Price Transmission in Thai Aquaculture Product Markets: An Analysis Along Value Chain and Across Species

This article was downloaded by: [University of Prince Edward Island], [Dr Kehar Singh]On: 09 March 2015, At: 06:44Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Click for updates

Aquaculture Economics & ManagementPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/uaqm20

Price Transmission in Thai AquacultureProduct Markets: An Analysis Along ValueChain and Across SpeciesKehar Singha, Madan M. Deyb, Amporn Laowapongc & Umesh Bastolad

a Agricultural/Resource Economics Division, Department of HealthManagement, University of Prince Edward Island, Charlottetown,Canadab Aquaculture/Fisheries Center, University of Arkansas at Pine Bluff,Pine Bluff, Arkansas, USAc Department of Fisheries, Ministry of Agriculture and Cooperative,Bangok, Thailandd School of Economic Sciences, Washington State University,Pullman, Washington, USAPublished online: 03 Mar 2015.

To cite this article: Kehar Singh, Madan M. Dey, Amporn Laowapong & Umesh Bastola (2015) PriceTransmission in Thai Aquaculture Product Markets: An Analysis Along Value Chain and Across Species,Aquaculture Economics & Management, 19:1, 51-81, DOI: 10.1080/13657305.2015.994236

To link to this article: http://dx.doi.org/10.1080/13657305.2015.994236

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the“Content”) contained in the publications on our platform. However, Taylor & Francis,our agents, and our licensors make no representations or warranties whatsoever as tothe accuracy, completeness, or suitability for any purpose of the Content. Any opinionsand views expressed in this publication are the opinions and views of the authors,and are not the views of or endorsed by Taylor & Francis. The accuracy of the Contentshould not be relied upon and should be independently verified with primary sourcesof information. Taylor and Francis shall not be liable for any losses, actions, claims,proceedings, demands, costs, expenses, damages, and other liabilities whatsoever orhowsoever caused arising directly or indirectly in connection with, in relation to or arisingout of the use of the Content.

This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &

http://crossmark.crossref.org/dialog/?doi=10.1080/13657305.2015.994236&domain=pdf&date_stamp=2015-03-03

http://www.tandfonline.com/loi/uaqm20

http://www.tandfonline.com/action/showCitFormats?doi=10.1080/13657305.2015.994236

http://dx.doi.org/10.1080/13657305.2015.994236

Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Dow

nloa

ded

by [

Uni

vers

ity o

f Pr

ince

Edw

ard

Isla

nd],

[D

r K

ehar

Sin

gh]

at 0

6:44

09

Mar

ch 2

015

http://www.tandfonline.com/page/terms-and-conditions

http://www.tandfonline.com/page/terms-and-conditions

Aquaculture Economics & Management, 19:51–81, 2015Copyright © Taylor & Francis Group, LLCISSN: 1365-7305 print/1551-8663 onlineDOI: 10.1080/13657305.2015.994236

PRICE TRANSMISSION IN THAI AQUACULTURE PRODUCT

MARKETS: AN ANALYSIS ALONG VALUE CHAIN AND

ACROSS SPECIES

Kehar Singh,1 Madan M. Dey,2 Amporn Laowapong,3 and Umesh Bastola4

1Agricultural/Resource Economics Division, Department of Health Management, Universityof Prince Edward Island, Charlottetown, Canada2Aquaculture/Fisheries Center, University of Arkansas at Pine Bluff, Pine Bluff, Arkansas,USA3Department of Fisheries, Ministry of Agriculture and Cooperative, Bangok, Thailand4School of Economic Sciences, Washington State University, Pullman, Washington, USA

& We have examined the presence of price transmission asymmetry along the value chain, and theprice transmission across four main aquaculture species in Thai fish market using monthly datafrom January 2001 to October 2010. This is an attempt to contribute to the literature on horizontaland vertical price transmission in the seafood markets including the price transmission asymmetryin the developing countries. We did not find any evidence of asymmetric price transmission inwalking catfish (except in long-run), vannamei shrimp and tilapia; however, it is evident inThai Asian sea bass market; wholesalers exercising some market power. In general, price of onespecies is not found to significantly affect price of the other species at the same level of value chain.

Keywords price transmission across species, price transmission asymmetry, pricetransmission models, thai fish market, vertical price transmission

INTRODUCTION

Understanding relationship between prices is an important area ofresearch in the food markets. Two common forms of price transmissionare price linkage along a value chain (vertical price transmission) and pricelinkages across market places and different commodities (horizontal pricetransmission). The extent to which a price shock at one market/level ofvalue chain affects a price in other market/value chain level provides anassessment of the functioning of markets.

Address correspondence to Kehar Singh, Research Scientist (Agricultural/Resource Economics),Department of Health Management, University of Prince Edward Island, Charlottetown PE C1A 4P3,Canada. E-mail: [email protected]

Dow

nloa

ded

by [

Uni

vers

ity o

f Pr

ince

Edw

ard

Isla

nd],

[D

r K

ehar

Sin

gh]

at 0

6:44

09

Mar

ch 2

015

mailto:[email protected]

http://dx.doi.org/10.1080/13657305.2015.994236

Horizontal price transmission studies include spatial as well as cross-commodity price transmissions. The theoretical foundation of spatialprice transmission is the spatial arbitrage and the consequent Law onOne Price. On the other hand, cross-commodity price transmission ismostly influenced by the substitutability and complementary relationsamong products. Recent studies on the horizontal price linkages in seafoodmarkets include Nielsen (2004, 2005); Asche et al. (2005, 2007, 2012,2014); Nielsen et al. (2007, 2009); Vinuya (2007); Norman-Lopez andAsche (2008); Norman-Lopez (2009); and, Jimenez-Toribio et al. (2010).Nielsen (2004) found that the “Law of One Price” is in force between theNorwegian and Danish herring markets. Asche et al. (2005) examinedmarket integration between wild and farmed salmon on the Japanesemarket and found that the species were close substitutes on the market,and that the expansion of farmed salmon had resulted in price decreasesfor all salmon species. Nielsen (2005) identified a partially integratedEuropean first-hand market for whitefish and as a part of this, a perfectlyspatially integrated cod market.

Asche et al. (2007), while studying the supply chain for salmon whichoriginates in Norway and the United Kingdom and is then sold at retaillevel in France as smoked salmon, found markets as integrated. Nielsenet al. (2007) found that markets for farmed trout are related to other fishmarkets in Germany, and that markets for these trout are more closelylinked to markets for captured fish than to farmed salmon. Using importprice data from Japan, United States, and European Union, Vinuya(2007) tested market integration and the Law of One Price in the worldshrimp market. Norman-Lopez and Asche (2008) found that imports offresh and frozen tilapia fillets lie in different market segments, while freshand frozen catfish fillets compete in the same market. Norman-Lopez(2009) showed that fresh farmed tilapia fillets compete with wild wholered snapper, wild fresh fillets of sea bass, and blackback flounder in theU.S. market.

Nielsen et al. (2009) identified a loose form of market integrationbetween 13 fresh and seven frozen fish species in Europe. They found thatthe Law of One Price is in force on the fresh market within the segments offlatfish and pelagic fish in Europe. Jimenez-Toribio et al. (2010) examinedthe degree of integration between the world market and the majorEuropean marketplaces of frozen and canned tuna through both verticaland spatial price relationships. They found that the European market forfinal goods is segmented between the Northern countries consuming low-priced canned skipjack tuna imported from Asia (mainly Thailand) andthe Southern countries (Italy, Spain) processing and importing yellowfin-based products sold at higher prices. Asche et al. (2012) used detailed dataon shrimp prices by size class and import prices to conduct a co-integration

52 K. Singh et al.

Dow

nloa

ded

by [

Uni

vers

ity o

f Pr

ince

Edw

ard

Isla

nd],

[D

r K

ehar

Sin

gh]

at 0

6:44

09

Mar

ch 2

015

analysis of market integration in the U.S. shrimp market. They found asignificant evidence of market integration, suggesting that the ‘Law ofOne Price’ holds for this industry. Asche et al. (2014) examined marketintegration and price transmission for salmon in France, and found thatprice determination processes are different for fresh and smoke salmon.

The literature analyzing vertical price linkages has concentrated onevaluations of the links between farm, wholesale and retail prices (Vavra& Goodwin, 2005). The price relationships along the value chain provideinsights into marketing efficiency, and consumer and farmer welfare(Aguiar & Santana, 2002). It is to mention here that the relationshipsbetween two stages in the value chain are well developed by the theory ofderived demand; however, the high data requirements to estimate suchrelationships often make it impossible to estimate. Therefore, analysis ofjust prices at different levels of the market chain is more commonlyemployed. Vertical price linkages in seafood markets are not studied much.

A few recent studies to consider are: Jimenez-Toribio et al. (2003,2010); Asche et al. (2007); Guillen and Franquesa (2008); Larsen andKinnucan (2009); Gonzales et al. (2013); Gordon and Maurice (2014).Jimenez-Toribio et al. (2003) used prices concerning ex-vessel markets,wholesale markets and foreign trade to study the impact of verticalintegration on price transmission in the fishing distribution channel ofthe striped venus (Chamellea gallina). Asche et al. (2007) found a highdegree of price transmission in supply chains for salmon which originatesin Norway and the United Kingdom and is then sold at retail level in Franceas smoked salmon.

Using weekly data, Guillen and Franquesa (2008) analyzed the pricetransmission elasticity of the main twelve seafood products in the Spanishmarket chain (ex-vessel, wholesale and retail stages). Larsen and Kinnucan(2009) presented a structural model that incorporates price linkage andmarketing margin identity into a common framework, and applied theframework to an international marketing channel using farmed salmon.The results of their study suggested that the markets were purely competi-tive and that the marketing margins only increased when costs of marketingservice increased.

Jimenez-Toribio et al. (2010) tested vertical price relationships betweenthe price of frozen tuna paid by the canneries and the price of canned fishin both Italy and France. The two species show an opposite pattern in pricestransmission along the value chain: price changes along the chain are farbetter transmitted for the “global” skipjack tuna than for the more“European” yellowfin tuna. Gonzales et al. (2013) studied asymmetric pricetransmission along the fish value chain by using a consistent thresholdautoregressive model, and found asymmetric price transmission for wildcod and farmed salmon marketed in France. Gordon and Maurice

Price Transmission in Thai Aquaculture Product Market 53

Dow

nloa

ded

by [

Uni

vers

ity o

f Pr

ince

Edw

ard

Isla

nd],

[D

r K

ehar

Sin

gh]

at 0

6:44

09

Mar

ch 2

015

(this issue) tested for vertical and horizontal co-integration for fish speciesin Uganda and found that ex-vessel prices are only weakly related to down-stream markets.

The asymmetric price transmission, i.e., increasing and decreasingprices at one level of value chain transmit at different rates to anotherlevel, has received considerable attention in agricultural economics.Meyer and von Cramon-Taubadel (2004) and Frey and Manera (2007)provide reviews of the literature on asymmetry price transmission. How-ever, the issue of asymmetric price transmission has been overlooked infish and fish product market studies (Jaffry, 2005). A few studies to men-tion are Jaffry (2005); Garcia (2006); Guillen and Franquesa (2008),Matsui et al. (2011); Nakajima et al. (2011). Jaffry (2005) found asym-metry in price transmission in the whole hake value chain in France.Garcia (2006) studied the hake prices transmission along the Spanishmarket chain. Guillen and Franquesa (2008) investigated the price trans-mission asymmetry in the 12 main seafood products in the Spanishmarket chain (ex-vessel, wholesale and retail levels). Matsui et al.(2011) and Nakajima et al. (2011) analyzed Japanese blue fin tuna marketand discussed that entities having the market power shifted fromupstream to downstream by tuna market structure change.

Common explanations of the existence of asymmetric farm-retail pricetransmission in the food sector include: market power, search costs, con-sumer response to changing prices, producer adjustment cost, and thebehavior of markups over the business cycle (Jaffry, 2005). The presenceof asymmetric price transmission is often considered as an evidence of mar-ket failure (Meyer & Cramon-Taubadel, 2004). Peltzman (2000) found thatasymmetric pricing is not just anecdotal, it’s closer to universal, and asym-metric pricing to be as common in unconcentrated industries as it was inconcentrated industries.

Though the number of studies on horizontal price linkages in the sea-food markets in the developed world has increased recently; however, it ishard to find studies in the developing countries. There are limited studieson vertical price transmission including the asymmetric price transmissionin seafood markets in the world. Given the large number of fish producedand consumed in seafood-producing developing Asian countries, it isimportant to analyze horizontal and vertical price transmission in the sea-food markets in those countries.

The present study contributes to the literature on horizontal and verti-cal price transmission in the seafood markets including the price trans-mission asymmetry in the developing countries. We have examined thepresence of price transmission asymmetry along the value chain, and theprice transmission across four main aquaculture species in Thai fish marketusing monthly data from January 2001 to October 2010. The fish species

54 K. Singh et al.

Dow

nloa

ded

by [

Uni

vers

ity o

f Pr

ince

Edw

ard

Isla

nd],

[D

r K

ehar

Sin

gh]

at 0

6:44

09

Mar

ch 2

015

considered in the analysis are vannamei shrimp (Penaeus vanamei), tilapia(Oreochromis niloticus), walking catfish (Clarius sp.) and Asian sea bass (Latescalcarifer). These species are different in terms of production and trade.Vannnamei shrimp is mainly for export, walking catfish and Asian sea bassare used mostly for domestic consumption, and tilapia is sold in both dom-estic and international markets. This article focuses on price transmissionin the Thai domestic market.

FISH PRODUCTION AND AQUACULTURE VALUE CHAIN IN

THAILAND

The fisheries and aquaculture sector plays a vital role in the food secur-ity and economy of Thailand. In 2011, total fish (including finfish, crusta-ceans and mollusks) production in the country was 3.78 million tons,valued at THB 131,053 million (approximately USD 4,700 million). Thecontribution of various sub-sectors to the total production in terms ofquantity included: marine capture (52%), inland capture (8%), coastalaquaculture (27%), and freshwater aquaculture (13%). Marine catchesin Thailand have decreased considerably over the last decade, from2,651,223 tons in 2003 to 1,612,073 tons in 2012. On the other hand, theshare of aquaculture to the country’s total fish production has increasedsubstantially over the years.

In 2012, Thailand was the 7th top aquaculture producer in the worldwith the total production of 1,233,877 tons. About half of country’s aqua-culture production is crustaceans, consisting mostly of Vannamei shrimp(Penaeus vanamei). In 2011, Vannamei shrimp contributed about 73% oftotal coastal aquaculture production. Asian Sea bass (Lates calcarifer) isthe main marine finfish cultured in Thailand; about 86% of all marine fin-fish farms cultured Asian sea bass during 2011 (Department of Fisheries,2013). In terms of freshwater aquaculture, tilapia (Oreochromis niloticus)and walking catfish (Clarius sp.) account for 41% and 26% of total freshwater production, respectively.

The aquaculture value chain in Thailand is complex. It comprises manytypes of markets and involves a large number of parties when fish aremoved from fish farmers to consumers. Fish is sold fresh (both live anddead) as well as processed. Walking catfish, snakehead fish and sand gobyare usually sold live. Poor transportation may cause fish to die (or decreaseits quality) and reduces the price substantially.

The market structure for aquaculture products can be classifiedinto three major market levels: primary market, intermediate market andterminal market. Marketing fish starts with the primary market. Fishfarmers sell their harvest to every level but the highest proportion is


Dow

nloa

ded

by [

Uni

vers

ity o

f Pr

ince

Edw

ard

Isla

nd],

[D

r K

ehar

Sin

gh]

at 0

6:44

09

Mar

ch 2

015

sold to primary markets. Fish are then transported to the intermediatemarket, which is comprised of central assembly markets and wholesalemarkets.

During the study period in 2011–2012, there were 21 fisheries assemblymarkets in Thailand: 14 state markets, 6 privately owned markets and onebelonging to a fishery cooperative. The assembly markets are located inboth urban (11 markets) and rural areas (10 markets).The Fish MarketingOrganization (FMO), a state enterprise under the Ministry of Agricultureand Cooperatives and located in central Bangkok, manages state assemblymarkets, which handle both marine and freshwater fishes. Fish agents, bothFMO and private, as well as fish collectors in the assembly markets sell mostfish to wholesalers. These wholesale markets are mostly located in big citiesand in good locations

Wholesalers distribute most of the fish directly to retailers, while someare sold to processors/cold storage and also are exported. Fish is distribu-ted to consumers through retail outlets, including retail markets, supermar-kets, restaurants and hotels. It is common to find fish in supermarkets,which are emerging throughout the country.

METHODOLOGY

We have used following procedure to fulfill the objectives of the study:

i. Testing for a presence of the unit-root, Granger causality, andcointegration;

ii. Testing for the price transmission asymmetry along the value chain;and

iii. Specifying and estimating the price transmission models.

Unit Root, Granger Causality and Cointegration Tests

Important issues in the price transmission analysis are: a) stationarity/non-stationarity of the time series, b) the Granger causation, and c) co-inte-gration of non-stationary time series having same order of integration.Addressing these issues is important to decide on the regression modelto adopt for the price transmission analysis (stationarity/non-stationarityand cointegration) and the right-hand side (RHS) variables in the model(the Granger causation). If the series under study are stationary at levels,one can use traditional econometric tools like ‘ordinary least square’ esti-mation procedure to determine relationships between those series. Thenon-stationary series having unit root may be co-integrated if their orderof integration is same; one can use the “error correction models” to

56 K. Singh et al.

Dow

nloa

ded

by [

Uni

vers

ity o

f Pr

ince

Edw

ard

Isla

nd],

[D

r K

ehar

Sin

gh]

at 0

6:44

09

Mar

ch 2

015

determine the relationships. The “models in difference” can be used fornon-cointegrated series having unit root.

There are two types of tests used to test whether a time series is station-ary or not: the unit root tests and the stationarity tests. The unit root teststest the null of a unit root against an alternative of stationarity, or meanreversion. If the unit root null hypothesis is rejected, then the series is saidto be stationary. The presence of a unit root in the time series represen-tation of a variable has important implications for both the econometricmethod used and the economic interpretation of the model in which thatvariable appears. The Augmented Dickey–Fuller (ADF) test of Dickey andFuller (1979), the generalized least squares ADF (DF-GLS), the Point Opti-mal tests (PT) of Elliott et al. (ERS) (1996), and the Phillips–Perron test(Phillips & Perron, 1988) are commonly used univariate unit root tests.The stationarity tests test the null hypothesis of stationarity against a unitroot alternative. If the test fails to reject the null, the time series is saidto be stationary. The most commonly used test to test the null hypothesisof stationarity is KPSS (Kwiatkowski et al., 1992).

As is well known in the applied economics literature, even a test withDF-GLS’s favorable characteristics may still lack power to distinguishbetween the null hypothesis of nonstationary behavior (I(1)) and thestationary alternative (I(0)). The Ng-Perron test (Ng & Perron, 2001)modifies the Phillips and Perron (1988) test in a number of ways inorder to increase the test’s size and power. This testing procedureensures that non-rejections of the null hypothesis of the unit root arenot due to a low probability of rejecting a false null hypothesis, whilerejections are not related to size distortions. The Ng-Perron test con-structs four test statistics that are based upon the GLS de-trended data.These test statistics are modified forms of Phillips (1987) Za statisticsand Phillips and Perron (1988) Zt statistics, the Bhargava (1986) R1 stat-istic, which is built on the work of Sargan and Bhargava (1983), and theERS (1996) Point Optimal statistic. Considering the improved size andpower of the Ng–Perron (2001) test over other univariate unit root tests,we have used the same to test the null hypothesis of presence of unit rootin the series.

The next step is to determine whether the series having unit root arecointegrated or not. Cointegration between two time series integrated ofsame order can be tested with either by the Engle and Granger (1987) testor by the Johansen (1988) test; we have used the latter one. The Johansen(1988) cointegration test is an unrestricted cointegration test; Gonzalo(1994) discussed advantages/disadvantages of this test.

The issue of testing whether or not a variable precedes another variable,i.e., the Granger causality (Granger, 1969), is increasingly gaining attentionin empirical research (Hatemi-J, 2012). We followed the Toda and


Dow

nloa

ded

by [

Uni

vers

ity o

f Pr

ince

Edw

ard

Isla

nd],

[D

r K

ehar

Sin

gh]

at 0

6:44

09

Mar

ch 2

015

Yamamoto (1995) procedure to test for the Granger causality: i)determining maximum order of integration of two series, ii) setting up aVAR model in levels, iii) selecting appropriate maximum lag length for vari-ables in the VAR model, iv) testing for serial autocorrelation in the model,v) re-estimating the VAR model with appropriate lag length, and vi) testingthe null hypothesis.

We have estimated appropriate maximum lag order using: i) FPE (Finalprediction error), ii) AIC (Akaike information criterion), iii) SIC (Schwarzinformation criterion), and iv) HQIC (Hannan-Quinn informationcriterion). Then we have estimated the VAR model with lag order equalto maximum lag length selected using different information criteriaplus maximum order of integration of two series. Then we conducted(post-estimation test) to check for autocorrelation in the model using theLagrange-multiplier test (H0: no autocorrelation at lag order). If autocorre-lation is found in the selected lag length, we increased the lag length untilautocorrelation issue resolved and re-estimated the model. In the end we,tested the null hypothesis using the Wald test, which has asymptoticallychi-square distribution with p degree of freedom under the null hypothesis.For this test, we included only lag length selected on the basis of differentinformation criteria; extra lags (maximum order of integration andincreased lags to resolve autocorrelation) used are just to fix up theasymptotics.

Testing for the Price Transmission Asymmetry along the Value

Chain

Meyer and von Cramon-Taubadel (2004) provide a survey of the asym-metric price transmission methods. Based on the results of the Johansen(1988) cointegration test, we followed the Houck (1977) and Ward(1982) approach. This approach basically splits the change in explanatoryvariable into positive and negative changes.

We have considered three levels along the value chain: farm, wholesaleand retail. Based on the pair-wise Granger causality test, we determined thedirection of causation. Depending on the Granger causality test results, wehave extended the Houck (1977) and Ward (1982) model to consider tworegressors. The empirical model used in this article for testing its asym-metry can be expressed as:

lnP�i ¼ a0t þ

Xpl¼0

aþl cumDðln Pþj Þ

h it�l

þXpl¼0

a�l cumDðln P�j Þ

h it�l

þXqm¼0

bþm cumDðlnPþm Þ

� �t�l þ

Xqm¼0

b�l cumDðlnP�m Þ

� �þ e;

ð1Þ

58 K. Singh et al.

Dow

nloa

ded

by [

Uni

vers

ity o

f Pr

ince

Edw

ard

Isla

nd],

[D

r K

ehar

Sin

gh]

at 0

6:44

09

Mar

ch 2

015

where, cum and ln stand for cumulative and natural logarithmic value, respect-ively. Subscripts ‘i’, ‘j’ and ‘k’ stands for value chain level; ‘l’ and ‘m’ denote lagnumber; t is the time; lnP�

i ¼ lnPt � lnPt¼0 ; DðlnPþt Þ ¼ lnPt � lnPt�1, if

lnPt > lnPt � 1 and 0 otherwise; and DðlnP�t Þ ¼ ln Pt � ln Pt�1, if lnPt < lnPt �

1 and 0 otherwise. et is the error component. If the price series on the left-handside (LHS) of the equation are stationary at levels without trend, we did not usethe time as a variable on the RHS of the equation.

The null hypotheses of no difference tested against the alternatehypotheses of inequality are as follows:

Null HS10

� �: aþl ¼ a�l against alternate HS1

1

� �: aþl 6¼ a�l for l ¼ 1;2;3; . . . ð2:1Þ

Null HS20

� �: bþm ¼ b�m against alternate HS2

1

� �: bþm 6¼ b�m for m¼ 1;2;3; ::: ð2:2Þ

Null HS30

� �: aþl ¼ bþm against alternate HS3

1

� �: aþl 6¼ bmþ for l ¼m ð2:3Þ

Null HS40

� �: a�l ¼ b�m against alternate HS4

1

� �: a�l 6¼ b�m for l ¼m ð2:4Þ

Null HL10

� �:Xol¼1

aþl ¼Xol¼1

a�l against alternate HL11

� �:Xol¼1

aþl 6¼Xol¼1

a�l ð2:5Þ

Null HL20

� �:Xqm¼1

bþm ¼Xqm¼1

b�m against alternate HL21

� �:Xqm¼1

bþm 6¼Xqm¼1

b�im ð2:6Þ

Null HL30

� �:Xol¼1

aþl ¼Xqm¼1

bþm against alternate HL31

� �:Xol¼1

aþl 6¼Xqm¼1

bþim ð2:7Þ

Null HL40

� �:Xol¼1

a�l ¼Xqm¼1

b�m against alternate HL31

� �:Xol¼1

a�l 6¼Xqm¼1

b�im ð2:8Þ

The equality of the coefficients of the positive change and negativechange HS1

0 and HS20

� �provides the test on short run asymmetry. The

equality of the coefficients for the sum of positive change and sum of nega-tive change HL1

0 and HL20

� �gives the information on long run price trans-

mission asymmetry. Testing the null hypotheses HS30 and HL3

0 provides theevidence whether degree of positive changes in two regressors on thechanges in the dependent variable are significantly different from eachother or not in short run and long run, respectively. Similarly rejectionof null hypotheses HS4

0 and HL40 provides evidence of significant difference

in the influence of negative changes in two independent variables on thedependent variable.


Dow

nloa

ded

by [

Uni

vers

ity o

f Pr

ince

Edw

ard

Isla

nd],

[D

r K

ehar

Sin

gh]

at 0

6:44

09

Mar

ch 2

015

Specifying and Estimating the Price Transmission Models

We have identified the regressors based on the Granger causality testresults. If a price series on the LHS and the RHS price series (price serieswhich are the Granger cause of the series on the RHS) do not have unitroot, we have used a price transmission in levels [Equation (3.1)]. However,if any of the price series on the LHS and RHS have a unit root and two ormore price series are not cointegrated, we have used a model in difference[Equation (3.2)].

lnPik ¼ a0t þXl 6¼0

ailkðln PikÞt�l þXv 6¼k

Xl

ailvðlnPivÞt�l þXj

Xl

ajlkðln PjkÞt�l

þ et ;

ð3:1ÞD lnPik� � ¼ a0t þ

Xl 6¼0

ailk DðlnPikÞ� �

t�l þXv 6¼k

Xl

ailv DðlnPivÞ� �

t�l

þXj

Xl

ajlk DðlnPjkÞh i

t�lþ et ; ð3:2Þ

where subscript ‘i’ and ‘j’ denote the species, ‘v’ and ‘k’ denote value chainlevel, ‘l’ denote lag order and ‘t’ denote time. ‘P’ stands for price series, ‘ln’is the natural logarithmic value, a denote parameter and et is the errorcomponent. If the price series on the LHS of the equation are stationaryat levels without trend, we did not use the time as a variable on the RHSof the equation. As we have used logarithmic form, the estimated para-meters (a) are short run price transmission elasticities. The long run elasti-cities along the value chain (gLRVC) and across species (gLRS ) are computed asfollows:

gLRVC ¼Xl

ailv � 1�Xl 6¼0

ail

!and gLRS ¼

Xl

ajl � 1�Xl 6¼0

ail

!ð4Þ

The analyses have been done on the STATA12 software (STATACORPLP, Texas, USA). Equations (1), (3.1) and (3.2) were estimated using theCochrane–Orcutt regression, which corrects for the auto-correlation, ifany, in the time series. We have used EView6 software (IHS Inc.) for theGranger causality test. Using F-test in STATA12, we tested the null hypoth-eses given in Equations (2.1) to (2.8).

We have used monthly price data on different fish species at differentlevels of supply chain, collected by different agencies. The time period of dataused ranges from January 2001 to October 2010 (Appendix 1). Data on farm-gate price and retail price of all four species were obtained from the Office ofAgriculture and Cooperative (Ministry of Agriculture and Cooperative) andthe Department of Internal Trade (Ministry of Commerce), respectively.

60 K. Singh et al.

Dow

nloa

ded

by [

Uni

vers

ity o

f Pr

ince

Edw

ard

Isla

nd],

[D

r K

ehar

Sin

gh]

at 0

6:44

09

Mar

ch 2

015

Wholesale level price of Asian sea bass, catfish, and tilapia were collected fromthe FishMarket Organization under the Ministry of Agriculture and Coopera-tives, Thailand, and the wholesale prices of shrimp were obtained fromshrimp wholesale market, Samut Sakom province, Thailand.

RESULTS AND DISCUSSION

Unit Root, Granger Causality and Cointegration

As mentioned in the methodology section, we have used Ng–Perron(2001) test to test the null hypothesis of the presence of the unit root inthe series. Table 1 presents the unit root test results for different time seriesunder study. The price series namely, shrimp farm, shrimp wholesale,shrimp retail, walking catfish wholesale, tilapia farm and tilapia wholesaleare stationary at levels, whereas the price series namely walking catfish farmand walking catfish retail are trend stationary (Table 1). Asian sea bassfarm, wholesale and retail price series, and tilapia retail price series haveunit root. These series are stationary in first difference without a lineartrend; we have taken the liberty not to present these results in Table 1. Itis to mention here that we have also used the ADF and Phillips–Perron unitroot tests, and the Kwiatkowski–Phillips–Schmidt–Shin stationarity test.

TABLE 1 Ng–Perron Unit Root Tests

SpeciesValue Chain

Level,

Constant ConstantþLinear Trend

MZa MZt MSB MPT LL MZa MZt MSB MPT LL

Shrimp F �16.20 �2.83 0.17 1.58 1 �24.91 �3.53 0.14 3.68 1W �12.21 �2.47 0.20 2.02 2 �20.42 �3.14 0.15 4.80 2R �18.11 �3.00 0.17 1.37 1 �31.12 �3.92 0.13 3.05 1

Asian sea bass F �3.01 �1.19 0.40 8.08 0 �6.26 �1.70 0.27 14.53 0W �0.10 �0.07 0.70 30.42 0 �7.53 �1.83 0.24 12.33 0R �5.17 �1.53 0.30 4.95 1 �15.03 �2.72 0.18 6.18 1

Walking Catfish F �5.17 �1.42 0.28 5.21 0 �22.21 �3.33 0.15 4.10 0W �13.87 �2.52 0.18 2.20 0 �27.56 �3.71 0.13 3.31 0R �0.53 �0.30 0.55 19.68 1 �18.31 �2.98 0.16 5.23 1

Tilapia F �12.96 �2.28 0.18 2.88 0 �23.11 �3.34 0.14 4.29 0W �8.69 �2.00 0.23 3.13 1 �29.74 �3.86 0.13 3.07 0R 1.16 0.55 0.47 21.37 0 �2.68 �0.97 0.36 28.26 0

Critical values* 1% �13.80 �2.58 0.17 1.78 �23.80 �3.42 0.14 4.035% �8.10 �1.98 0.23 3.17 �17.30 �2.91 0.17 5.4810% �5.70 �1.62 0.28 4.45 �14.20 �2.62 0.19 6.67

MZa and MZt¼Modified forms of Phillips (1987) Za statistics, Phillips and Perron (1988) Zt statistics,MSB¼ the Bhargava (1986) R1 statistics, MPT¼ the Elliott, Rothenberg, and Stock Point Optimal (ERS,1996) Point Optimal statistic, LL¼Lag length using Spectral GLS-detrended AR based on SIC.

The unit root hypothesis is rejected in favor of stationarity when MBS and ERS are smaller than theirrespective critical values, and MZa and MZt are greater than their respective critical values.

Figures in bold indicate rejection of null hypothesis up to the 0.05 level of significance.*Ng-Perron (2001, Table 1).


Dow

nloa

ded

by [

Uni

vers

ity o

f Pr

ince

Edw

ard

Isla

nd],

[D

r K

ehar

Sin

gh]

at 0

6:44

09

Mar

ch 2

015

The ADF test concluded that walking catfish wholesale and tilapia farmprice series are stationary at levels at the 0.05 level of significance, andshrimp retail, walking catfish farm, and tilapia wholesale price series aretrend stationary. The Phillips–Perron test showed shrimp farm, walkingcatfish wholesale, and tilapia farm and wholesale series as stationaryat levels, and walking catfish farm as trend stationary at the 0.05 level ofsignificance. The Kwiatkowski–Phillips–Schmidt–Shin test concluded thatshrimp farm, wholesale and retail price series, Asian sea bass retail priceseries, and walking catfish and tilapia wholesale prices series are trendstationary at the 0.05 level of significance. All tests conducted includingNg–Perron test showed that tilapia retail price series, Asian sea bass farm,wholesale and retail price series have unit root.

TABLE 2 Lag Length Selection for the Pairwise Granger Causality Test

Price Series FPE AIC HQIC SBIC Selected Lag*

Vannamei Shrimp Farm and Wholesale Price 3 3 3 1 3þ 0þ 0¼ 3Vannamei Shrimp Wholesale and Retail Price 3 3 3 1 3þ 0þ 0¼ 3Vannamei Shrimp Farm and Retail Price 2 2 1 1 2þ 0þ 0¼ 2Walking Catfish Farm and Wholesale Price 1 1 1 1 1þ 0þ 0¼ 1Walking Catfish Wholesale and Retail Price 2 2 2 2 2þ 0þ 0¼ 2Walking Catfish Farm and Retail Price 2 2 2 1 2þ 0þ 0¼ 2Asian Sea Bass Farm and Wholesale Price 2 2 1 1 2þ 1þ 0¼ 3Asian Sea Bass Wholesale and Retail Price 2 2 1 1 2þ 1þ 0¼ 3Sea bass Farm and Retail Price 2 2 2 1 2þ 1þ 0¼ 3Tilapia Farm and Wholesale Price 5 5 1 1 6þ 0þ 0¼ 6Tilapia Wholesale and Retail Price 1 1 1 1 1þ 1þ 0¼ 2Tilapia Farm and Retail Price 1 1 1 1 1þ 1þ 0¼ 2Vannamei Shrimp and Asian Sea bass Retail Price 2 2 2 1 2þ 1þ 0¼ 3Vannamei Shrimp and Walking Catfish Retail Price 2 5 2 2 2þ 1þ 0¼ 3Vannamei Shrimp and Tilapia Retail Price 2 2 1 1 2þ 1þ 0¼ 3Asian Sea Bass and Walking Catfish Retail Price 5 5 4 2 6þ 0þ 1¼ 6Asian Sea Bass and Tilapia Retail Price 2 2 2 1 2þ 1þ 0¼ 3Walking Catfish and Tilapia Retail Price 2 2 2 1 3þ 0þ 0¼ 3Shrimp and Asian Sea Bass Farm Price 1 1 1 1 1þ 1þ 0¼ 2Shrimp and Walking Catfish Farm Price 2 2 2 1 2þ 1þ 0¼ 3Shrimp and Tilapia Farm Price 2 2 1 1 2þ 0þ 0¼ 2Asian Sea bass and Walking Catfish Farm Price 2 2 2 1 2þ 1þ 0¼ 3Asian Sea Bass and Tilapia Farm Price 1 1 1 1 1þ 1þ 0¼ 2Walking Catfish and Tilapia Farm Price 1 1 1 1 1þ 1þ 0¼ 2Shrimp and Asian Sea bass Wholesale Price 3 3 3 1 3þ 1þ 0¼ 4Shrimp and Walking Catfish Wholesale Price** 5 5 2 2 5þ 0þ 1¼ 6Shrimp and Tilapia Wholesale Price 3 3 2 2 3þ 0þ 0¼ 3Asian Sea Bass and Walking Catfish Wholesale Price 1 6 1 1 1þ 1þ 0¼ 2Asian Sea Bass and Tilapia Wholesale Price 6 6 1 1 6þ 1þ 0¼ 7Walking Catfish and Tilapia Wholesale Price 2 2 1 1 2þ 0þ 0¼ 2

*Selected lag is equal to lag length based on different criteria plus maximum order of integration plusadditional lag(s) to adjust for autocorrelation. All shrimp and walking catfish price series, and tilapiafarm and WPs series are I(0) levels. All sea bass and tilapia RP series are I(1).**Autocorrelation at lag 5.

62 K. Singh et al.

Dow

nloa

ded

by [

Uni

vers

ity o

f Pr

ince

Edw

ard

Isla

nd],

[D

r K

ehar

Sin

gh]

at 0

6:44

09

Mar

ch 2

015

For the pairwise Granger causality test, Table 2 presents the selectednumber of lags. Tables 3A (along the value chain) and 3B (across the spe-cies) show results of the pair-wise Granger causality test. The test rejectedthe following null hypotheses (Table 3A):

i. Shrimp wholesale price does not Granger cause shrimp farm price,ii. Shrimp retail price does not Granger cause shrimp wholesale price,iii. Shrimp wholesale price does not Granger cause shrimp retail price,iv. Shrimp retail price does not Granger cause shrimp farm price,v. Walking catfish farm price does not Granger cause walking catfish

wholesale price,vi. Walking catfish retail price does not Granger cause walking catfish

wholesale price,vii. Asian sea bass farm price does not Granger cause Asian sea bass whole-

sale price,viii. Tilapia wholesale price does not Granger cause tilapia farm price,ix. Tilapia retail price does not Granger cause tilapia farm price; and,x. Tilapia farm price does not Granger cause tilapia retail price.

TABLE 3A Pairwise Granger Causality Tests on Prices of Fish Species in Thailand along Value Chain

Null Hypothesis Obs F-Statistic Prob.

Vannamei Shrimp WP does not Granger Cause Vannamei Shrimp FP 78 5.78 0.00Vannamei Shrimp FP does not Granger Cause Vannamei Shrimp WP 0.89 0.45Vannamei Shrimp RP does not Granger Cause Vannamei Shrimp WP 66 2.42 0.08Vannamei Shrimp WP does not Granger Cause Vannamei Shrimp RP 2.82 0.05Vannamei Shrimp RP does not Granger Cause Vannamei Shrimp FP 67 5.29 0.01Vannamei Shrimp FP does not Granger Cause Vannamei Shrimp RP 0.62 0.54Walking Catfish WP does not Granger Cause Walking Catfish FP 91 0.59 0.45Walking Catfish FP does not Granger Cause Walking Catfish WP 14.13 0.00Walking Catfish WP does not Granger Cause Walking Catfish RP 90 0.36 0.70Walking Catfish RP does not Granger Cause Walking Catfish WP 14.74 0.00Walking Catfish RP does not Granger Cause Walking Catfish FP 91 2.15 0.12Walking Catfish FP does not Granger Cause Walking Catfish RP 0.35 0.71Asian Sea Bass WP does not Granger Cause Asian Sea bass FP 63 0.44 0.64Asian Sea Bass FP does not Granger Cause Asian Sea bass WP 3.28 0.04Asian Sea Bass RP does not Granger Cause Asian Sea bass WP 62 0.76 0.52Asian Sea Bass WP does not Granger Cause Asian Sea bass RP 1.09 0.36Asian Sea Bass RP does not Granger Cause Asian Sea bass FP 64 0.24 0.87Asian Sea Bass FP does not Granger Cause Asian Sea bass RP 0.71 0.55Tilapia WP does not Granger Cause Tilapia FP 83 1.93 0.09Tilapia FP does not Granger Cause Tilapia WP 1.70 0.13Tilapia RP does not Granger Cause Tilapia WP 87 0.60 0.55Tilapia WP does not Granger Cause Tilapia RP 3.06 0.05Tilapia RP does not Granger Cause Tilapia FP 91 2.62 0.08Tilapia FP does not Granger Cause Tilapia RP 3.42 0.04

RP¼Retail Price, WP¼Wholesale Price, RP¼Retail Price, F-statistics in bold shows rejection ofhypothesis up to the 0.10 level of significance.


Dow

nloa

ded

by [

Uni

vers

ity o

f Pr

ince

Edw

ard

Isla

nd],

[D

r K

ehar

Sin

gh]

at 0

6:44

09

Mar

ch 2

015

Across the value chain, the Granger causality tests rejected the followingnull hypotheses up to 0.10 levels of significance (Table 3B):

i. Walking catfish retail price does not Granger cause Asian sea bass retailprice,

ii. Walking catfish retail price does not Granger cause tilapia retail price,iii. Walking catfish retail price does not Granger cause shrimp retail price,iv. Walking catfish farm price does not Granger cause Asian sea bass farm

price,v. Walking catfish farm price does not Granger cause tilapia farm price,

TABLE 3B Granger Causality Tests on Prices of Fish Species in Thailand across Value Chain

Null Hypothesis Obs F-Statistic Prob.

Sea Bass RP does not Granger Cause Vannamei Shrimp RP 64 1.32 0.28Vannamei Shrimp RP does not Granger Cause Asian Sea bass RP 0.21 0.89Walking Catfish RP does not Granger Cause Asian Sea bass RP 61 3.77 0.00Asian Sea Bass RP does not Granger Cause Walking Catfish RP 0.62 0.72Tilapia RP does not Granger Cause Walking Catfish RP 91 0.75 0.53Walking Catfish RP does not Granger Cause Tilapia RP 2.58 0.06Walking Catfish RP does not Granger Cause Shrimp RP 66 2.38 0.08Vannamei Shrimp RP does not Granger Cause Walking Catfish RP 0.76 0.52Tilapia RP does not Granger Cause Vannamei Shrimp RP 66 1.60 0.20Vannamei Shrimp RP does not Granger Cause Tilapia RP 1.31 0.28Tilapia RP does not Granger Cause Asian Sea bass RP 64 1.43 0.24Asian Sea Bass RP does not Granger Cause Tilapia RP 0.11 0.96Asian Sea Bass FP does not Granger Cause Vannamei Shrimp FP 65 0.43 0.65Vannamei Shrimp FP does not Granger Cause Asian Sea bass FP 0.78 0.46Walking catfish FP does not Granger Cause Vannamei Shrimp FP 78 0.65 0.58Vannamei Shrimp FP does not Granger Cause Walking Catfish FP 0.89 0.45Tilapia FP does not Granger Cause Vannamei Shrimp FP 79 1.89 0.16Vannamei Shrimp FP does not Granger Cause Tilapia FP 0.70 0.50Walking Catfish FP does not Granger Cause Asian Sea bass FP 64 2.47 0.07Asian Sea Bass FP does not Granger Cause Walking Catfish FP 0.43 0.73Tilapia FP does not Granger Cause Asian Sea bass FP 65 1.16 0.32Asian Sea bass FP does not Granger Cause Tilapia FP 1.13 0.33Tilapia FP does not Granger Cause Walking Catfish FP 91 0.77 0.47Walking Catfish FP does not Granger Cause Tilapia FP 3.92 0.02Asian Sea Bass WP does not Granger Cause Vannamei Shrimp WP 61 1.54 0.20Vannamei Shrimp WP does not Granger Cause Asian Sea bass WP 0.95 0.44Walking Catfish WP does not Granger Cause Vannamei Shrimp WP 74 1.52 0.19Vannamei Shrimp WP does not Granger Cause Walking Catfish WP 3.22 0.01Tilapia WP does not Granger Cause Vannamei Shrimp WP 74 1.01 0.39Vannamei Shrimp WP does not Granger Cause Tilapia WP 1.81 0.15Walking Catfish WP does not Granger Cause Asian Sea bass WP 63 0.26 0.77Asian Sea Bass WP does not Granger Cause Walking Catfish WP 2.75 0.07Tilapia WP does not Granger Cause Asian Sea bass WP 58 0.66 0.70Asian Sea Bass WP does not Granger Cause Tilapia WP 1.71 0.13Tilapia WP does not Granger Cause Walking Catfish WP 87 3.35 0.04Walking Catfish WP does not Granger Cause Tilapia WP 0.29 0.75

RP¼Retail Price, WP¼Wholesale Price, RP¼Retail Price, F-statistics in bold shows rejection ofhypothesis up to the 0.10 level of significance.

64 K. Singh et al.

Dow

nloa

ded

by [

Uni

vers

ity o

f Pr

ince

Edw

ard

Isla

nd],

[D

r K

ehar

Sin

gh]

at 0

6:44

09

Mar

ch 2

015

vi. Shrimp wholesale price does not Granger cause walking catfish whole-sale price,

vii. Asian sea bass wholesale price does not Granger cause walking catfishwholesale price; and,

viii. Tilapia wholesale price does not Granger cause walking catfish whole-sale price.

Therefore, we conclude that shrimp retail price, shrimp wholesale priceand walking catfish farm price are the Granger cause of shrimp farm prices.Shrimp wholesale and walking catfish retail prices are the Granger cause ofshrimp retail prices. Shrimp retail price is a Granger cause of shrimp whole-sale price. Walking catfish farm price, walking catfish retail price and Asiansea bass farm price are the Granger cause of shrimp farm price, Asian seabass retail price, and Asian sea bass wholesale price, respectively. Walkingcatfish farm price, walking catfish retail price, and wholesale prices ofshrimp, Asian sea bass and tilapia are the Granger cause of walking catfishwholesale price. Retail and wholesale prices of tilapia and walking catfishfarm price are the Granger cause of tilapia farm price, whereas farm andwholesale prices of tilapia and walking catfish retail price are the Grangercause of tilapia retail price. None of the price series considered is a Grangercause of walking catfish farm and retail prices, and tilapia wholesale price.

Our results suggest that prices in the Thai fish sector are not deter-mined at one end and then passed down or up along the supply channel.That is, pricing patterns in the Thai fish sector are not just cost or demanddriven. We found the direction of causality from retail to farm prices in van-namei shrimp; however, the direction of causality was also found fromwholesale to retail prices. In case of walking catfish, the pricing patternsare both supply and demand driven. The retail market shocks in case of tila-pia are directly transmitted to farmers, and vice-versa. The wholesale pricesof Asian sea bass adjust to shocks in farm prices; however, shocks in retailmarket remains confined to retail market. Tiffin and Dawson (2000), whilestudying the United Kingdom lamb market, found that lamb prices weredetermined in the retail market, and then passed upward along the supplychain. Goodwin and Holt (1999) and Goodwin and Harper (2000) foundthat retail market shocks were confined in retail markets for the most part,but farm markets adjusted to shocks in wholesale markets. However, Ben-Kaabia et al. (2002) found both supply and demand shocks were fullypassed through the marketing channel; i.e., they found complete pricetransmission. Saghaian (2007) found that beef price causality in the U.S.markets at different levels of the supply channel are bi-directional, influen-cing and being influenced by each other at each stage.

We have tested the cointegration along value chain for Asian sea bass;and at the retail level of value chain between Asian sea bass and tilapia.


Dow

nloa

ded

by [

Uni

vers

ity o

f Pr

ince

Edw

ard

Isla

nd],

[D

r K

ehar

Sin

gh]

at 0

6:44

09

Mar

ch 2

015

Other price series are either stationary at levels or trend stationary or thereis only one price series having unit root at farm/wholesale level of valuechain. Table 4 presents the results of the Johansen Cointegration test.The Trace and Eigenvalue statistics failed to reject the null hypothesis ofmaximum rank equal to ‘0’ in all other cases, which shows absence of coin-tegration between those price series.

Price Transmission Analysis

Equations (3.1) and (3.2) are a general model used to study the pricetransmission relations in Thai fish market. These models have AR-terms;

TABLE 4 Unrestricted Cointegration Rank Test: Trace and Maximum Eigenvalue

Maximum rank

Constant Constant þLinear Trend

Trace statistic Max Eigenvalue Trace statistic Max Eigenvalue

Asian Sea Bass: Farm and Wholesale Prices0 13.50 8.59 21.80 15.171 4.91 4.91 6.63 6.63Asian Sea Bass: Retail and Wholesale Prices0 18.27 13.38 19.63 13.561 4.89 4.89 6.07 6.07Asian Sea Bass: Retail and Farm Prices0 17.21 11.11 19.19 13.171 6.10 6.10 6.02 6.02Retail Prices: Asian Sea Bass and Tilapia0 15.71 14.09 20.62 15.241 1.61 1.61 5.38 5.38Critical Value 5%0 19.96 15.67 25.32 18.961 9.42 9.24 12.25 12.52

TABLE 5 Lag Length Selection for Auto-regression in Price Transmission Model

Price Series FPE AIC HQIC SBIC Selected Lag

Vannamei Shrimp Farm Price 1 1 1 1 1Vannamei Shrimp Wholesale Price 3 3 3 3 3Vannamei Shrimp Retail Price 2 2 2 2 2Walking Catfish Farm Price 3 3 3 1 3Walking Catfish Wholesale Price 1 1 1 1 1Walking Catfish Retail Price 2 2 2 2 2Asian Sea Bass Farm Price 2 2 1 1 2Asian Sea Bass Wholesale Price 1 1 1 1 1Asian Sea Bass Retail Price 3 3 3 3 3Tilapia Farm Price 1 1 1 1 1Tilapia Wholesale Price 2 2 2 2 2Tilapia Retail Price 1 1 1 1 1

66 K. Singh et al.

Dow

nloa

ded

by [

Uni

vers

ity o

f Pr

ince

Edw

ard

Isla

nd],

[D

r K

ehar

Sin

gh]

at 0

6:44

09

Mar

ch 2

015

therefore, it is necessary to decide the number of lags of AR terms. We haveselected the lags using FPE, AIC, HQIC and SBIC criteria (Table 5).

Walking CatfishTable 6A presents the estimates of Equation (1) for walking catfish

wholesale price and asymmetry price tests results. F-tests failed to reject allnull hypotheses of no difference up to 0.10 levels of significance exceptfor long run asymmetry test hypothesis for walking catfish retail prices, wherethe difference of the sum of positive change and negative change coefficientsis statistically significant at 0.07 levels. The long run elasticity (sum of coeffi-cients) of wholesale price with increasing retail prices (0.62) is significantlylower than decreasing retail prices (1.14). This means positive demandshocks in the walking catfish retail market are transmitted at a lower ratethan negative shocks to the walking catfish wholesale market in the long run.

TABLE 6A Estimates and Tests for Asymmetric Price Transmission in Walking Catfish Prices inThailand Along the Value Chain

Variable Lag

Coefficient

Sig. LevelSymbol Estimates

Dependent Variable: Cumulative Change in (D) Walking Catfish Wholesale PriceCumulative þ D Walking Catfish Farm Price 0 aþt�0 0.5874 0.0250

1 aþt�1 �0.0327 0.8760Cumulative - D Walking Catfish Farm Price 0 a�t�0 0.2120 0.2300

1 a�t�1 0.3379 0.1360Cumulative þ D Walking Catfish Retail Price 0 bþt�0 0.1652 0.6090

1 bþt�1 1.0967 0.01402 bþt�2 �0.6397 0.0600

Cumulative - D Walking Catfish Retail Price 0 b�t�0 0.5761 0.31201 b�t�1 1.6022 0.00902 b�t�2 �1.0324 0.0630

Constant k0 0.0214 0.5530Adjusted R-squared¼ 0.4344, Durbin-Watson Statistics¼ 1.8394

Testing for AsymmetryNull Hypotheses (H0) F-stat (df¼ 1, 77) Sig. Level

aþt�0 ¼ a�t�0 1.38 0.2438aþt�1 ¼ a�t�1 1.15 0.2862aþt�0 þ aþt�1 ¼ a�t�0 þ a�t�1 0.00 0.9575bþt�0 ¼ b�t�0 0.37 0.5436bþt�1 ¼ b�t�1 0.41 0.5232bþt�2 ¼ b�t�2 0.33 0.5682bþt�0 þ bþt�1 þ bþt�2 ¼ b�t�0 þ b�t�1 þ b�t�2 3.30 0.0731aþt�0 ¼ bþt�0 1.02 0.3158aþt�1 ¼ a b�t�1 0.31 0.5766aþt�0 þ aþt�1 ¼ bþt�0 þ bþt�1 þ bþt�2 0.07 0.7950a�t�0 þ a�t�1 ¼ b�t�0 þ b�t�1 þ b�t�2 1.27 0.2625

þ D¼positive change, � D¼negative change.


Dow

nloa

ded

by [

Uni

vers

ity o

f Pr

ince

Edw

ard

Isla

nd],

[D

r K

ehar

Sin

gh]

at 0

6:44

09

Mar

ch 2

015

Table 6B provides the estimates of the price transmission models forthe walking catfish farm, wholesale and retail prices. As stated earlier, wedid not find any of the price series along the value chain and across the spe-cies at the same level of value chain as a Granger cause for farm and retailprices (Tables 3A and 3B). Also these price series are trend stationary(Table 1), and lag length selection criteria showed optimum lag lengthof three for farm prices and lag length of two for retail price (Table 5).The estimated models show very low but positive trends in walking catfishfarm and retail prices (Table 6B). Both farm and retail current prices ofwalking catfish are positively influenced by its previous month prices andnegatively with two-month lagged price (Table 6B).

TABLE 6B Estimated Price Transmission Equations for Walking Catfish in Thailand

Variable Lag Coefficient Sig. Level

Dependent Variable: Walking Catfish Farm PriceWalking Catfish Farm Price 1 0.9165 0.0000

2 �0.2910 0.04503 0.2003 0.0660

Trend 0.0006 0.0230Constant 0.5455 0.0130Adjusted R-squared¼ 0.8744, Durbin-Watson Statistics¼ 1.9843

Dependent Variable: D Walking Catfish Wholesale PriceD Walking Catfish Wholesale Price 1 �0.1580 0.2060D Walking Catfish Farm Price 0 0.1971 0.1850

1 0.1239 0.3970D Walking Catfish Retail Price 0 0.0774 0.8440

1 1.2524 0.00202 �0.9776 0.0060

D Vannamei Shrimp Wholesale Price 0 �0.0523 0.68901 0.1178 0.43202 0.4051 0.02203 �0.3158 0.10104 0.1654 0.33805 0.0430 0.7720

D Asian Sea Bass Wholesale Price 0 �1.1133 0.01201 1.1883 0.0100

D Tilapia Wholesale Price 0 0.2052 0.00801 �0.0794 0.37102 0.0698 0.3730

Constant 0.0014 0.8520Adjusted R-squared¼ 0.4274, Durbin-Watson Statistics¼ 1.9822

Dependent Variable: Walking Catfish Retail PriceWalking Catfish Retail Price 1 1.3004 0.0000

2 �0.4159 0.0000Trend 0.0005 0.0220Constant 0.4204 0.0050Adjusted R-squared¼ 0.9719, Durbin-Watson Statistics¼ 2.0082

D¼ first difference.Note: All price series are in Natural Logarithmic form.

68 K. Singh et al.

Dow

nloa

ded

by [

Uni

vers

ity o

f Pr

ince

Edw

ard

Isla

nd],

[D

r K

ehar

Sin

gh]

at 0

6:44

09

Mar

ch 2

015

Walking catfish wholesale price series is influenced by its farm and retailprices, and also by vannamei shrimp, Asian sea bass and tilapia wholesaleprices. Asian Sea bass wholesale price has unit root, and walking catfishwholesale price is stationary at levels without trend (Table 1). Therefore,we have used model in difference without trend. The estimates of themodel (Table 6B) show that walking catfish farm price does not have anysignificant influence on its wholesale price, whereas its retail price affectedits wholesale price significantly. Walking catfish current month retail pricedoes not affect its current wholesale price, whereas one- and two-monthlagged retail price have positive (short run elasticity ¼ 1.25) and negative(short run elasticity ¼ �0.97), respectively, on walking catfish wholesaleprices. Two-month lagged vannamei shrimp wholesale price affects walkingcatfish current month wholesale price significantly (short run elas-ticity ¼ 0.40). Current Asian sea bass wholesale price has negative and pre-vious month has positive influence on walking catfish wholesale price. Onlycurrent month tilapia wholesale prices influence walking catfish wholesaleprices significantly. In nutshell a positive and a negative changes in currentmonth tilapia and Asian sea bass wholesale prices, respectively, lead to apositive change in current month walking catfish wholesale price. Thereverse is true for effects of previous month wholesale prices of tilapiaand Asian sea bass on current month wholesale price of walking catfish.

Vannamei ShrimpWe have presented the estimated price transmission asymmetry models

[Equation (1)] for vannamei shrimp farm, wholesale and retail prices inTable 7A, and the asymmetric price transmission hypotheses tests resultsin Table 7B. None of the estimated coefficients in vannamei shrimp farmprice model are statistically significant up to 0.10 levels of significance.However, in case of wholesale/retail price models, current price coeffi-cients of positive as well as negative cumulative changes in retail/wholesaleprices are significant, and magnitudes of coefficients are almost equal. Thisindicates absence of asymmetric price transmission in Thai vannameishrimp markets at farm, wholesale and retail levels of value chain. This isconfirmed by the hypotheses test results presented in Table 7B.

Vannamei shrimp farm, wholesale and retail prices are stationary atlevels without trend. Walking catfish retail price, which is trend stationaryat levels, is the Granger cause of vannamei shrimp retail price. At the samelevel of value chain, price of none of the species understudy is theGranger cause of vannamei shrimp farm and wholesale prices. The testresults showed the absence of asymmetric price transmission in Thai van-namei shrimp market along the value chain. Therefore, we have usedmodel given in Equation (3.1) (Table 7C) to work out price transmissionrelationships.


Dow

nloa

ded

by [

Uni

vers

ity o

f Pr

ince

Edw

ard

Isla

nd],

[D

r K

ehar

Sin

gh]

at 0

6:44

09

Mar

ch 2

015

One-month lagged price of vannamei shrimp has significant influenceon its current price at respective levels of value chain; however, degree ofinfluence is considerably higher at wholesale and retail levels than at farmlevel. Vannamei shrimp current wholesale price also affects vannamei

TABLE 7A Estimates for Asymmetric Price Transmission in Vannamei Shrimp Prices in Thailand

Variable Lag,

Coefficient


Dependent Variable: Cumulative Change in (D) Vannamei Shrimp Farm PriceCumulativeþD Vannamei Shrimp Wholesale Price 0 aþt�0 0.2824 0.1590

1 aþt�1 �0.1714 0.39602 aþt�2 �0.1417 0.48703 aþt�3 0.1190 0.2700

Cumulative�D Vannamei Shrimp Wholesale Price 0 a�t�0 0.2542 0.19201 a�t�1 0.1710 0.40402 a�t�2 �0.2196 0.30203 a�t�3 �0.0813 0.5000

CumulativeþD Vannamei Shrimp Retail Price 0 bþt�0 0.0166 0.93001 bþt�1 0.3601 0.06002 bþt�2 0.0610 0.7410

Cumulative�D Vannamei Shrimp Retail Price 0 b�t�0 0.1926 0.29301 b�t�1 �0.0964 0.61802 b�t�2 0.3200 0.1000


Dependent Variable: Cumulative Change in (D) Vannamei Shrimp Wholesale PriceCumulativeþD Vannamei Shrimp Retail Price 0 aþt�0 0.9203 0.0000

1 aþt�1 0.1137 0.31902 aþt�2 �0.1296 0.25403 aþt�3 0.0929 0.4040

Cumulative�D Vannamei Shrimp Retail Price 0 a�t�0 0.7814 0.00001 a�t�1 0.1819 0.15202 a�t�2 0.0755 0.53903 a�t�3 �0.0664 0.6040

Constant k0 �0.0243 0.5980Adjusted R-squared¼ 0.7538, Durbin-Watson Statistics¼ 1.8153

Dependent Variable: Cumulative Change in (D) Vannamei Shrimp Retail PriceCumulativeþD Vannamei Shrimp Wholesale Price 0 aþt�0 0.8564 0.0000

1 aþt�1 �0.0533 0.67802 aþt�2 0.2062 0.10703 aþt�3 �0.2159 0.0520

Cumulative�D Vannamei Shrimp Wholesale Price 0 a�t�0 0.8845 0.00001 a�t�1 �0.0423 0.74802 a�t�2 �0.1276 0.34103 a�t�3 0.1120 0.3760

Constant k0 0.0203 0.6440Adjusted R-squared¼ 0.7659, Durbin–Watson Statistics¼ 1.8440


70 K. Singh et al.

Dow

nloa

ded

by [

Uni

vers

ity o

f Pr

ince

Edw

ard

Isla

nd],

[D

r K

ehar

Sin

gh]

at 0

6:44

09

Mar

ch 2

015

shrimp farm price significantly; the short run price transmission elasticity ofvannamei shrimp farm price with respect to its wholesale price is very low(0.30). Current and one-month lagged vannamei shrimp wholesale/retailprices affect current vannamei shrimp retail/wholesale prices significantly.The long run price transmission elasticity of vannamei shrimp wholesale/retail price with respect to its retail/wholesale price is 0.84/0.77.

Asian Sea BassAll Asian sea bass price series have the unit roots (Table 1); however,

they are not cointegrated (Table 4). Asian sea bass farm price is theGranger cause of its wholesale price; the hypothesis of the Granger causalityis rejected in other price pairs of Asian sea bass along value chain. Keepingin view these results, we have estimated Equation (1) for Asian sea basswholesale price (Table 8A). The coefficient of current cumulative positivechange in Asian sea bass retail price is significant; however, the coefficient

TABLE 7B Tests for Asymmetric Price Transmission in Vannamei Shrimp Prices in Thailand

Null Hypotheses (H0) F-stat Sig. Level

Dependent Variable: Cumulative D Vannamei Shrimp Farm Price (df¼ 1, 49)aþt�0 ¼ a�t�0 0.01 0.9219aþt�1 ¼ a�t�1 1.23 0.2736aþt�2 ¼ a�t�2 0.06 0.7999aþt�3 ¼ a�t�3 1.27 0.2660aþt�0 þ aþt�1 þ aþt�2 þ aþt�3 ¼ a�t�0 þ a�t�1 þ a�t�2 þ a�t�3 0.02 0.8866bþt�0 ¼ b�t�0 0.42 0.5192bþt�1 ¼ b�t�1 2.48 0.1216bþt�2 ¼ b�t�2 0.89 0.3489bþt�0 þ bþt�1 þ bþt�2 ¼ b�t�0 þ b�t�1 þ b�t�2 0.01 0.9384aþt�0 ¼ bþt�0 0.53 0.4682a�t�1 ¼ ab�t�1 0.00 0.9731aþt�0 þ aþt�1 þ aþt�2 þ aþt�3 ¼ bþt�0 þ bþt�1 þ bþt�2 0.63 0.4310a�t�0 þ a�t�1 þ a�t�2 þ a�t�3 ¼ b�t�0 þ b�t�1 þ b�t�2 0.30 0.5851

Dependent Variable: Cumulative D Vannamei Shrimp Wholesale Price (df¼ 1, 54)aþt�0 ¼ a�t�0 0.54 0.4643aþt�1 ¼ a�t�1 0.14 0.7133aþt�2 ¼ a�t�2 1.28 0.2630aþt�3 ¼ a�t�3 0.72 0.4009aþt�0 þ aþt�1 þ aþt�2 þ aþt�3 ¼ a�t�0 þ a�t�1 þ a�t�2 þ a�t�3 0.33 0.5696

Dependent Variable: Cumulative D Vannamei Shrimp Retail Price (df¼ 1, 57)aþt�0 ¼ a�t�0 0.02 0.8801aþt�1 ¼ a�t�1 0.00 0.9555aþt�2 ¼ a�t�2 0.09 0.0921aþt�3 ¼ a�t�3 3.20 0.0791aþt�0 þ aþt�1 þ aþt�2 þ aþt�3 ¼ a�t�0 þ a�t�1 þ a�t�2 þ a�t�3 1.22 0.2746

df ¼degree of freedom for F-test.


Dow

nloa

ded

by [

Uni

vers

ity o

f Pr

ince

Edw

ard

Isla

nd],

[D

r K

ehar

Sin

gh]

at 0

6:44

09

Mar

ch 2

015

of current cumulative negative change in retail Asian sea bass price is non-significant up to 0.10 level of significance. The coefficients of lagged (one-and three-month lags) cumulative negative change in retail Asian sea bassprice are significant too. This means that if the Asian sea bass wholesalerspay higher prices (say 1%) to the farmers, they immediately receive higherprices (0.58%) from the Asian sea bass retailers.

However, if the wholesalers pay lower prices to the farmers, they do notpass the decrease to the retailers immediately. They pass around 20% ofdecreased price to the retailers in the next month and about 26% in thethird month. Less than 50% of decrease and 70% of increase in wholesa-lers’ purchase price is passed to the retailers in the long run. This indicates,and is confirmed by asymmetry hypotheses tests results (Table 8A),

TABLE 7C Estimated Price Transmission Equations for Vannamei Shrimp in Thailand


Dependent Variable: Vannamei Shrimp Farm PriceVannamei Shrimp Farm Price 1 0.4947 0.0000Vannamei Shrimp Retail Price 0 0.0649 0.5620

1 0.1416 0.25402 0.0704 0.5460

Vannamei Shrimp Wholesale Price 0 0.3054 0.00901 �0.1717 0.23402 �0.1408 0.29203 0.0523 0.4090


Dependent Variable: Vannamei Shrimp Wholesale PriceVannamei Shrimp Wholesale Price 1 0.8658 0.0000

2 �0.3884 0.02503 0.3469 0.0110

Vannamei Shrimp Retail Price 0 0.7778 0.00001 �0.4878 0.00102 0.1414 0.36503 �0.2840 0.0350


Dependent Variable: Vannamei Shrimp Retail PriceVannamei Shrimp Retail Price 1 0.9086 0.0000

2 �0.1737 0.2010Vannamei Shrimp Wholesale Price 0 0.8321 0.0000

1 �0.7700 0.00002 0.1482 0.36903 �0.0064 0.9290

Walking Catfish Retail Price 0 �0.2388 0.23701 0.4223 0.19102 �0.2511 0.1730


Note: All price series are in Natural Logarithmic form.

72 K. Singh et al.

Dow

nloa

ded

by [

Uni

vers

ity o

f Pr

ince

Edw

ard

Isla

nd],

[D

r K

ehar

Sin

gh]

at 0

6:44

09

Mar

ch 2

015

presence of short run as well as long run asymmetric price transmissionbetween Asian sea bass price in Thailand. It is to mention here that Asiansea bass production is mainly based on cage culture, which requires veryhigh investments. Asian sea bass farmers are well organized too. Retailershave very low, if any, control over prices.

We have estimated models in difference form presented in Equation(3.2) for sea bass farm, wholesale and retail prices (Table 8B). One-monthlagged sea bass retail and farm prices influence respective prices. One-month lagged farm price of walking catfish affects Asian sea bass farm price;the corresponding short-term price transmission elasticity is 0.19. Asian seabass current farm price is only factor that affects sea bass wholesale pricesignificantly; the coefficient of Asian sea bass current price in Asian sea basswholesale price transmission equation is 0.26. Three-month lagged walkingcatfish retail price has significant influence on sea bass retail price; thecorresponding coefficient is 0.95.

TilapiaTilapia wholesale and retail prices are the Granger cause of tilapia farm

price; so are the wholesale and farm prices of the retail price. Tilapia retailprice series have unit roots, whereas wholesale and farm price series arestationary. The results of the asymmetric price transmission model show thatsix month lagged cumulative positive change in wholesale price and five

TABLE 8A Tests for Asymmetric Price Transmission in Asian Sea Bass Prices in Thailand

Variable Lag

Coefficient


Dependent Variable: Cumulative Change in (D) Sea bass Wholesale PriceCumulative þ D Asian Sea Bass Farm Price 0 aþt�0 0.5874 0.0000

1 aþt�1 �0.0474 0.71402 aþt�2 0.2063 0.11403 aþt�3 �0.0886 0.4560

Cumulative - D Asian Sea Bass Farm Price 0 a�t�0 0.1183 0.17701 a�t�1 0.2005 0.04302 a�t�2 �0.1372 0.16503 a�t�3 0.2645 0.0030


Testing for AsymmetryNull Hypotheses (H0) F-stat(df¼ 1, 51) Sig. Level

aþt�0 ¼ a�t�0 8.18 0.0061aþt�1 ¼ a�t�1 1.79 0.1871aþt�2 ¼ a�t�2 3.32 0.0743aþt�3 ¼ a�t�3 4.55 0.0378

aþt�0 þ aþt�1 þ aþt�2 þ aþt�3 ¼ a�t�0 þ a�t�1 þ a�t�2 þ a�t�3 91.29 0.0000



Dow

nloa

ded

by [

Uni

vers

ity o

f Pr

ince

Edw

ard

Isla

nd],

[D

r K

ehar

Sin

gh]

at 0

6:44

09

Mar

ch 2

015

month lagged cumulative change in retail price affect tilapia farm price sig-nificantly (Table 9A); however, there is no evidence of asymmetric price trans-mission in tilapia markets along the value chain in Thailand (Table 9B).

The estimates of the price transmission model (Table 9C) for tilapiaretail price show that walking catfish retail price influences tilapia retailprice significantly (price transmission elasticity in current month ¼ 0.32,and long-run price transmission elasticity ¼ -0.06). Recent historical pricesaffect tilapia prices at all levels of the value chain.

CONCLUSIONS AND POLICY IMPLICATIONS

We have examined the presence of price transmission asymmetry alongthe value chain and price transmission across species in Thai fish market.This is an attempt to contribute to the horizontal and vertical price trans-mission in the seafood markets literature including the price transmissionasymmetry in the developing countries.

TABLE 8B Estimated Price Transmission Equations for Asian Sea Bass in Thailand


Dependent Variable: D Sea Bass Retail PriceD Asian Sea Bass Retail Price 1 0.7074 0.0000

2 �0.1671 0.31603 �0.1626 0.2160

D Walking Catfish Retail Price 0 0.5428 0.14501 0.0928 0.85802 �0.5254 0.32803 0.9517 0.04704 �0.0213 0.96105 �0.3462 0.2980


Dependent Variable: D Sea Bass Farm PriceD Asian Sea Bass Farm Price 1 0.2381 0.0590D Walking Catfish Farm Price 0 0.0376 0.6640

1 0.1914 0.02002 0.0492 0.5480

Constant 0.0002 0.9520Adjusted R-squared¼ 0.0766, Durbin–Watson Statistics¼ 2.0226

Dependent Variable: D Sea Bass Wholesale PriceD Asian Sea Bass Wholesale Price 1 0.0842 0.4960D Asian Sea Bass Farm Price 0 0.2639 0.0000

1 0.1074 0.1300Constant 0.0033 0.0830Adjusted R-squared¼ 0.3840, Durbin-Watson Statistics¼ 1.9870

D ¼change/first difference.Note: All price series are in Natural Logarithmic form.

74 K. Singh et al.

Dow

nloa

ded

by [

Uni

vers

ity o

f Pr

ince

Edw

ard

Isla

nd],

[D

r K

ehar

Sin

gh]

at 0

6:44

09

Mar

ch 2

015

We found unidirectional Granger causation in some cases and bidirec-tional Granger causation in other; however, in some of the cases the priceat one level of value chain is Granger caused by the prices at other levels ofvalue chain. Therefore, we have extended the Houck (1977) and Ward(1982) asymmetric price transmission model to consider two regressors,which allow the researchers to test the hypotheses “whether degree of

TABLE 9A Estimates for Asymmetric Price Transmission in Tilapia Prices in Thailand

Variable Lag

Coefficient


Dependent Variable: Cumulative Change in (D) Tilapia Farm PriceCumulative þ D Tilapia Wholesale Price 0 aþt�0 0.2639 0.2150

1 aþt�1 0.1071 0.68002 aþt�2 �0.1528 0.59103 aþt�3 0.0206 0.93704 aþt�4 �0.0701 0.77905 aþt�5 �0.2190 0.39306 aþt�6 0.4350 0.0620

Cumulative � D Tilapia Wholesale Price 0 a�t�0 0.2903 0.15901 a�t�1 �0.1105 0.62102 a�t�2 �0.0259 0.90403 a�t�3 �0.2455 0.24804 a�t�4 �0.0949 0.68005 a�t�5 0.4362 0.06206 a�t�6 �0.0570 0.7970

Cumulative þ D Tilapia Retail Price 0 bþt�0 0.4995 0.55701 bþt�1 0.4465 0.66602 bþt�2 �1.0667 0.2000

Cumulative � D Tilapia Retail Price 0 b�t�0 0.4902 0.59401 b�t�1 �0.7139 0.50502 b�t�2 0.7437 0.4040


Dependent Variable: Cumulative Change in (D) Tilapia Retail PriceCumulative þ D Tilapia Wholesale Price aþt�0 �0.0867 0.1190

aþt�1 0.0559 0.3500aþt�2 0.0771 0.1790

Cumulative � D Tilapia Wholesale Price a�t�0 �0.0439 0.3600a�t�1 0.0133 0.8180a�t�2 0.0219 0.7130

Cumulative þ D Tilapia Farm Price bþt�0 0.0327 0.5970bþt�1 0.0203 0.7520bþt�2 0.0877 0.1840

Cumulative � D Tilapia Farm Price b�t�0 0.0450 0.4450b�t�1 �0.0489 0.4020b�t�2 �0.0084 0.8910

Constant k0 2.0702 0.2080Adjusted R-squared¼ 0.0617, Durbin–Watson Statistics¼ 1.7371



Dow

nloa

ded

by [

Uni

vers

ity o

f Pr

ince

Edw

ard

Isla

nd],

[D

r K

ehar

Sin

gh]

at 0

6:44

09

Mar

ch 2

015

positive/negative changes in two regressors on the changes in the depen-dent variable are significantly different from each other or not both inthe short run and the long run”. We estimated the price transmission rela-tionships using regressors along the value chain and across the species atthe same level of value chain.

There is no evidence of short run asymmetric price transmission fromeither retail or farm level to wholesale level; however, there is weak evi-dence of long run asymmetric price transmission from retail to wholesaleprice. We did not find any evidence of asymmetric price transmission inThai fish market for vannamei shrimp and tilapia in short- and long run.Short run and long run price transmission asymmetry is evident in ThaiAsian sea bass market; wholesalers exercising some market power.

TABLE 9B Tests for Asymmetric Price Transmission in Tilapia Prices in Thailand

Null Hypotheses (H0) F-stat Sig. Level

Dependent Variable: Cumulative Change in (D) Tilapia Farm Price (df¼ 1, 48)aþt�0 ¼ a�t�0 0.01 0.9354aþt�1 ¼ a�t�1 0.29 0.5934aþt�2 ¼ a�t�2 0.09 0.7631aþt�3 ¼ a�t�3 0.48 0.4913aþt�4 ¼ a�t�4 0.00 0.9478aþt�5 ¼ a�t�5 3.07 0.0862aþt�6 ¼ a�t�6 2.23 0.1418

aþt�0 þ aþt�1 þ aþt�2 þ aþt�3 þ aþt�4 þ aþt�5 þ aþt�6

¼ a�t�0 þ a�t�1 þ a�t�2 þ a�t�3 þ a�t�4 þ a�t�5 þ a�t�6 0.53 0.4694bþt�0 ¼ b�t�0 0.00 0.9947bþt�1 ¼ b�t�1 0.44 0.5080bþt�2 ¼ b�t�2 1.64 0.2071bþt�0 þ bþt�1 þ bþt�2 ¼ b�t�0 þ b�t�1 þ b�t�2 0.29 0.5936aþt�0 ¼ bþt�0 0.07 0.7866a�t�1 ¼ ab�t�1 0.04 0.8356aþt�0 þ aþt�1 þ aþt�2 þ aþt�3 þ aþt�4 þ aþt�5 þ aþt�6 ¼ bþt�0 þ bþt�1 þ bþt�2 0.21 0.6503a�t�0 þ a�t�1 þ a�t�2 þ a�t�3 þ a�t�4 þ a�t�5 þ a�t�6 ¼ b�t�0 þ b�t�1 þ b�t�2 0.12 0.7310

Dependent Variable: Cumulative Change in (D) Tilapia Retail Price (df¼ 1, 60)aþt�0 ¼ a�t�0 1.24 0.2695aþt�1 ¼ a�t�1 0.13 0.7159aþt�2 ¼ a�t�2 0.56 0.4554aþt�0 þ aþt�1 þ aþt�2 ¼ a�t�0 þ a�t�1 þ a�t�2 0.00 0.9897bþt�0 ¼ b�t�0 0.02 0.8967bþt�1 ¼ b�t�1 0.50 0.4843bþt�2 ¼ b�t�2 0.89 0.3494bþt�0 þ bþt�1 þ bþt�2 ¼ b�t�0 þ b�t�1 þ b�t�2 0.88 0.3510aþt�0 ¼ bþt�0 1.91 0.1721a�t�1 ¼ ab�t�1 0.11 0.7396aþt�0 þ aþt�1 þ aþt�2 ¼ bþt�0 þ bþt�1 þ bþt�2 0.29 0.5916a�t�0 þ a�t�1 þ a�t�2 ¼ b�t�0 þ b�t�1 þ b�t�2 0.08 0.7765

df ¼degree of freedom for F-test.

76 K. Singh et al.

Dow

nloa

ded

by [

Uni

vers

ity o

f Pr

ince

Edw

ard

Isla

nd],

[D

r K

ehar

Sin

gh]

at 0

6:44

09

Mar

ch 2

015

In most of the cases, none of the species considered affect signifi-cantly prices of other species at the same level of value chain. The excep-tions to this are: i) walking catfish price affects tilapia price at retail levelin short as well as long run, ii) three-month lagged walking catfish retailprice affects Asian sea bass current retail price, iii) one-month laggedwalking catfish farm price influences Asian sea bass farm price, and iv)vannamei shrimp two-month lagged price, current tilapia price and cur-rent and one-month lagged Asian sea bass price significantly affect walk-ing catfish prices at wholesale level. In all these cases, the pricetransmission elasticities are positive except for long-run elasticity in casei (where it is negative but close to zero) and current month Asian sea basswholesale price in case iv where it is �1.11. These results indicate lack of

TABLE 9C Estimated Price Transmission Equations for Tilapia in Thailand


Dependent Variable: D Tilapia Retail PriceD Tilapia Retail Price 1 �0.2825 0.0090D Walking Catfish Retail Price 0 0.3259 0.0040

1 �0.1252 0.27902 �0.2847 0.0110

D Tilapia Wholesale Price 0 �0.0109 0.68401 �0.0175 0.5250

D Tilapia Farm Price 0 0.0193 0.48601 �0.0314 0.2630


Dependent Variable: Tilapia Farm Price*Tilapia Farm Price 1 0.5449 0.0000Walking Catfish Farm Price 0 0.1080 0.5550

1 0.2709 0.1390Constant 0.1026 0.7960Adjusted R-squared¼ 0.4635, Durbin-Watson Statistics¼ 1.9731

Dependent Variable: Tilapia Wholesale PriceTilapia Wholesale Price 1 0.8043 0.0000

2 �0.0458 0.6850Constant 0.7182 0.0030Adjusted R-squared¼ 0.5635, Durbin-Watson Statistics¼ 1.9922

D ¼first difference.Note: All price series are in Natural Logarithmic form.*Walking Catfish Farm Price, Tilapia Wholesale Price and Tilapia Retail Price are the Granger cause

of Tilapia Farm Price. However, we did not find significant fit for Tilapia Farm Price¼ f(Walking CatfishFarm Price, Tilapia Wholesale Price, Tilapia Retail Price, Autoregressive terms of Tilapia Farm Price).The null hypotheses that Tilapia Wholesale Price is a Granger cause of Tilapia Farm Price, and TilapiaRetail Price is a Granger cause of Tilapia Farm Price were rejected at 0.1 levels. Therefore, we droppedTilapia Wholesale Price and Tilapia Retail Price from the model, and re-run the model Tilapia FarmPrice¼ f(Walking Catfish Farm Price, Autoregressive terms of Tilapia Farm Price). Since Tilapia FarmPrice and Walking Catfish Farm Price at stationary at levels without a trend, therefore, we used a modelin levels.


Dow

nloa

ded

by [

Uni

vers

ity o

f Pr

ince

Edw

ard

Isla

nd],

[D

r K

ehar

Sin

gh]

at 0

6:44

09

Mar

ch 2

015

competition among different species in Thai seafood market. However,walking catfish faces some competition from tilapia in short run atwholesale level.

Price transmission relationships along the value chain shows thatwalking catfish retail prices (one-month and two-month lagged) influ-ence significantly its wholesale price in short run. Vannamei shrimp retailand wholesale prices affects each other in short run as well as long run.Vannamei shrimp’s current wholesale price also influences its currentfarm price. Asian sea bass current farm price affects its wholesale price.None of the prices along value chain in tilapia affect each othersignificantly.

The results of the study have important policy implications. Variousstudies (Dey et al. 2008a; Dey et al., 2008b) indicate that, given elasticincome elasticity of demand for fish, there will be tremendous increasein demand for various types of fish in Thailand over time due to popu-lation growth and increases in per capita income. Dey et al. (2008a) alsoindicates that fish exports from Thailand are expected to rise particularlyfor tilapia, cultured shrimp and high-value marine fish like Asian seabass. It is projected that consumer prices of the various species studiedwill rise faster than the posited inflation rate of 3.5% during 2005–2020, except for tilapia (with a yearly rise of 2.6%) (Dey et al., 2008a).The findings of no asymmetric price transmission of retail prices ofaquaculture products, indicating that increases in the retail price ofthe aquaculture products are likely to pass fully to the primary markets,are beneficial to aquaculture farmers in the country. In recent years,almost all increases in fish production have come from aquaculturesector. However, increasing fish supply from aquaculture will exert adownward pressure on prices of aquaculture products. But if marketprices fall due to the expansion of products, retailers might also be ableto easily pass through falling prices to farmers, and thereby farmers’ rev-enue might fall. Thus, there is a need to monitor the likely effect of aqua-culture expansion on farm prices. The aquaculture products should havea favorable market outlook to ensure economic viability of the con-cerned farm enterprises.

Aquaculture harvests are seasonal in nature. Like in other developingcountries, many fish farmers in Thailand are often forced to sell their pro-duce during the harvesting season because of perishable nature of the fish.If retail and/or wholesale prices drop due to some market phenomenon,farmers will have to sell their produces at that low price, i.e., fish farmersbecome price takers. This signifies the importance of better storage facili-ties and transport infrastructure in rural markets. Policies that encouragesmall-scale farmers to form collective arrangement for marketing will behelpful.

78 K. Singh et al.

Dow

nloa

ded

by [

Uni

vers

ity o

f Pr

ince

Edw

ard

Isla

nd],

[D

r K

ehar

Sin

gh]

at 0

6:44

09

Mar

ch 2

015

ACKNOWLEDGMENTS

The University of Arkansas at Pine Bluff provided staff time for some ofthe authors. The Canada Excellence Chairs Program supported KeharSingh’s time. The authors are thankful to the anonymous reviewers fortheir comments/suggestions.

FUNDING

The primary sponsors of the project were the Norwegian Agency forDevelopment Cooperation (Norad) and the Food and Agricultural Organi-zation of the United Nations (FAO).

REFERENCES

Aguiar, D.R.D. & J.A. Santana (2002) Asymmetry in farm to retail price transmission: evidencefrom Brazil. Agribusiness, 18(1), 37–48.

Asche, F., L.S. Bennear, A. Oglend & M.D. Smith (2012) U.S. shrimp market integration. MarineResource Economics, 27(2), 181–192.

Asche, F., R.E. Daho, D. Valderrama, & D. Zhang (2014) Price transmission in new supply chains-The case of salmon in France. Aquaculture Economics and Management, 18, 205–219.

Asche, F., A.G. Guttormsen, T. Sebulonsen, & E.H. Sissener (2005) Competition between farmedand wild salmon: the Japanese salmon market. Agricultural Economics, 33, 333–340.

Asche, F., S. Jaffry, & J. Hartmann (2007) Price transmission and market integration: vertical andhorizontal price linkages for salmon. Applied Economics, 39(19), 2535–2545.

Ben-Kaabia, M., O.M. Gill, & L. Boshnjaku (2002) Price transmission asymmetries in the Spanish lambsector. Paper presented at the Xth Congress of European Association of Agricultural Econo-mists, Zaragoza, Spain, 28–31 August.

Bhargava, A. (1986) On the theory of testing for unit roots in observed time series. Review of Econ-omic Studies, 53, 369–384.

Department of Fisheries (2013) Fisheries Statistic of Thailand 2011. Information Technology Cen-ter, Department of Fisheries, Ministry of Agriculture and Cooperatives, Bangkok.

Dey, M.M., R.M. Briones, Y.T. Garcia, A. Nissapa, U.P. Rodriguez, R.K. Talukder, A. Senaratne, I.H.Omar, S. Koeshendrajana, N.T. Khiem, T.S. Yew, M. Weimin, D.S. Jayakody, P. Kumar, R.Bhatta, M.S. Haque, M.A. Rab, O.L. Chen, L. Luping, & F.J. Paraguas (2008a) Strategies andoptions for increasing and sustaining fisheries and aquaculture production to benefit poorer householdsin Asia. WorldFish Center Studies and Reviews No. 1823. WorldFish Center, Penang, Malaysia.

Dey, M.M., Y.T. Garcia, P. Kumar, S. Piumsombun, M.S. Haque, L. Li, A. Adam, A. Senaratne, N.T. Khiem, & S. Koeshendrajana (2008b) Demand for fish in Asia: A cross-country analysis.Australian Journal of Agricultural and Resource Economics, 52(3), 321–338.

Dickey, D.A. & W.A. Fuller (1979) Distribution of the estimators for autoregressive time serieswith a unit root. Journal of the American Statistical Association, 74, 427–431.

Elliott, G., T.J. Rothenberg, & J.H. Stock (1996) Efficient tests for an autoregressive unit root.Econometrica, 64, 813–836.

Engle, R.F. & C.W.J. Granger (1987) Cointegration and error correction: representation, esti-mation and testing. Econometrica, 55, 251–276.

Frey, G.F. & M. Manera (2007) Econometric models of asymmetric price transmission. Journal ofEconomic Surveys, 21(2), 349–415.

Garcia, J.G. (2006) The Spanish hake market: price, product and market relationships. Master’s Thesis,University of Barcelona, Barcelona, Spain.


Dow

nloa

ded

by [

Uni

vers

ity o

f Pr

ince

Edw

ard

Isla

nd],

[D

r K

ehar

Sin

gh]

at 0

6:44

09

Mar

ch 2

015

Gonzalo, J. (1994) Five alternative methods for estimating long-run relationships. Journal of Econo-metrics, 60(1/2), 203–233.

Gonzales, F., P. Guillotreau, L.L Grel, & M. Simioni (2013) Detecting asymmetric price trans-mission with consistent threshold along the fish supply chain. Canadian Journal of AgriculturalEconomics, 61, 37–60.

Goodwin, B.K. & D.C. Harper (2000) Price transmission, threshold behavior, and asymmetricadjustment in the U.S. pork sector. Journal of Agricultural and Applied Economics, 32(3),543–553.

Goodwin, B.K. & M.T. Holt (1999) Price transmission and asymmetric adjustment in the U.S. beefsector. American Journal of Agricultural Economics, 81(3), 630–637.

Granger, C.W.J. (1969) Investigating causal relation by econometric and cross-sectional method.Econometrica, 37(3), 424–438.

Guillen, J. & R. Franquesa (2008) Price transmission and market power analysis in the Spanishseafood market chain. Espai de Recerca en Economia Working Papers in Economics No.190. Retrieved from http://www.ere.ub.es/dtreball/E08190.rdf/at_download/file

Hatemi-J, A. (2012) Asymmetric causality tests with an application. Empirical Economics, 43(1), 447–456.

Houck, P.J. (1977) An approach to specifying and estimating non-reversible functions. AmericanJournal of Agricultural Economics, 59(3), 570–572.

Jaffry, S. (2005) Asymmetric price transmission: a case study of the French hake value chain. Mar-ine Resource Economics, 19(4), 511–523.

Jimenez-Toribio, R., J.J. Garcia-Del-Hoyo, & F. Garcia-Ordaz (2003) Vertical integration and pricetransmission in the Spanish distribution channel of the striped venus. XV EAFE Conference,Ifremer, Brest, FRANCE 15–16 May 2003. Retrieved from http://s3.amazonaws.com/zanran_storage/www.ifremer.fr/ContentPages/6626367.pdf

Jimenez-Toribio, R., P. Guillotreau & R. Mongruel (2010) Global integration of European tunamarkets. Progress in Oceanography, 86(1/2), 166–175.

Johansen, S. (1988) Statistical analysis of cointegration vectors. Journal of Economic Dynamics andControl, 12(2/3), 231–254.

Kwiatkowski, D., P.C.B. Phillips, P. Schmidt, & Y. Shin (1992) Testing the null hypothesis of sta-tionarity against the alternative of a unit root. Journal of Econometrics, 54(1/3), 159–178.

Larsen, T.A. & H. Kinnucan (2009) The effect of exchange rates on international marketing mar-gins. Aquaculture Economics and Management, 13(2), 124–137.

Matsui T., T. Nakajima, Y. Sakai, & N. Yagi (2011) Structural change and price transmission inJapanese marine products market: the case of blue fin tuna. Kokusai Gyogyo Kenkyu, 10, 1–13.

Meyer, J. & S.von Cramon-Taubadel (2004) Asymmetric price transmission: a survey. Journal ofAgricultural Economics, 55(3), 581–611.

Nakajima, T., T. Matsui, Y. Sakai, & N. Yagi (2011) Capturing changes in asymmetric price trans-mission: a rolling window tar estimation using bluefin tuna case study. Journal of InternationalFisheries, 10, 1–18.

Ng, S. & P. Perron (2001) Lag length selection and the construction of unit root tests with goodsize and power. Econometrica, 69(6), 1519–1554.

Nielsen, M. (2004) International market integration and demand: an analysis of the Norwegianand Danish herring market. Food Economics, 1(3), 176–184.

Nielsen, M. (2005) Price formation and market integration on the European first-hand market forwhitefish. Marine Resource Economics, 20(2), 185–202.

Nielsen, M., J. Setala, J. Laitinen, K. Saarni J. Virtanen, & A. Honkanen (2007) Market integrationof farmed trout in Germany. Marine Resource Economics, 22(2), 195–213.

Nielsen, M.J.Smit, & J. Guillen (2009) Market integration of fish in Europe. Journal of AgriculturalEconomics, 60(2), 367–385.

Norman-López, A. (2009) Competition between different wild and farmed species: The US tilapiamarket. Marine Resource Economics, 24(3), 237–251.

Norman-López, A. & F. Asche (2008) Competition between imported tilapia and US catfish in theUS market. Marine Resource Economics, 23(2), 199–214.

Peltzman, S. (2000) Prices rise faster than they fall. Journal of Political Economy, 108(3), 466–502.

80 K. Singh et al.

Dow

nloa

ded

by [

Uni

vers

ity o

f Pr

ince

Edw

ard

Isla

nd],

[D

r K

ehar

Sin

gh]

at 0

6:44

09

Mar

ch 2

015

http://www.ere.ub.es/dtreball/E08190.rdf/at_download/file

http://s3.amazonaws.com/zanran_storage/www.ifremer.fr/ContentPages/6626367.pdf

http://s3.amazonaws.com/zanran_storage/www.ifremer.fr/ContentPages/6626367.pdf

Phillips, P.C.B. (1987) Time series regression with unit roots. Econometrica, 55(2), 277–301.Phillips, P.C.B. & P. Perron (1988) Testing for a unit root in time series regression. Biometrika,

75(2), 335–346.Saghaian, S.H. (2007) Beef safety shocks and dynamics of vertical price adjustment: the case of

BSE discovery in the U.S. beef sector. Agribusiness, 23(3), 333–348.Tiffin, R. & P.J. Dawson (2000) Structural breaks, cointegration and the farm retail price spread

for lamb. Applied Economics, 32(10), 1281–1286.Toda, H.Y. and T. Yamamoto (1995) Statistical inference in Vector Autoregressions with possibly

integrated processes. Journal of Econometrics, 66(1-2), 225–250.Vavra, P. & B.K. Goodwin (2005)Analysis of price transmission along the food chain. OECD Food,

Agriculture and Fisheries Working Papers, No. 3, OECD Publishing doi: 10.1787/752335872456. Retrieved from: http://www.oecd.org/agriculture/agricultural-policies/40459642.pdf

Vinuya, F.D. (2007) Testing for market integration and the law of one price in world shrimp mar-kets. Aquaculture Economics and Management, 11(3), 243–265.

Ward, R.W. (1982) Asymmetry in retail, wholesale and shipping point pricing for fresh vegetables.American Journal of Agricultural Economics, 64(2), 205–212.

APPENDIX 1 Price Data and Summary Statistics

Species Level Time periodAverage

(Baht/kg) S.D. C.V.

Asian Sea Bass Farm Jan 05 – July 10 108.75 9.59 8.82Wholesale Jan 05 – May 10 116.48 10.65 9.14Retail Jan 05 – July 10 140.84 23.92 16.98

Hybrid Walking Catfish Farm Jan 03 – Sep 10 26.98 3.08 11.41Wholesale Jan 03 – Aug 10 30.12 3.15 10.46Retail Jan 03 – Oct 10 46.45 6.33 13.62

Tilapia Farm Jan 03 – Sep 10 19.47 3.11 15.98Wholesale Jan 03 – May 10 29.25 4.57 15.61Retail Jan 03 – Oct 10 37.91 4.64 12.25

Vannamei Shrimp(50 pcs/kg)

Farm Jan 05 – Sep 10 119.28 14.68 12.31Wholesale Jan 05 – Sep 10 135.58 18.31 13.50Retail Jan 05 – Sep10 215.48 19.24 8.93


Dow

nloa

ded

by [

Uni

vers

ity o

f Pr

ince

Edw

ard

Isla

nd],

[D

r K

ehar

Sin

gh]

at 0

6:44

09

Mar

ch 2

015

http://www.oecd.org/agriculture/agricultural-policies/40459642.pdf

http://www.oecd.org/agriculture/agricultural-policies/40459642.pdf

Price Transmission in Thai Aquaculture Product Markets: An Analysis Along Value Chain and Across Species

Documents