEANWFOREIGN=@MEAN(2); LL=(DI1=1); MSD(WEIGHT=LL,NOPRINT) WQALL; SET MEAN WQALL1=@MEAN; LL=(DI2=1); MSD(WEIGHT=LL,NOPRINT) WQALL; SET MEANWQALL2=@MEAN; LL=(DI3=1); MSD(WEIGHT=LL,NOPRINT) WQALL; SET MEANWQALL3=@MEAN; LL=(DI4=1); MSD(WEIGHT=LL,NOPRINT) WQALL; SET MEANWQALL4=@MEAN; LL=(DX1=1); MSD(WEIGHT=LL,PRINT) WRP; SET MEAN WRP1=@MEAN; LL=(DX2=1); MSD(WEIGHT=LL,PRINT) WRP; SET MEANWRP2=@MEAN; LL=(DX3=1); MSD(WEIGHT=LL,PRINT) WRP; SET MEANWRP3=@MEAN; LL=(DX4=1); MSD(WEIGHT=LL,PRINT) WRP; SET MEANWRP4=@MEAN; SET ADJ_WRP=1.0; SET ADJ_WQALL=1.0; SET ADJ_WFOREIGN=1.0; SET SWRP= [ (MEANWRP1 (1 ADJ_WRP))*SDX1 + (MEANWRP2 (1 ADJ_WRP))*SDX2 + (MEANWRP3 (1 ADJ_WRP))*SDX3 + (MEANWRP4 (1 ADJ_WRP))*(SDX1=0 & SDX2=0 & SDX3=0) ];SET SWFOREIGN=MEANWFOREIGN*ADJ_WFOREIGN; SET SWQALL=[ MEANWQALL1*SDI1*ADJ_WQALL + MEANWQALL2* SDI2*ADJ_WQALL + MEANWQALL3*SDI3*ADJ_WQALL + MEANWQALL4*(SDI1=0 & SDI2=0 & SDI3=0)*ADJ_WQALL ]; PRINT MEANWRP1 MEANWRP2 MEANWRP3 MEANWRP4 ; ENDPROC; MFORM(TYPE=GEN,NROW=5500,NCOL=50) MBEEFM=0; SET SIMNUM=0; SET SIMVAR=0; == ======================================================================; SIMULATOR FOR PREDICTING BOTH Xij AND THE MARKET SHARES ; ========================================================================; PROC ZBSIMZ; SET I=I+1;

PAGE 164

164 SET SWRP= [ (MEANWRP1 (1 ADJ_WRP))*SDX1 + (MEANWRP2 (1 ADJ_WRP))*SDX2 + (MEANWRP3 (1 ADJ_WRP))*SDX3 + (MEANWRP4 (1 ADJ_WRP))*(SDX1=0 & SDX2=0 & SDX3=0) ]; SET SWQALL=[ MEANWQALL1*SDI1*ADJ_WQALL + MEANWQALL2*SDI2*ADJ_WQALL + MEANWQALL3*SDI3*ADJ_WQALL + MEANWQALL4*(SDI1=0 & SDI2=0 & SDI3=0)*ADJ_WQALL ]/1000; SET SWFOREIGN=(MEANWFOREIGN)*ADJ_WFOREIGN; SET SAAID= 1/[ 1 + EXP[ KK0 + KKI1*SDI1 + KKI2*SDI2 ]]; SET SAAIJ = 1 / [1+ EXP[GG0 + GGI1*SDI1 + GGX1*SDX1 + GGX2*SDX2 + GGX3*SDX3 + GGF1*SDF1 + GG G1*(SWFOREIGN/10000)] ] ; SET SBBIJ = [EXP[HH0 + [HHYR2*SDYR2 + HHYR3*SDYR3+ HHYR4*SDYR4 + HHYR5*SDYR5 + HHYR6*SDYR6 + HHYR7*SDYR7 + HHYR8*SDYR8 + HHYR9*SDYR9 + HHYR10*SDYR10 + HHYR11*SDYR11] + [HHYR1X1 + HHYR2X1*SDYR2 + HHYR3X1*SDYR3+ HHYR4X1*SDYR4 + HHYR5X1*SDYR5 + HHYR6X1*SDYR6 + HHYR7X1*SDYR7 + HHYR8X1*SDYR8 + HHYR9X1*SDYR9 + HHYR10X1*SDYR10 + HHYR11X1*SDYR11 ]*SDX1 + [HHYR2X2*SDYR2 + HHYR3X2*SDYR3+ HHYR4X2*SDYR4 + HHYR5X2 *SDYR5 + HHYR6X2*SDYR6 + HHYR7X2*SDYR7 + HHYR8X2*SDYR8 + HHYR9X2*SDYR9 + HHYR10X2*SDYR10 + HHYR11X2*SDYR11]*SDX2 + [HHYR2X3*SDYR2 + HHYR3X3*SDYR3+ HHYR4X3*SDYR4 + HHYR5X3*SDYR5 + HHYR6X3*SDYR6 + HHYR7X3*SDYR7 + HHYR8X3*SDYR8 + HHYR9X3*SDYR9 + HHYR10X3*SDYR10 + HHYR11X3*SDYR11]*SDX3 ] ]; SET STT0= (1/(SAAIJ 1) )*( LOG(SAAID) LOG(SAAIJ) LOG(SBBIJ) ); SET STT1 = ( 1/(SAAIJ 1) ); SET STT2 = ( 1/(SAAIJ 1) )*(SAAID 1); SET SXIJ = [ EXP(STT0) [ SWRP**STT1 ] [ (SWQALL)* *(STT2)] ]; SET SXIJ_SHARE = SXIJ/SWQALL; SET ELIJ=STT1*(1 SXIJ_SHARE*SWRP); SET SIMNUM = SIMNUM; SET SIMVAR = SIMVAR; SET J=1; SET MBEEFM(I,J)= SIMNUM; SET J=2; SET MBEEFM(I,J)= SIMVAR; SET J=3; SET MBEEFM(I,J)= SXIJ; ? PRODUCT DEMAND; SET J=4 ; SET MBEEFM(I,J)= SXIJ_SHARE; ? MARKET SHARE BY QUANTITY; SET J=5; SET MBEEFM(I,J)= ELIJ; ? ELASTICITY; SET J=6; SET MBEEFM(I,J)= SWRP; SET J=7; SET MBEEFM(I,J)= ADJ_WRP; SET J=8; SET MBEEFM(I,J)= SWQALL; SET J=9; SET MBEEFM(I,J)= ADJ_WQALL; SET J=10;SET MBEEFM(I,J)= SWFOREIGN; SET J=11;SET MBEEFM(I,J)= ADJ_WFOREIGN;; DOT(CHAR=#,INDEX=K) STT0 STT1 STT2 SDI1 SDI2 SDI3 SDX1 SDX2 SDX3 SDF1 SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; SET J=K+11; SET MBEEFM (I,J)= .#; ENDDOT; PRINT I; ENDPROC ZBSIMZ;

PAGE 165

165 ===============================================================================; STARTING THE BEEF TRADE SIMULATOR ; SIM #1 IS THE BASE STARTING VALUES ; ===============================================================================; SET SIMNUM = 1; ? BASE ZRESETZ; SET SIMVAR = 1; ZBSIMZ; ===============================================================================; STA RTING THE BEEF TRADE SIMULATOR ; SIM #2 TRADE BY IMPORTER AND EXPORTER ; ===============================================================================; SET SIMNUM = 2; ZR ESETZ; SET SIMVAR = 1; SET SDI1=1; SET SDX1=1; ZBSIMZ; ? US TO KOREA; ZRESETZ; SET SIMVAR = 2; SET SDI1=1; SET SDX2=1; ZBSIMZ; ? AUSTRALIA TO KOREA; ZRESETZ; SET SIMVAR = 3; SET SDI1=1; SET SDX3=1; ZBSIMZ; ? NEW ZEALAND TO KOREA; ZRESETZ; S ET SIMVAR = 1; SET SDI2=1; SET SDX1=1; ZBSIMZ; ? US TO JAPAN; ZRESETZ; SET SIMVAR = 2; SET SDI2=1; SET SDX2=1; ZBSIMZ; ? AUSTRALIA TO JAPAN; ZRESETZ; SET SIMVAR = 3; SET SDI2=1; SET SDX3=1; ZBSIMZ; ? NEW ZEALAND TO JAPAN; ZRESETZ; SET SIMVA R = 1; SET SDI3=1; SET SDX1=1; ZBSIMZ; ? US TO TAWAIN; ZRESETZ; SET SIMVAR = 2; SET SDI3=1; SET SDX2=1; ZBSIMZ; ? AUSTRALIA TO TAWAIN; ZRESETZ; SET SIMVAR = 3; SET SDI3=1; SET SDX3=1; ZBSIMZ; ? NEW ZEALAND TO TAWAIN; ZRESETZ; SET SIMVAR = 1 ; SET SDX1=1; ZBSIMZ; ? US TO HONG KONG; ZRESETZ; SET SIMVAR = 2; SET SDX2=1; ZBSIMZ; ? AUSTRALIA TO HONG KONG; ZRESETZ; SET SIMVAR = 3; SET SDX3=1; ZBSIMZ; ? NEW ZEALAND TO HONG KONG; ============ ===================================================================; STARTING THE BEEF TRADE SIMULATOR ; SIM #3 US TRADE OVER TIME TO COUNTIRES ; ============================= ==================================================; SET SIMNUM = 3; ZRESETZ; SET SIMVAR = 1; SET SDI1=1; SET SDX1=1; ZBSIMZ; ? US TO KOREA; DOT(INDEX=H,CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; ZRESETZ; SET SIMVAR = H+1; SET .X=1; SET SDI1=1; SET SDX1=1; ZBSIMZ; ENDDOT; SET HH=SIMVAR; ZRESETZ; SET SIMVAR =1; SET SDI2=1; SET SDX1=1; ZBSIMZ; ? US TO JAPAN; DOT(INDEX=H,CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; ZRESETZ; SET SIMVAR = H+1; SET .X=1; SET SDI2=1; SET SDX1=1; ZBSIMZ; ENDDOT; SET HH=SIMVAR; ZRESETZ; SET SIMVAR =1; SET SDI3=1; SET SDX1=1; ZBSIMZ; ? US TO TAWIN; DOT(INDEX=H,CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; ZRESETZ; SET SIMVA R = H+1; SET .X=1; SET SDI3=1; SET SDX1=1; ZBSIMZ; ENDDOT; SET HH=SIMVAR; ZRESETZ; SET SIMVAR =1; SET SDX1=1; ZBSIMZ; ? US TO HONG KONG; DOT(INDEX=H,CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11;

PAGE 166

166 ZRESETZ; S ET SIMVAR = H+1; SET .X=1; SET SDX1=1; ZBSIMZ; ENDDOT; SET HH=SIMVAR; ===============================================================================; STARTING THE BEEF TRADE SIMULATOR ; SIM #4 AUSTRAL IA TRADE OVER TIME TO COUNTIRES ; ===============================================================================; SET SIMNUM = 4; ZRESETZ; SET SIMVAR = 1; SET SDI1=1; SET SDX2=1; ZBSIMZ; ? AUSTRALIA TO KOREA; DOT(INDEX= H,CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; ZRESETZ; SET SIMVAR = H+1; SET .X=1; SET SDI1=1; SET SDX2=1; ZBSIMZ; ENDDOT; SET HH=SIMVAR; ZRESETZ; SET SIMVAR =0+1; SET SDI2=1; SET SDX2=1; ZBSIMZ; ? AUSTRALIA TO JAPA N; DOT(INDEX=H,CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; ZRESETZ; SET SIMVAR = H+1; SET .X=1; SET SDI2=1; SET SDX2=1; ZBSIMZ; ENDDOT; SET HH=SIMVAR; ZRESETZ; SET SIMVAR =0+1; SET SDI3=1; SET SDX2=1; ZBSIMZ; ? AUST RALIA TO TAWIN; DOT(INDEX=H,CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; ZRESETZ; SET SIMVAR = H+1; SET .X=1; SET SDI3=1; SET SDX2=1; ZBSIMZ; ENDDOT; SET HH=SIMVAR; ZRESETZ; SET SIMVAR =0+1; SET SDX2=1; ZBSIMZ; ? AU STRALIA TO HONG KONG; DOT(INDEX=H,CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; ZRESETZ; SET SIMVAR = H+1; SET .X=1; SET SDX2=1; ZBSIMZ; ENDDOT; SET HH=SIMVAR; ============================================================= ==================; STARTING THE BEEF TRADE SIMULATOR ; SIM #5 NEW ZEALAND TRADE OVER TIME TO COUNTIRES ; ========================================================================== =====; SET SIMNUM = 5; ZRESETZ; SET SIMVAR = 1; SET SDI1=1; SET SDX3=1; ZBSIMZ; ? NEW ZEALAND TO KOREA; DOT(INDEX=H,CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; ZRESETZ; SET SIMVAR = H+1; SET .X=1; SET SDI1=1; SET SDX3 =1; ZBSIMZ; ENDDOT; SET HH=SIMVAR; ZRESETZ; SET SIMVAR =0+1; SET SDI2=1; SET SDX3=1; ZBSIMZ; ? NEW ZEALAND TO JAPAN; DOT(INDEX=H,CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; ZRESETZ; SET SIMVAR = H+1; SET .X=1; SET SDI 2=1; SET SDX3=1; ZBSIMZ; ENDDOT; SET HH=SIMVAR; ZRESETZ; SET SIMVAR =0+1; SET SDI3=1; SET SDX3=1; ZBSIMZ; ? NEW ZEALAND TO TAWIN; DOT(INDEX=H,CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; ZRESETZ; SET SIMVAR = H+1; SET .X=1; SET SDI3=1; SET SDX3=1; ZBSIMZ; ENDDOT; SET HH=SIMVAR;

PAGE 167

167 ZRESETZ; SET SIMVAR =0+1; SET SDX3=1; ZBSIMZ; ? NEW ZEALAND TO HONG KONG; DOT(INDEX=H,CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; ZRESETZ; SET SIMVAR = H +1; SET .X=1; SET SDX3=1; ZBSIMZ; ENDDOT; SET HH=SIMVAR; ===============================================================================; STARTING THE BEEF TRADE SIMULATOR ; SIM #6 RELATIVE PRICE A DJUSTMENTS IN KOREA ; THIS SHOWS EACH EXPORTERS SHARE TO KOREA AS RELATIVE PRICES CHANGE ; ===============================================================================; SET SIMNUM = 6; ZRESETZ; SET SIMVAR = 1; SET SDI1=1; SET SDX1=1; ZBSIMZ; ? US DO ADJ_WRP = .90 TO 1.10 BY .025; SET SIMVAR = SIMVAR+1; SET SDI1=1; SET SDX1=1; ZBSIMZ; ENDDO; SET H=SIMVAR; ZRESETZ; SET SIMVAR = 1; SET SDI1=1; SET S DX2=1; ZBSIMZ; ? AUSTRALIA DO ADJ_WRP = .90 TO 1.10 BY .025; SET SIMVAR = SIMVAR+1; SET SDI1=1; SET SDX2=1; ZBSIMZ; ENDDO; SET H=SIMVAR; ZRESETZ; SET SIMVAR = 1; SET SDI1=1; SET SDX3=1; ZBSIMZ; ? NEW ZEALA ND; DO ADJ_WRP = .90 TO 1.10 BY .025; SET SIMVAR = SIMVAR+1; SET SDI1=1; SET SDX3=1; ZBSIMZ; ENDDO; SET H=SIMVAR; ===============================================================================; STARTING THE BEEF TRADE SIMULATOR ; SIM #7 RELATIVE PRICE ADJUSTMENTS IN JAPAN ; THIS SHOWS EACH EXPORTERS SHARE TO JAPAN AS RELATIVE PRICES CHANGE ; ======================================================== =======================; SET SIMNUM = 7; ZRESETZ; SET SIMVAR = 1; SET SDI2=1; SET SDX1=1; ZBSIMZ; ? US DO ADJ_WRP = .90 TO 1.10 BY .025; SET SIMVAR = SIMVAR+1; SET SDI2=1; SET SDX1=1; ZBSIMZ; ENDDO; SET H=SIMVAR; ZRESETZ; SET SIMVAR = 1; SET SDI2=1; SET SDX2=1; ZBSIMZ; ? AUSTRALIA DO ADJ_WRP = .90 TO 1.10 BY .025; SET SIMVAR = SIMVAR+1; SET SDI2=1; SET SDX2=1; ZBSIMZ; ENDDO; SET H=SIMVAR; ZRESETZ; SET SIMVAR = 1; SET SDI2=1; SET SDX3=1; ZBSIMZ; ? NEW ZEALAND; DO ADJ_WRP = .90 TO 1.10 BY .025; SET SIMVAR = SIMVAR+1; SET SDI2=1; SET SDX3=1; ZBSIMZ; ENDDO; SET H=SIMVAR;

PAGE 168

168 ============================================================== =================; STARTING THE BEEF TRADE SIMULATOR ; SIM #8 RELATIVE PRICE ADJUSTMENTS IN TAWAIN ; THIS SHOWS EACH EXPORTERS SHARE TO TAWAIN AS RELATIVE PRICES CHANGE ; ===============================================================================; SET SIMNUM = 8; ZRESETZ; SET SIMVAR = 1; SET SDI3=1; SET SDX1=1; ZBSIMZ; ? US DO ADJ_WRP = .90 TO 1.10 BY .025; SET SIMVAR = SIMVAR+1; SET SDI3=1; SET SDX1=1; ZBSIMZ; ENDDO; SET H=SIMVAR; ZRESETZ; SET SIMVAR = 1; SET SDI3=1; SET SDX2=1; ZBSIMZ; ? AUSTRALIA DO ADJ_WRP = .90 TO 1.10 BY .025; SET SIMVAR = SIMVAR+1; SET SDI3=1; SET SDX2=1; ZB SIMZ; ENDDO; SET H=SIMVAR; ZRESETZ; SET SIMVAR = 1; SET SDI3=1; SET SDX3=1; ZBSIMZ; ? NEW ZEALAND; DO ADJ_WRP = .90 TO 1.10 BY .025; SET SIMVAR = SIMVAR+1; SET SDI3=1; SET SDX3=1; ZBSIMZ; ENDDO; SET H=SIMVAR; == =============================================================================; STARTING THE BEEF TRADE SIMULATOR ; SIM #9 RELATIVE PRICE ADJUSTMENTS IN HONG KONG ; THIS SHOWS EA CH EXPORTERS SHARE TO HONG KONG AS RELATIVE PRICES CHANGE ; ===============================================================================; SET SIMNUM = 9; ZRESETZ; SET SIMVAR = 1; SET SDX1=1; ZBSIMZ; ? US DO A DJ_WRP = .90 TO 1.10 BY .025; SET SIMVAR = SIMVAR+1; SET SDX1=1; ZBSIMZ; ENDDO; SET H=SIMVAR; ZRESETZ; SET SIMVAR = 1; SET SDX2=1; ZBSIMZ; ? AUSTRALIA DO ADJ_WRP = .90 TO 1.10 BY .025; SET SIMVAR = SIMVAR+1; SET SDX2=1; ZBSIMZ; ENDDO; SET H=SIMVAR; ZRESETZ; SET SIMVAR = 1; SET SDX3=1; ZBSIMZ; ? NEW ZEALAND; DO ADJ_WRP = .90 TO 1.10 BY .025; SET SIMVAR = SIMVAR+1; SET SDX3=1; ZBSIMZ; ENDDO; SET H=SIMVAR; ========================== =====================================================; STARTING THE BEEF TRADE SIMULATOR ; SIM #10 RELATIVE TOTAL DEMAND IS ADJUSTMENTS IN KOREA ; ========================== =====================================================; SET SIMNUM = 10; ZRESETZ; SET SIMVAR = 1; SET SDI1=1; SET SDX1=1; ZBSIMZ; ? US DO ADJ_WQALL = .80 TO 1.20 BY .05; SET SIMVAR = SIMVAR+1;

PAGE 169

169 SET SDI1=1; SET SDX1=1; ZBSIMZ; ENDDO; SET H=SIMVAR; ZRESETZ; SET SIMVAR = 1; SET SDI1=1; SET SDX2=1; ZBSIMZ; ? AUSTRALIA DO ADJ_WQALL = .80 TO 1.20 BY .05; SET SIMVAR = SIMVAR+1; SET SDI1=1; SET SDX2=1; ZBSIMZ; ENDDO; SET H=SIMVAR; ZRESETZ; SET SIMVAR = 1; SET SDI1=1; SET SDX3=1; ZBSIMZ; ? NEW ZEALAND; DO ADJ_WQALL = .80 TO 1.20 BY .05; SET SIMVAR = SIMVAR+ 1; SET SDI1=1; SET SDX3=1; ZBSIMZ; ENDDO; SET H=SIMVAR; ===============================================================================; STARTING THE BEEF TRADE SIMULATOR ; SIM #11 RELATIVE TOTAL DEMAND IS ADJUSTMENTS IN JAPAN ; ===============================================================================; SET SIMNUM = 11; ZRESETZ; SET SIMVAR = 1; SET SDI2=1; SET SDX1=1; ZBSIMZ; ? US DO ADJ_WQALL = .80 TO 1.20 BY .05; SET SIMVAR = SIMVAR+1; SET SDI2=1; SET SDX1=1; ZBSIMZ; ENDDO; SET H=SIMVAR; ZRESETZ; SET SIMVAR = 1; SET SDI2=1; SET SDX2=1; ZBSIMZ; ? AUSTRALIA DO ADJ_WQALL = .80 TO 1.20 BY .05; SET SIMVAR = SIMVAR+1; SET SDI2=1; SET SDX2=1; ZBSIMZ; ENDDO; SET H=SIMVAR; ZRESETZ; SET SIMVAR = 1; SET SDI2=1; SET SDX3=1; ZBSIMZ; ? NEW ZEALAND; DO ADJ_WQALL = .80 TO 1.20 BY .05; SET SIMVAR = SIMVAR+1; SET SDI2=1; SET SDX3=1; ZBSIMZ; ENDDO; SET H=SIMVAR; ===============================================================================; STARTING THE BEEF TRADE SIMULATOR ; SIM #12 RELATIVE TOTAL DEMAND IS ADJUSTMENTS IN TAWAIN ; ===============================================================================; SET SIMNUM = 12; ZRESETZ; SET SIMVAR = 1; SET SDI3=1; SET SDX1=1; ZBSIMZ; ? US DO ADJ_WQALL = .80 TO 1.20 BY .05; SET SIMVAR = SIMVAR+1; SET SDI3=1; SET SDX1=1; ZBSIMZ; ENDDO; SET H=SIMVAR; ZRESETZ; SET SIMVAR = 1; SET SDI3=1; SET SDX2=1; ZBSIMZ; ? AUSTRALIA D O ADJ_WQALL = .80 TO 1.20 BY .05; SET SIMVAR = SIMVAR+1; SET SDI3=1; SET SDX2=1; ZBSIMZ; ENDDO; SET H=SIMVAR; ZRESETZ; SET SIMVAR = 1; SET SDI3=1; SET SDX3=1; ZBSIMZ; ? NEW ZEALAND;

PAGE 170

170 DO ADJ_WQALL = .80 TO 1.20 BY .05; SET SIMVAR = SIMVAR+1; SET SDI3=1; SET SDX3=1; ZBSIMZ; ENDDO; SET H=SIMVAR; ===============================================================================; STARTING THE BEEF TRADE SIMULATOR ; SIM #13 RELATIVE TOTAL DEMAND IS ADJUSTMENTS IN HONG KONG ; ===============================================================================; SET SIMNUM = 13; ZRESETZ; SET SIMVAR = 1; SET SDX1=1; ZBSIMZ ; ? US DO ADJ_WQALL = .80 TO 1.20 BY .05; SET SIMVAR = SIMVAR+1; SET SDX1=1; ZBSIMZ; ENDDO; SET H=SIMVAR; ZRESETZ; SET SIMVAR = 1; SET SDX2=1; ZBSIMZ; ? AUSTRALIA DO ADJ_WQALL = .80 TO 1.20 BY .05; SET SIMVAR = SIMVAR+1; SET SDX2=1; ZBSIMZ; ENDDO; SET H=SIMVAR; ZRESETZ; SET SIMVAR = 1; SET SDX3=1; ZBSIMZ; ? NEW ZEALAND; DO ADJ_WQALL = .80 TO 1.20 BY .05; SET SIMVAR = SIMVAR+1; SET SDX3=1; ZBSIMZ; ENDDO; SET H=SIMVAR; ========================== =====================================================; STARTING THE BEEF TRADE SIMULATOR ; SIM #14 RELATIVE TOTAL FOREIGN PROMOTIONS IS ADJUSTMENTS IN KOREA ; ==================================== ===========================================; SET SIMNUM = 14; ZRESETZ; SET SIMVAR = 1; SET SDI1=1; SET SDX1=1; ZBSIMZ; ? US DO ADJ_WFOREIGN = .70 TO 1.30 BY .05; SET SIMVAR = SIMVAR+1; SET SDI1=1; SET SDX1=1; ZBSI MZ; ENDDO; SET H=SIMVAR; ZRESETZ; SET SIMVAR = 1; SET SDI1=1; SET SDX2=1; ZBSIMZ; ? AUSTRALIA DO ADJ_WFOREIGN = .70 TO 1.30 BY .05; SET SIMVAR = SIMVAR+1; SET SDI1=1; SET SDX2=1; ZBSIMZ; ENDDO; SET H=SIMVAR; ZRESETZ; SET SIMVAR = 1; SET SDI1=1; SET SDX3=1; ZBSIMZ; ? NEW ZEALAND; DO ADJ_WFOREIGN = .70 TO 1.30 BY .05; SET SIMVAR = SIMVAR+1; SET SDI1=1; SET SDX3=1; ZBSIMZ; ENDDO; SET H=SIMVAR; ========================= ======================================================; STARTING THE BEEF TRADE SIMULATOR ; SIM #15 RELATIVE TOTAL FOREIGN PROMOTIONS IS ADJUSTMENTS IN JAPAN ; ======================================= ========================================;

PAGE 171

1 71 SET SIMNUM = 15; ZRESETZ; SET SIMVAR = 1; SET SDI2=1; SET SDX1=1; ZBSIMZ; ? US DO ADJ_WFOREIGN = .70 TO 1.30 BY .05; SET SIMVAR = SIMVAR+1; SET SDI2=1; SET SDX1=1; Z BSIMZ; ENDDO; SET H=SIMVAR; ZRESETZ; SET SIMVAR = 1; SET SDI2=1; SET SDX2=1; ZBSIMZ; ? AUSTRALIA DO ADJ_WFOREIGN = .70 TO 1.30 BY .05; SET SIMVAR = SIMVAR+1; SET SDI2=1; SET SDX2=1; ZBSIMZ; ENDDO; SET H=SIMV AR; ZRESETZ; SET SIMVAR = 1; SET SDI2=1; SET SDX3=1; ZBSIMZ; ? NEW ZEALAND; DO ADJ_WFOREIGN = .70 TO 1.30 BY .05; SET SIMVAR = SIMVAR+1; SET SDI2=1; SET SDX3=1; ZBSIMZ; ENDDO; SET H=SIMVAR; ================ ===============================================================; STARTING THE BEEF TRADE SIMULATOR ; SIM #16 RELATIVE TOTAL FOREIGN PROMOTIONS IS ADJUSTMENTS IN TAWAIN ; ========================== =====================================================; SET SIMNUM = 16; ZRESETZ; SET SIMVAR = 1; SET SDI3=1; SET SDX1=1; ZBSIMZ; ? US DO ADJ_WFOREIGN = .70 TO 1.30 BY .05; SET SIMVAR = SIMVAR+1; SET SDI3=1; SE T SDX1=1; ZBSIMZ; ENDDO; SET H=SIMVAR; ZRESETZ; SET SIMVAR = 1; SET SDI3=1; SET SDX2=1; ZBSIMZ; ? AUSTRALIA DO ADJ_WFOREIGN = .70 TO 1.30 BY .05; SET SIMVAR = SIMVAR+1; SET SDI3=1; SET SDX2=1; ZBSIMZ; ENDDO ; SET H=SIMVAR; ZRESETZ; SET SIMVAR = 1; SET SDI3=1; SET SDX3=1; ZBSIMZ; ? NEW ZEALAND; DO ADJ_WFOREIGN = .70 TO 1.30 BY .05; SET SIMVAR = SIMVAR+1; SET SDI3=1; SET SDX3=1; ZBSIMZ; ENDDO; SET H=SIMVAR; === ============================================================================; STARTING THE BEEF TRADE SIMULATOR ; SIM #17 RELATIVE TOTAL FOREIGN PROMOTIONS IS ADJUSTMENTS IN HONG KONG ; ============= ==================================================================; SET SIMNUM = 17; ZRESETZ; SET SIMVAR = 1; SET SDX1=1; ZBSIMZ; ? US DO ADJ_WFOREIGN = .70 TO 1.30 BY .05; SET SIMVAR = SIMVAR+1; SET SDX1=1; ZBSIMZ; ENDDO; SET H=SIMVAR; ZRESETZ; SET SIMVAR = 1; SET SDX2=1; ZBSIMZ; ? AUSTRALIA D O ADJ_WFOREIGN = .70 TO 1.30 BY .05; SET SIMVAR = SIMVAR+1;

PAGE 172

172 SET SDX2=1; ZBSIMZ; ENDDO; SET H=SIMVAR; ZRESETZ; SET SIMVA R = 1; SET SDX3=1; ZBSIMZ; ? NEW ZEALAND; DO ADJ_WFOREIGN = .70 TO 1.30 BY .05; SET SIMVAR = SIMVAR+1; SET SDX3=1; ZBSIMZ; ENDDO; SET H=SIMVAR; ===============================================================================; STARTING THE BEEF TRADE SIMULATOR ; SIM #18 RELATIVE PRICE ADJUSTMENTS AND TIME FOR THE US ; == =============================================================================; SET SIMNUM = 18; ZRESETZ; SET SIMVAR = 1; SET SDI1=1; SET SDX1=1; ZBSIMZ; ? US DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 S DYR10 SDYR11; ZRESETZ; DO ADJ_WRP = .90 TO 1.10 BY .025; SET SIMVAR = SIMVAR+1; SET .X=1; SET SDI1=1; SET SDX1=1; ZBSIMZ; ENDDO; ENDDOT; ZRESETZ; SET SIMVAR = 1; SET SDI2=1; SET SDX1=1; ZBSIMZ; ? US DOT(C HAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; ZRESETZ; DO ADJ_WRP = .90 TO 1.10 BY .025; SET SIMVAR = SIMVAR+1; SET .X=1; SET SDI2=1; SET SDX1=1; ZBSIMZ; ENDDO; ENDDOT; ZRESETZ; SET SIMVAR = 1; SET SDI3=1; SET SDX1=1; ZBSIMZ; ? US DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; ZRESETZ; DO ADJ_WRP = .90 TO 1.10 BY .025; SET SIMVAR = SIMVAR+1; SET .X=1; SET SDI3=1; SET SDX1=1; ZBSIMZ; ENDDO ; ENDDOT; ZRESETZ; SET SIMVAR = 1; SET SDX1=1; ZBSIMZ; ? US DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; ZRESETZ; DO ADJ_WRP = .90 TO 1.10 BY .025; SET SIMVAR = SIMVAR+1; SE T .X=1; SET SDX1=1; ZBSIMZ; ENDDO; ENDDOT; ===============================================================================; STARTING THE BEEF TRADE SIMULATOR ; SIM #19 RELATIVE PRI CE ADJUSTMENTS AND TIME FOR AU ; ===============================================================================; SET SIMNUM = 19; ZRESETZ; SET SIMVAR = 1; SET SDI1=1; SET SDX2=1; ZBSIMZ; ? AU DOT(CHA R=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; ZRESETZ; DO ADJ_WRP = .90 TO 1.10 BY .025; SET SIMVAR = SIMVAR+1; SET .X=1; SET SDI1=1; SET SDX2=1; ZBSIMZ; ENDDO; ENDDOT; ZRESETZ; SET SIMVAR = 1; S ET SDI2=1; SET SDX2=1; ZBSIMZ; ? AU DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11;

PAGE 173

173 ZRESETZ; DO ADJ_WRP = .90 TO 1.10 BY .025; SET SIMVAR = SIMVAR+1; SET .X=1; SET SDI2=1; SET SDX2=1; ZBSIMZ; ENDDO; ENDDOT; ZRE SETZ; SET SIMVAR = 1; SET SDI3=1; SET SDX2=1; ZBSIMZ; ? AU DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; ZRESETZ; DO ADJ_WRP = .90 TO 1.10 BY .025; SET SIMVAR = SIMVAR+1; SET .X=1; SET SD I3=1; SET SDX2=1; ZBSIMZ; ENDDO; ENDDOT; ZRESETZ; SET SIMVAR = 1; SET SDX2=1; ZBSIMZ; ? AU DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; ZRESETZ; DO ADJ_WRP = .90 TO 1.10 BY .025 ; SET SIMVAR = SIMVAR+1; SET .X=1; SET SDX2=1; ZBSIMZ; ENDDO; ENDDOT; ===============================================================================; STARTING THE BEEF TRADE SIMULATOR ; SIM #20 RELATIVE PRICE ADJUSTMENTS AND TIME FOR NZ ; ===============================================================================; SET SIMNUM = 20; ZRESETZ; SET SIMVAR = 1; SET SDI1=1; SET SDX3=1; ZBSIMZ; ? NZ DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; ZRESETZ; DO ADJ_WRP = .90 TO 1.10 BY .025; SET SIMVAR = SIMVAR+1; SET .X=1; SET SDI1=1; SET SDX3=1; ZBSIMZ; ENDDO; ENDDOT; ZRESETZ; SET SIMVAR = 1; SET SDI2=1; SET SDX3=1; ZBSIMZ; ? NZ DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; ZRESETZ; DO ADJ_WRP = .90 TO 1.10 BY .025; SET SIMVAR = SIMVAR+1; SET .X=1; SET SDI2=1; SET SDX3=1; ZBSIMZ; ENDDO; ENDDOT; ZRESETZ; SET SIMVAR = 1; SET SDI3=1; SET SDX3=1; ZBSIMZ; ? NZ DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; ZRESETZ; DO ADJ_WRP = .90 TO 1.10 BY .025; SET SIMVAR = SIMVAR+1; SET .X=1 ; SET SDI3=1; SET SDX3=1; ZBSIMZ; ENDDO; ENDDOT; ZRESETZ; SET SIMVAR = 1; SET SDX3=1; ZBSIMZ; ? NZ DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; ZRESETZ; DO ADJ_WRP = .90 TO 1.10 BY .025; SET SIMVAR = SIMVAR+1; SET .X=1; SET SDX3=1; ZBSIMZ; ENDDO; ENDDOT;

PAGE 174

174 ===============================================================================; STARTING THE BEEF TRADE SIMULATOR ; SIM #21 RELATIVE DEMAND ADJUSTMENTS AND TIME FOR THE US ; ===============================================================================; SET SIMNUM = 21; ZRESETZ; SET SIMVAR = 1; SET SDI 1=1; SET SDX1=1; ZBSIMZ; ? US DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; ZRESETZ; DO ADJ_WQALL = .80 TO 1.20 BY .05; SET SIMVAR = SIMVAR+1; SET .X=1; SET SDI1=1; SET SDX1=1; ZBSIMZ; ENDDO; ENDDOT; ZRESETZ; S ET SIMVAR = 1; SET SDI2=1; SET SDX1=1; ZBSIMZ; ? US DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; ZRESETZ; DO ADJ_WQALL = .80 TO 1.20 BY .05; SET SIMVAR = SIMVAR+1; SET .X=1; SET SDI2=1; SET SDX1=1; ZBSIMZ; ENDDO; ENDDOT; ZRESETZ; SET SIMVAR = 1; SET SDI3=1; SET SDX1=1; ZBSIMZ; ? US DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; ZRESETZ; DO ADJ_WQALL = .80 TO 1.20 BY .05; SET SIMV AR = SIMVAR+1; SET .X=1; SET SDI3=1; SET SDX1=1; ZBSIMZ; ENDDO; ENDDOT; ZRESETZ; SET SIMVAR = 1; SET SDX1=1; ZBSIMZ; ? US DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; ZRES ETZ; DO ADJ_WQALL = .80 TO 1.20 BY .05; SET SIMVAR = SIMVAR+1; SET .X=1; SET SDX1=1; ZBSIMZ; ENDDO; ENDDOT; ===============================================================================; STARTING THE BEEF TRADE SIMULATOR ; SIM #22 RELATIVE DEMAND ADJUSTMENTS AND TIME FOR AU ; ===============================================================================; SET SIMNUM = 22; ZRESETZ; SET SIMVAR = 1; SET SDI1=1; SET SDX2=1; ZBSIMZ; ? AU DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; ZRESETZ; DO ADJ_WQALL = .80 TO 1.20 BY .05; SET SIMVAR = SIMVAR+1; SET .X=1; SET SDI1=1; SET SDX2=1; ZBSIMZ; ENDDO ; ENDDOT; ZRESETZ; SET SIMVAR = 1; SET SDI2=1; SET SDX2=1; ZBSIMZ; ? AU DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; ZRESETZ; DO ADJ_WQALL = .80 TO 1.20 BY .05; SET SIMVAR = SIMVAR+1; SET .X=1; SET SDI2=1; SET SDX2=1; ZBSIMZ; ENDDO; ENDDOT; ZRESETZ; SET SIMVAR = 1; SET SDI3=1; SET SDX2=1; ZBSIMZ; ? AU DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; ZRESETZ; DO ADJ_WQALL = .80 TO 1.20 BY .05; SET SIMVAR = SIMVAR+1;

PAGE 175

175 SET .X=1; SET SDI3=1; SET SDX2=1; ZBSIMZ; ENDDO; ENDDOT; ZRESETZ; SET SIMVAR = 1; SET SDX2=1; ZBSIMZ; ? AU DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDY R9 SDYR10 SDYR11; ZRESETZ; DO ADJ_WQALL = .80 TO 1.20 BY .05; SET SIMVAR = SIMVAR+1; SET .X=1; SET SDX2=1; ZBSIMZ; ENDDO; ENDDOT; ===============================================================================; STARTING THE BEEF TRADE SIMULATOR ; SIM #23 RELATIVE DEMAND ADJUSTMENTS AND TIME FOR NZ ; ===============================================================================; SET SIMNUM = 23; ZRESETZ; SET SIMVAR = 1; SET SDI1=1; SET SDX3=1; ZBSIMZ; ? NZ DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; ZRESETZ; DO ADJ_WQALL = .80 TO 1.20 BY .05; SET SIMVAR = SIMVAR+1; SET .X=1; SET SDI1=1; SET SDX3=1; ZBSIMZ; ENDDO; ENDDOT; ZRESETZ; SET SIMVAR = 1; SET SDI2=1; SET SDX3=1; ZBSIMZ; ? NZ DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; ZRESETZ; DO ADJ_WQALL = .80 TO 1.20 BY .05; SET SIMVAR = SIMVAR+1; SET .X=1; SET SDI2=1; SET SDX3=1; ZBSIMZ; ENDDO; ENDDOT; ZRESETZ; SET SIMVAR = 1; SET SDI3=1; SET SDX3=1; ZBSIMZ; ? NZ DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDYR5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; ZRESET Z; DO ADJ_WQALL = .80 TO 1.20 BY .05; SET SIMVAR = SIMVAR+1; SET .X=1; SET SDI3=1; SET SDX3=1; ZBSIMZ; ENDDO; ENDDOT; ZRESETZ; SET SIMVAR = 1; SET SDX3=1; ZBSIMZ; ? NZ DOT(CHAR=X) SDYR2 SDYR3 SDYR4 SDY R5 SDYR6 SDYR7 SDYR8 SDYR9 SDYR10 SDYR11; ZRESETZ; DO ADJ_WQALL = .80 TO 1.20 BY .05; SET SIMVAR = SIMVAR+1; SET .X=1; SET SDX3=1; ZBSIMZ; ENDDO; ENDDOT; ?WRITE(FORMAT=EXCEL,FILE='C: \ ZSTUDENT \ OFERRARA \ TSPPRG \ ARMCOEF#6.XLS') ?WYEAR WIMP WEXP WTYPE AID AIJ BIJ TH0 TH1 TH2 AAID AAIJ BBIJ TT0 TT1 TT2; WRITE(FORMAT=EXCEL,FILE='C: \ ZSTUDENT \ OFERRARA \ TSPPRG \ BEEF_SIM#7_Fresh.XLS') MBEEFM; END;

PAGE 176

176 LIST OF REFERENCES Adda, J "Behavior Towards Health Risks: An Empirical Study Using the "Mad Cow" Crisis as an Experiment." Journal of Risk and Uncertainty 35, No. 3 (2007): 285 305. Allen J. W., and T. R. Pierson. Food Marketing and Distribution: Implications for Agriculture Food Industry Reports; No. 9301. East Lansing, Mich.: Food Industry I nstitute, Michigan State University, 1993. Alston, J. M., C. A. Carter, R. Green, and D. Pick. "Whither Armington Trade Models?" American Journal of Agricultural Economics 72, No. 2 (1990): 455 67. Armington, P. S. "A Theory of Demand for Products Distingu ished by Place of Production." Staff Papers International Monetary Fund 16, No. 1 (1969): 159 78. Arrow, K. J., H. B. Chenery, B. S. Minhas, and R. M. Solow. Labor Substitution and Economic Efficiency." The Review of Economics and Statistics 43, No. 3 (1961): 225 50. Artus, J. R., and R. R. Rhomberg. "A Multilateral Exchange Rate Model." Staff Papers International Monetary Fund 20, No. 3 (1973): 591 611. Bhuiyan, M. A. "Measuring Revealed Competitiveness: Applications to World Vegetable Market. Ph.D. Dissertation, University of Florida, 1993. Blonigen, B. A., and W. Wesley "Explaining Armington: What Determines Substitutability between Home and Foreign Goods?" The Canadian Journal of Economics / Revue Canadienne d'Economique 32, No. 1 (1999): 1 21. Brester, G. W. "Estimation of the U.S. Import Demand Elasticity for Beef: The Importance of Disaggregation." Review of Agricultural Economics 18, No. 1 (1996): 31 42. Bureau, J. C., W. Jones, E. Gozlan, and S. Marette. "Overview and Conceptual Founda tion." In Global Food Trade and Consumer Demand for Quality edited by Barry Krissoff, Mary Bohman and Julie A. Caswell, 252. New York, NY: Kluwer Academics/ Plenum Publishers, 2002. Bureau, J. C., M. P. Ramos and L. Salvatici. "Tariffs, Trq's and Import C omposition: The Case of Beef Trade between the EU and Mercosur." In UMR Economie Publique edited by INRA INAPG, 28. Paris, France: European Trade Study Group, 2005. Burton, M. and Trevor Young. "The Impact of BSE on the Demand for Beef and Other Meats in Great Britain." Applied Economics 28, No. 6 (1996): 687 93.

PAGE 177

177 Buzby, J. C. "International Trade and Food Safety: Econom ic Theory and Case Studies." Agricultural Economic Report No. 828 Washington DC 2003. Cameron, A. C., a nd F. A. G. Windmeijer. "An R S quared Measure of Goodness of Fit for Some Common Nonlinear Regression Models." Journal of Econometrics 77, No. 2 (1997): 329 42. Capps, O., R. Tsai, R. Kirby, and G.W Williams. "A Compari son of Demands for Meat Products in the Pacific Rim Region." Journa l of Agricultural and Resource Economics 19, No. 1 (1994): 210 24. Carwell, R. T. "The Importance of Dynamics in Commodity Advertizing Demand Systems." Ph.D. Dissertation, University of Saskatchewan, 2005. Chiang, A. C. Fundamental Methods of Mathematica l Economics Third edition New York, NY: McGraw Hill/Irwin, 1987. Codron, J. M., J. A. Sterns, and T. Reardon. "Strategic Choices in Produce Marketing: Issues of Compatible Use and Exclusion Cost." Journal of Food Distribution Research Volume 34, N o. 03 ( 2003): 1 12. Coffey, B., J. Mintert, S. Fox, T. Schoeder, and L. Valentin. "The Economic Impact of BSE on the U.S. Beef Industry: Product Value Losses, Regulatory Cost, and Consumer Reactions." In Kansas State University Agricultural Experiment Station and Cooperative Extension Service 64: Kansas State University, 2005. Cong L.T., H. M. Kaiser, and W. Tomek "Export Promotion and Import Demand for Us Red Meat in Selected Pacific Rim Countries." Agribusiness 14, N o. 2 (1998): 95 105. Davis, G. C., and N. C Kruse. "Consistent Estimation of Armington Demand Models." American Journal of Agricultural Economics 75, No. 3 (1993): 719 23. Devadoss, S., D. W. Holland, L. Stodick and J. Ghosh. "A General Equilibrium Analysis of Foreign and Domestic Demand Shocks Arising from Mad Cow Disease in the United States." Journal of Agricultural and Resource Economics 31, No. 2 (2006): 441 53. Dyck, J.H., and K.E. Nelson. "Structure of the Global Markets for Meat." In Agriculture Information Bulletin edited by Economic Re search Service (ERS) Market and Trade Economics Division, 24. Washington, DC: United States Department of Agriculture, 2003. Fmnia, F. and A. Gohin. "Estimating Price Elasticities of Food Trade Functions: How Relev ant Is the Gravity Approach?". TRADEAG Working Papers edited by INRA France, 37 2007.

PAGE 178

178 Food and Agriculture Organi zation of the United Nations ( FAO ) FAOSTAT 2007 Available from http://faostat.fao.org/?alias=faostat Forker, O. and R.W. Wa rd. Commodity Advertizing: The Economics and Measurement of Generic Programs New York, NY: Lexington Books, 1993. Gallaway, M. P., C A. McDaniel, and S. A. Rivera. "Short Run and Long Run Industry Level Estimates of U.S. Armington Elasticities." The Nort h American Journal of Economics and Finance 14, No. 1 (2003): 49 68. Golan, A., J. M. Perloff, and E. Z. Shen. a Demand System with Non negativity Constraints: Mexican Meat Demand." The Review of Economics and Statistics 83, No. 3 (2001): 541 5 0. Greene, W. H. Econometric Analysis 5th ed. Upper Saddle River, NJ: Prentice Hall, 2003. Gujarati, D. N. Basic Econometrics 4th ed ition : McGraw Hill/Irwin, 2003. Gumbel, P. "Pasta Panic Strikes Italy." Fortune November 26, 200 7 Hall, B. H., and C. C ummins. TSP 5.0 : User's Guide Vol. 1. Palo Alto, CA: TSP International, 2005. Hanoch, G. "Cresh Production Functions." Econometrica 39 (1971): 695 712. Hayes, D. J., T. I. Wahl, and G W. Williams. "Testing Restrictions on a Model of Japanese Meat Demand. American Journal of Agricultural Economics 72, No. 3 (1990): 556 66. Herrmann, R., S R. Thompson, and S Krischik Bautz,. "Bovine Spongiform Encephalopathy and Generic Promotion of Beef: An Analysis for "Quality from Bavaria"." Agribusiness 18, No. 3 (2 002): 16. Hillberry, R. H., M. A. Anderson, E. J. Balistreri, and A. K. Fox. "Taste Parameters as Model Residuals: Assessing the "Fit" of an Armington Trade Model are Distributed across Exporters in Proportion to the Expenditure Elasticities." Review of In ternational Economics 13, No. 5 (2005): 973 84. Jin, H. J., A. Skripnitchenko, and W. W. Koo. "The Effects of th e BSE Outbreak in the United States on the Beef and Cattle Industry." S pecial Report 17. Fargo, ND: University of North Dakota, Center for Agri cultural Policy and Trade Studies, 2004. Jin, H. J., and W. W. Koo. "The Effect of the BSE Outbreak in Japan on Consumers' Preferences." Euro Rev Agric Econ 30, No. 2 (2003): 173 92.

PAGE 179

179 Jung, J "Understanding the Compass Model: Assumptions, Structure, and El asticity of Substitution." Ph.D. Dissertation, University of Florida, 2004. Kapuscinski, C. A., and P. G. Warr. "Estimation of Armington Elasticities: An Application to the Philippines." Economic Modeling 16, No. 2 (1999): 257 78. Kinnucan, H. W., P. A. Du ffy, and K. Z. Ackerman. "Effects of Price Versus Non Price Export Promotion: The Case of Cotton." Review of Agricultural Economics 17, No. 1 (1995): 91 1 00. Kinnucan, H. W., H. Xiao, C. J. Hsia, and J. D. Jackson. "Effects of Health Information and Generi c Advertising on U.S. Meat Demand." American Journal of Agricultural Economics 79, No. 1 (1997): 13 23. A Note." Agribusiness 20, No. 2 (2004): 181 188. Kinnucan, H. W. and O. Myrland. "On Generic vs Brand Promotion of Farm Products in Foreign Markets." Applied Economics 40, No. 6 (2008): 673 84. Kotler, P Marketing Management 11 th ed ition New York, NY: Prentice Hall, 2002. Latouche, K., P. Rainelli, and D. Vermersch. "Food Safety Issues and the BSE Scare: Some Lessons from the French Case." Food Policy 23, No. 5 (1998): 347 56. Lloyd P.J ., and X.G. Zhang. "The Armington Model." Productivity Commission Staff Working Paper (2006): 30. Mangen, M.J.J. and A.M. Burrell. Decomposing Preference Shifts for Meat and Fish in the Netherlands." 2001. Marsh, J. M., G. W. Brester and V. H. Smith. "Effects of North American BSE Events on U.S. Cattle Prices." Review of Agricultural Economics 30, No. 1 (2008): 136 50. McDaniel, C A. and E J. Balistreri. "A Discussion on Armington Trade Substitution Elasticities." Office of Economics Working Paper (USITC) No. 2002 01 A (2002): 17. McLaren, D. "On Aspects of Food Safety and International Trade." Singapore Economic Review 51, No. 2 ( 2006): 135 45. Miljkovic, D., and J. Hyun. "Import Demand for Quality in the Japanese Beef Market." Agricultural and Resource Economics Review 35, No. 2 (2006): 276 84. Mukerji, V "A Generalized S.M.A.C. Function with Constant Ratios of Elasticity of Su bstitution." The Review of Economic Studies 30, No. 3 (1963): 233 36.

PAGE 180

180 Mutondo, J. E., and S. Henneberry. "Competitiveness of U.S. Meats in Japan and South Korea: A Source Differentiated Market Study." Paper presented at the American Agricultural Economics Association, AAEA, Portland, O R, July 29 August 1, 2007. Nicholson, W Microeconomic Theory: Basic Principles and Extensions Eighth edition USA. South Western College /Thomson Learning, 2002 Peterson, H. H., and Y. J. Chen. "The Impact of BSE on Japanes e Retail Meat Demand." Agribusiness 21, No. 3 (2005): 313 27. Piggott N. E., J. A. Chalfant, J. M. Alston and G. R. Griffith. Demand Response to Advertising in the Australian Meat Industry. American Journal of Agricultural Economics 78, No. 2. (1996): 26 8 279. Poulin, D., and A. K. Boane. "Mad Cow Disease and Beef Trade." edited by Yvan Gervais, 7. Ottawa, Canada: Statistics Canada, 2003. Preckel, P. V., J. Cranfield, and T. W. Hertel. "Implicit Additive Preferences: A Further Generalization of the CES ." West Lafayette, Indiana: Dept. of Agricultural Economics, Purdue University, 2005. Regmi, A "Changing Structure of Global Food Consumption and Trade." Agricultural Economic Report WRS 01 1 Washington DC, 2001. Reinert, K. A., and D. W. Roland Holst. "Arm ington Elasticities for United States Manufacturing Sectors." Journal of Policy Modeling 14, No. 5 (1992): 631 39. Roig Franzia, M. "A Culinary and Cultural Staple in Crisis." The Washington Post Saturday, January 27, 2007. Saito, M "Armington Elasticiti es in Intermediate Inputs Trade: A Problem in Using Multilateral Trade Data." Canadian Journal of Economics/Revue Canadienne d'economique 37, No. 4 (2004): 1097 117. Sato, K. Two Level Constant Elasticity of Substitution Production Function." The Review of Economic Studies 34, No. 2 (1967): 201 18. Satyanarayana, V. and D Johnson. "Credit Guarantee Programs and U.S. Markets Share in Sel ected Wheat Import Markets." Agricultural Economics Report No. 13. Fargo ND 1998. Shiells, C. R., and K. A. Reinert. "Armington Models and Terms of Trade Effects: Some Econometric Evidence for North America." The Canadian Journal of Economics / Revue Canadienne d'Economique 26, No. 2 (1993): 299 316.

PAGE 181

181 Solomon, H. and H. W. Kinnucan. "Effects on Non Price Export Promotion: Some Evidence for Cotton." Australian Journal of Agricultural Economics 37, No. 1 (1993): 1 15. Sparks, A. L. and R. W. Ward. "A Simultaneous Econometric Model of World Fresh Vegetable Trade, 1962 82: An Application of Nonlinear Simultaneous Equations." J ournal of Agricultural Economics Research 44, No. 2 (1992). U. S. Department of Agriculture Economic Research Service (ERS) Foreign Agricultural Trade of the United States. 2 007 Available from http://www.fas.usda.gov/psdonline/psdReport.aspx U.S. Beef and Cattle Industry: Background Statistics and Information 2007 Available from http://www.ers.usda.gov/news/BSECoverage.htm U. S. Department of Agriculture Foreign Agricultural Service (FAS). Understanding Korea's Beef Import System Livestock and Poultry: World Markets and Trade No. 4. Washington DC, 1996. Commodity and Market Analysis: World Beef Trade Overview 2004. Avai lable from http://www.fas.usda.gov/dlp2/circular/2004/04 10LP/toc.htm Production, Supply and Distribution Online. 2007. Available from http://www.fas.usda.gov/psdonline/psdReport.aspx Unite d States Meat Export Federation ( USMEF ) "The Value of U.S. Beef and Beef Variety Meat Exports." 5. Denver CO: USMEF, 2007. Van Osselaer, S. M. and J. W. Alba. "Consumer Learning and Brand Equity." Journal of Consumer Research 27, No. 1 (2000): 16. Varian, Hal R. Microeconomic Analysis 3rd ed ition New York: W.W. Norton & Company, 1992. Verbeke, W. and R. W. Ward. "A Fresh Meat Almost Ideal Demand System Incorporating Negative TV Press and Advertising Impact." Agricultural Economics 25, No. 2 3 (2001): 359 74. Ward, R. W. Advertising and Promotions." Agro Food Marketing edited by et al Daniel. I. Padberg, 319 48: CAB International, 1997. Ward R. W., and C. Lambert. "Generic Pro motion of Beef Measuring the Impact of the United States Beef Checkoff." Journal of Agricultural Economics 44, No. 3 (1993): 456 65. Weatherspoon, D. D., and J. L. Seale. "Do the Japanese Discriminate against Australian Beef Imports ? Evidence from the Di fferential Approach." Journal of Agricultural and Applied Econometric 27, No. 2 (1995): 536 43.

PAGE 182

182 Webb, A. J., E. E. Figueroa, W. E. Wecker, and Alex F. McCalla. "Imp act of the Soviet Grain Embargo: a Comparison of Methods." Journal of Policy Modeling 11, No 3 (1989): 361 89. Winters, L. A "Separability and the Specification of Foreign Trade Functions." Journal of International Economics 17, No. 3 4 (1984): 239 63. World Customs Organization ( WCO ) What Is the Harmonized System (H S ) Nomenclature? 2008. Avai lable from http://www.wcoomd.org/home_wco_topics_hsoverviewboxes.htm World Organization for Animal Health ( OIE ) World Animal Health Situation Bovine Spongiform Encephalopathy ( BSE ) 2007 Available from http://www.oie.int/eng/info/en_esb.htm World Trade Atlas, GTIS. World Trade Atlas (Internet Version 4.3e.) Global Trade Information Services, Inc., 2007. Yang, S R. and W. W. Koo. "A Generalized Armington Trade Model: Re specification." Agricultural Economics No. 9 (1993): 374 76.

PAGE 183

BIOGRAPHICAL SKETCH Oscar Ferrara was born and raised in Asuncion, Paraguay. In 1992 he earned a B.S. degree in agricultural engineering at the Universidad Nacional de Asuncin in Paraguay. In August 2001 he earned a B.S. degree in applied economics from the University of Minnesota and in August 2005 he earned a M.S. degree in food and resource economics at the University of Florida. In November 2007 he was admitted to candidacy for Doctor of Philosophy in food and resource economics at the University of Florida. His research interests are focused on agricultural marketing, trade and food safety.