Indian Seafood Exports: Issues of Instability, Commodity ...

-- - - - -.- - -- --
Indian Seafood Exports: Issues of Instability, Commodity
Concentration and Geographical Spread
C. Sarada, T. Ravisankar, M. Krishnan and C. Anandanarayanan*
INTRODUCTION
Fluctuations in fish production from capture and culture, varlalions 111
lnternat~onal prices. adjustments in exchange rates and finally - the var~ablc values 01'
export earnings are grlm concerns for developing countries of South Asia and Soul11
East Asia. Most of the developing countries (including India) obtain thu~r niaj~,r
share of export earnlngs from selected few items or commodities and the trade is
concentrated with a couple of nations. These developing coutitrics as commodity
producers and exporlers have hardly any instruments at their disposal to licdge
ag,ainst the adverse effects of instabiltty of export earntngs. Managing commodity
prlcu risk and stabilising export earnings are still important policy issues for virtually
all low Income and commodity exporting countrlcs (Pal. 1992).
Imperative effects of variability of export earnings at the macroeconomic level
are, disruptions in the investment planning process, misallocation ol'rcsou~ccs and
disturbances of the internal balance of public finances, impacts on tlic ratc of
domestic savings, increastng internal and external indebtedness and problen~s ill tl~c
balance of payments, which might result in unstable earnings and discourage fanners
liom producing for export and can lead to a "further" fall in export earnlngr and gross
National Product (GNP). The export oriented producers who suffer from eartilng
shortfall, will also cut back their consumption, which affects the puhlic finances I'hc
perslstencc of abnormally depressed prices globally dur~ng the 1980s has also
resulted in a sharp reduction in the living standards of developing coutitries 'l'hu> the
export-earning instability has to be considered as a development probleni becaurc it
dampens the growth rate, particularly as a result of its negative effect on product~vity
of capital. Considering the strengths of Indian fisheries sector and steady
development in the past four decades, the causes of instabilities in production and
*Sctnl~,t(SS). Smwr Sc~mt~rt. Prtnctpat Sclcntlrl nnd Ilcad, Teehn~eill OWmr, rrspccl~vely boccd 5imre$
Dlvl,lon. Ccnual Inn~lule ol HracktvhwaVr Aquatullurc (CIHAI. It,dunc!l 01 Agrzcullural Rcrcarcb. Chcnsl, -
bO0 028
The avthon .re gatcful to Ihr tchrca for ~aluahlc wggcrllons lurlhcr. Ihc authors arc thankful lri A (I
Ponnlah. Dlnctor. CIRA, Chennsl md P Ravtchandran. Ibvrmcr i)crccliir fir, CIIIA, Chonns~, lhr lhclr
encoutlgcmcnt

INlllAlr 51 AFCIOD LXPORIS ISStII 501 INII*Lltl lTY.COMMOI>tlY CONCI NrK4IION 210
cxpon of sealhod assume immense Importance in the ccanamic scenarto of India and
hence the present stud> was conducted wlth the follou~ng ohjectivcs: (1) 7'0 compute
concelllralion measures such as commodity concentrdtlon and geographic concentration
coefficients (br lndian seafood exports. (2) lo estimate the ~nstability indices
for seafood exports from India and (3) '1 o determine the s~gnifica~~t factors that aflbct
the instability in xafood exports from lndla
Tlic paper is arranged as fbllows Sectlo11 I1 d~scusses the structure and pattern of
firlieries dcvelop~nent In lndia durlng the past four decndes The tli~rd rectlon deal,
\*1t11 data and transformation prc~cedl~rcs Sect~on IV descrthcs the theoretical and
computational framework. Sectlon V prcscnts the resullr and thc final sectloll gives a
summary of findings and d~scusses pollcy alternat~ves.
With an annual fish production of 6 ~n~lll~on cnctrlc tonnes India occupies the
fourth position in fish production and second III aquaculture productio~i globally The
annual domestlc per capita fish availability is 9 kilograms and seafoud cnpurt
earnlngs of lndia is consistently over K\ 6,000 crorc* a ycar I:ish cuntrabuter 1.4 per
ccnt of gross domest~c prr,doct (Gl)P) i~nd 4 5 per ccnt ofthe agr~cult~lral 601' 'I ill
the close of 1960 the ehpOIi ot Illdlan .;eak,od products iilalnly c

240 INDIAN JOUKNALOt AGRICULTURAL ECONOhlCS
export liom a few million US$ in 1961-62 to US$ 1330.76 million in 2003-04.
accounting for approximately 3.32 per cent of the total exports from India. During the
eighties, the canned items slowly disappeared and frozen items gained prominence in
India's seafood trade. Amongst the frozen items also, there were changes in the
demand for differentiated products from various countries. While Japan ind~cated
their preference for headless sliell on shrimp, the lJ S.A. demanded peeled shrimp
meat and the European countrtes preferred the IQF shrimp in frozen and cooked
Ibrm. The Europeali market also absorbed the major share of cephalopods while
Japdn had taken a small share of' it. 'These frozen fish items had greater demand in the
buuth East Astan countries as well as In the M~ddle East In the 1970s. the export was
depend~ny mainly o ~i shrimp hut due to the expon promotional measures, it hecame
possible to d~vrrbify the products In the cighties adding cephalopods (cultlefish, squid
and octopus) and frozen fish (such as pomfiet, rihhon fish. seer fish, mackcrcl. reef
cod. croakers, snapper, etc). While all these items hold good prospects, l~ve fish,
chilled fresh water fish etc arc promising ~tc~ns for the future (Ayyappan and
Krishnan, 2005). Duc lo the introduction of scientific shrimp fanning, the export of
fiozen value added shrlmp continued as the major foreign exchange earner among
seafood products and the volume of fiozen shrimp exported during 2003-04 was
1.29.768 metric tonnes (Tahle 2)
111
DATA AN11 Mt I HOPS
The value of total seafood cxpons from India from 1981-82 to 2003-04 were
obtained from various issues of Hevieus of Marrne 1'ruduct.s Erporb published by
Marltie Products txports Development Authority (MPEDA). Cochin. Indla. The data
on shrimp product~on were obtained from MPEDA Data regarding GDP, F~slieries
Gross Doniestic Product (IYGDP) was obtained from Central Statistical
Organisation's web site \\\\I< n11,hpl IC In. GDP excluding Fisher~es (NFCDP) was
coniputed by subtracting FGDP from GDP.
Mr.r~sure.\ oj Co'o,,cw~rrrrlmn
Several concentration measures are ava~lahle In the literature, e.g., Tongan
(1994). Tepegne (2000). Erlat and Akyuz (2001). Campa and Fernandes (2004) used
different measures for different studies. l.he appropriateness of a measure depends on
the data on which the estimates are based and with the purpose of analysis. ~ h c
present study utilises the most widely employed Gini-Hirschman coefficient of
concentration, which defines the degree of concentration in a country's export as

1hl)lAN C L 4 f t XPORTS ISSlJtOF INFIAHILITY COMMI)UIIY l'ON('IN~KA1ION 241

Where G - Cornnlodity concentratton coeffic~ent (CC)/Geographic concentration
coefficient (GC), X,, - Export earnings of commodlt). group i in year titxport
earnings from country i in year t and X, = '1 otal export earnings in year 1.
A varlety of instability ind~ces are available in the literature from simple to
complex. The variance of expon grov'th 15 the simplest measure of export instability.
But &ng tn lluctuat~otls in export volulnes and values, deviations from trend in
exports could be more ideal measure of export ~nstability Varlous corrections for
trend are available in the Ilterature, "11,!moving avcragcs, llnear and exponential
trends (Pinsuwana, 1991, Rhat and Nlrmala. 2001. Drvkota, 20041. I:.ach of these
methods has its own advantages and d~sadvantages. The present study anenlpts to
compute the Instability of ehparts hy using the measme, based on the avcrage percent
devlatlon of the ohscrved values pmcecd from an exponcntial path (Paudyal, 1988).
'The instability index (11) can he expressed by the formula.
Whcre 11 = Instability indcx, Xt = Est~mdtcd trend \aloe, X, = Actual value. X =
Mean of the actual valuc.
'The literature on the determinants of export instability is well established. The
deternittlants are commodity concentratton (CC), geographical concentration (GC),
the ratio of food and raw lnatcrials lo lotal exports, per capita income of exporting
country, openness of the economy and export shares in world trade, etc. The
empirical evidence on Ihc relationsh~p between these determ~nant~ and export
instability appears inconclusive. Studies like Massell (1970) and O'Brien (1972)
show no significant link between expon instability and its alleged determinants.
However. studies like Paudyal (1988), and Tegegne (2000) showed that these
determinants do affect export ~nstah~lity. The cross-country analysis by earlier studies
implicitly assume a unique relationship between a given ehplanatoV variable and the
degree of export instability across the countries. Thus, estimates using cross section
data to find the average relationshtps does not provlde much lnfomation on the

oliaviour of producers of specific commod~tlcs 111 thc choset~ cuuntrlcs. Only few
studics such as Love (1992). Wilson (1994). Slnha (1999). Tegegne (2000). used time
series analysis on an indwidual country hasis Hut, most ofthe ava~lahlc time series
studies do not address the Issues of non-statlonary nature of data. Hmce it could not
he ruled out that these estilnates are estimated from spurious rcgressioti. Mullor-
Sebast~an (1988) argued that studlcs, which lump together the exports of all goods.
are misleading because export instahllity II~ a gtven product is influenced hy the
characteristics of the indiv~dual product and degree of development of the exporting
country. Accordingly. the present study colifi~ie itself to instability in llidia~l seafood
exporl earnings and uses thc tltne serles data after considering the non-statlonary
nature oftlic data.
I:arlier statistical cv~denceb conclude that illstability index of exports are largcly
associated with the dcgrcc of commodity cu~ice~~tration of cxports, per capita it~con~c
and with the concentration of exportr by eeoeraphical area of dest~nation (Paudyal.
1988. Sinha, 1999. 'I egegtle. 2000). I'hus, instahil~ty index of seafood ehports
earnings could he expressed as a funct~oti of commodity concentratloti (CC).
Geographic Conce~itratiot~ ((iC) and Instability ill countp's GDl' ~ h reflects ~ h ~CI.
capita lncomc ofthe chporting country.
IISFFX f(C'C. GC, Il(il)P) .. . (7)
To separate out the effects of the fishcries GDP on instability of scafood expons
equation 3 is redefined as follows
IISFEX = f (CC. GC, IIFGDP. IINFGDP) .... (4)
Wliere IIFGDP represents lnstabillty index of fisheries GDI' and IINFGDP Instability
index ofnon-fisheries GVP.
'I'egegne (2000) argued that apan from the other key determinants of exports
income fluctuations, the relative Importance of major commodity, global demand
conditions affecting the major commodity, internal supply condit~oti should also be
considered. Among seafood exports, frozen shrimp is the single largest expon item.
The proportion of exports from cultured shrimp production has kept rising, implying
that ally instability caused ill seafood cxports could be mainly due to fluctuations in
the cultured shrimp production Thus to capture the flucluations In seafood exports,
cultured shrimp production is also cons~dercd as one of the determinants Thus
equation 4 can he written as
IISFEX = f (CC, GC, IIFGDP, IINFGDP, IISHPR) ....(5)

244 NDlAN JOlJKNALOT AGRICULTURAL kCONOMICS
(.'~~mpu/u/ronol Framework
To establish cause and effect relationship between major determinants and
seafood export instabil~ty a double log-linear function is preferred because of ease of
tliterpretation and best lit Thus the long-run seafood export instab~lity function is
specified as follows
liisfex =Pa+ PIIcc+ p,Igc+ p,liifgdpi P, liinfgdp+ p, liishpr+ E , ... .(6)
Where, l = Natural logarithm, ii = Instability Index, sfex = Indian .seafood exports, cc
.- Commodity Concentration, gc = Geographic Concentration, fgdp = Fisheries
GlIP, nfgdp = non-fisheries GDP, shpr = Cultured shrimp production, E, = error
term assumed to be identical and independent and E - N(o,o').
Po, PI. I) ,P, IDd. PI are the coefficients to be estimated.
Estimat~ng long-run relatiollsliip such as in equation (6) is likely to pose some
prohlcms because thc valiables in the analysis typically exhibit multicollinearity and
tion-stationarity. 'Thcsc problenis arc oRcn dealt with by tak~ng first differences of
the va~iahlcs, before any estimations are done, to mahe the series stationary. This
procedure of differencing has a ~najor drawback: 11 eliminates all ~nformation about
the Ikmg run relatiomhip and thcreby only short-run effects was explained (Maddala,
1092). Thus the modelling ol'srafood exports ~nstab~lity should be based on lnethods
which cxplicltl) take the non-stationarity features of the data into account. The theory
of cointegration techniques and vector error correction lnodels (VECM) addresses
these issues in an efficient and s~gll~ficant manner (Engle and Granger. 1991) which
i~npl~es that cointegration tests are superior when investigating the relationships that
are believed to be ofa long run nature.
There are several advantages In using VEC'M. I'irstly, the VECM approach treats
the variables as determined wlthin the salne system, without a prim assumptions
about the nature of interrelationship. Secondly, it clearly distinguishes between longrun
and short-run effects since both levels and first differences of the variables enter
the VECM. Thirdly, the speed of adjustment towards the long-run relationship can be
directly estimated. Finally, the VECM has a sound statistical foundation based on the
theory of cointegration developed by Engle and Granger (1987). Despite the above
advantages these models can become easily over-parameterised, as each variable is
allowed to affect the other variable at a number of lags. The results can also be
sensitive to the chosen lag length, although there are significance tests that can be
used to detennine the appropriate number of lags to be chosen. Thus transformation
of equation 6 by incorporating error correction (EC) term can be represented as
follows.

INDIAN SEAFOOD EXPORTS ISSllES OF INS1AHII.II Y. ('OMMODITY CONCFNTHATION 213
Aliisfex =~,,+fi,fblcc,., t ~ ,f Alngc,, + fi,f~lliif~d~,,, +p,fAli!nfgdp,,
to, i~ltish~r,, +BECM(-I)+ c,
. . , (7)
Whcre A is the difference operator. ECM(-I) is error-correction term lagged by one
period In the cotntegrating regression, for integrating short term dynamics tn the
long-run seafood export functton, This functlon allows to estimate short-run
relationships betwccn Instability index of Indian seafood exports and its
determinants, e,, the error term follows normal ltidependent and identically
distributed (i.i.d) properties. The coefficient 6 measures the response of instability
Index of seafood expons In each prriod from the long-ru11 equiltbrium with the
co~ntegration equation normaliscd on ~yfex. The coefficient 6 represents the
propartton of the d~sequllihrium in iisfex in one period corrected in the next period 6.
is ehpected to have a negative sign and be statistically significant.
The modelling strategy adopted to estimate VECM involvcs three steps
Step I Tesrjor Srarronorrh
Before conducting cointegration tests, it is necessary to establi5h the univariate
time series properties of the variables lo confirm all the variables are non-stationary
and integrated of the same order. This is performed by unit-root test, viz., Augmented
Dickey-Fuller (ADF) test. This test fitids out the order of integration, which is the
minimal number oftimes a series has to be differenced until 11 becomes stationary.
Step 2: Derermrnal~on of Oprimum Lag Lengrh
The cointcgration test is based on vector auto regression and is sensitivc to the
number of lags included in the model, therefore first we should determine the optimal
number of lags used in the cointegration test One way to determine the number of
lags is to select the model with minimum information criterion which are based 011
log-likelihood and penalise the inclusion of additional regressors (Greene, 1993). The
study utilises Akaike ~nformation criterion (AIC) to choose the optimum lag length
Step:3 Coinregrarion Tesr and VECM
The purpose of cointegration test is to determine whether a group of nonstationary
series is cointegrated or not. The presence of cointegration enables to
form a vector error correction mechanism to analyse both the short and long-run
nlationship among cointegrated series.

246 INDIAN JOIIKNAI.OF AORICIII.TURN. ECONOMIC
'There are a number of alternative cointepratlon tests (Engle and Granger 1987;
Stock and Watson. 1993 and Johansen, 1988 and 1995). The common objective of
these tests is to determine if there exists a long-run relationship among all variables.
Johansen cointegration procedure has properties which are superior to alternative
methods of cointegration testing. This method treats all the variables as endogenous
and take care of the endogenetty ~rohlem by prov~ding an estimation procedure that
does not require arbitrary cho~ce of a variable for normalisation. It also allows tests
for tnultiple cointegrations (Chosh. 2003). Even though this methodology is quite
complex. the underlying inference is straightforward. In order to find the possible
cointegrating vectors the data is div~ded into two groups. the variables in their levels
and thelr first differences, using the technique of canonical correlation, the linear
combinations are stationary atid thus so are the cointegrating vectors (Jaffry el al..
1998) The present study empir~cally evaluates cointegration bchveen variables by
utilising MI. method of cointegration developed by Sohansen (1988, 1995) and
exlcnded hy Johansen and Juselius (1994).
Ihe ind~vidual variahles uerc processed according to thc neth hods described.
hefore attempting the estimation of ~tiodel as detailed in the foregoing sections. The
sumniary statistics are presented in Table 3.
TAIIt.1 3 IIIMMAKY SIAIIFIICS (It MOI1EL VARIABLES
-. --
1981.1991
1991-2MM 1981-2004
Standwd Standard Standard
Vr8tahlc Mean Ikvlrinn Mean Dcv8auon Mran Ilrvm!nn .. .
11) PL_.- a A- 151 (6) (71
CC XI 3hi 5308 70600 1931 75748 6701
1iC hhXV3 1162 $5 041 3 011 60109 1510
IlhrFY 4 113 3 450 30 100 35 843 22491 31022
111 24905 13 381 20013 1 I560 22358 12427
IINI'ODP 0810 0614 11914 IU7IJ 6624 9486
IISHPH I 872 I320 13448 11 884 7912 1 0 3 1 3
Commodity concentration reflects the compositton of exports. It also ~ndicates
the direction of growth in lerms of variety, product differentiation and sources of
value realisation. It is therefore an important parameter to be examined for export
data analysis. For computing commodity concentration (CC), careful selection of
specific commodities is crucial. As composition of Indian sea food exports have
changed with the passage of time as detailed under Section 11, the items contributing
maximum value were selected. In the early 1970s Indian seafood exports basket

INDIAN S!-AFWD LIPORT!, ISSIIFS OF MST4BII.IrY. COMMOOITY CONCENFRATION 247
contained merely 37 Items (~nclud~ng processed) By 2000. it rose to 141 items
These product forms can be ma~nly categorised into the following: frozen shrimp (6
product forms), frozen fin fish (22). live items (5) chilled items (5), canned items (5)
dried items (30). shell Items (3). cunle fish (13) squid (14). frozen lobster (6) others
(32). t:ven though different product forms are exported from India, the major
commod~ties are frozen shrimp, fro~en fish. frozen cuttle fish, frozen squid which
constitutes nearly 90 per cent of the earnings from seafood expons. Hence, these
items arc conridered in the computation of commodity concentration and rest were
included as "others". Mean value of commodity concentration has decreased
ind~cati~ig the fact that lndia divers~fied her exports geographrcally during the 1990s
as compared to the 1980s since the index value decreased from 81.37 per cent to
70.60 percent (Table 3) The results (not presented here) also ~ndicate a gradual
dccrease In commod~ty concentration from 87.13 per cent in 1981-82 to 68.16 per
cent In 2003-04 which reflects the country's diversifying profile of exports Th~s is
confirmed by the facl that though the value added portion in seafood exports IS onl:
20 per cent. this segment is increasing at 55 per cent annually, tlowever, major
commodities still ~nake a substantial contribution to the total value realised from
cxports in value terms
Geographic concentration (GC) was computed co~is~dering Japan, USA, and
Western European countries and rest as "others" In splte of decrease in GC index
value from 66.89 per cent (1981-91) to 55.04 per cent (1991-2004) ind~cating in
general that thc geographic concct~tratio~~ of exports was h~gh. It also meant that the
markets for Indian seafood exports remained stable during th~r period. No structural
shift was observed durinp, the period under study. The figures remaining h~ghly
etahle, as can be seen from decrease in standard deviation from 5.46 per cent to 3 5
percent, the country's seafood exports rema~ned stable. Again, looking at the data on
the geographical spread of exports, with the ascendancy of the Euro, lndia has been
making deeper inroads into the EU markets; 27.42 per cent of Indian seafood was
exponed to the EU in 2004-05 with 23 37 per cent to the US and 18 per cent to Japan
Us~ng equation 2, the Indian seafood export instability was worked out, following
absolute average percentage deviations method of the observed values uslng an
exponentla1 path of trend values. The mean values of ~nstability index of seafood
exports were 4.37 per cent during the 1980s and 39.10 per cent during the 1990s,
indicating that the instability in sea food exports has increased over the period of
time. The potential of seafood exports as a source of enhanced revenue for propping
up the foreign exchange reserves and the national CDP was discovered only in the
early 1990s. The instabil~ty in the 1990s may be due to the gaps in institutionalising
the growth hhich led to short term fluctuations in the sector (Shanger al., 2001).
The decrease in instability indices of fisheries GDP from 24.91 per cent in the
1980s to 20.02 per cent in the 1990s with reduced standard deviation values of 13.38
per cent in the 1980s and 11.56 per cent in the 1990s indicate robust growth of the
fisheries sector. But the estimates also show that there is a need for stabilising the

24R
INDIAN IOIIRNAI. Uk AOKlCULTtJRAL ECONOMICS
growth of the sectors w~thin the economy. The value of instability indices of
non-fisheries CiDP increased from 0.81 per cent to 11.95 per cent with its
corresponding standard deviation increas~ng from 0.67 per cent to 10.71 per cent.
Instability in growth of lead sectors has been attributed to slow growth of the
agricultural sector (1.5 per cent) in the current plan period leading to an overall
decline in the perfornlance of the economy.
Hefore estimating the relationship hetwecll seafood exports instabil~ty and its
determinants, the integration properties or stationar~fy properties has to be
determined, The unlt root test werc perlnr~ned for all the variahles by employing
ADF test on hotli levels and First d~fferences of the variahles. The results presented in
l'able 4 indicate that the AIIF test in levels docs not all,>% for rejecting the null
lhypothcs~s (Ho) hut it can he rejccted at first differences. 'This means that all these
variables are integrated of the order onc, i.e. I( 1).
-.-- IAHI 1.4 AII(IMCN'T1 I) IIICKFY IliLi.1 K UNrl ROOTTt'SI KLSULT'1
- . ..- . .- --
1'81111111~1 lcrcli At llnl d~flcren~cs Ilcc~s~on
. -_JPL . 14)
Vrr~shle$ lririr~ -I lui -4 101
(51
!(I)
It / -il 8.16 -4 905 I(!)
ipc -I 283 .5 13R l(ll
12!/gdp -2 922 .5 380 1(1)
-1421 -5216 1111
As mentioned in the methodology, the Johnnsen ~~integration test is sensitive to
lag length. It is essential to identify the appropriate lag length before proceeding for
the co~ntegrat~on test. One of the most commonly used procedures to identify the lag
length is to estimate VAR using un-differenced data and compare their AIC (Chin
and Fang, 2003). Bascd on AIC results, optimum lag length indicated is I for the
equation. Since all the variables are I( 1) Johansen multivariate cointegration test was
applied to determine rank of impact matrix (n) and estimate the cointegration
equations. An unrestricted intercept and a linear trend in the variables, but not in the
co-integrating vectors enter the system. The results of A,,.,, and hmax in Table 5
Indicate that the tank of Il can be set to 2 based on A,. statistic at both 5 per cent
and I per cent level of significance and L., indicated that there exists one cointegrating
equation at both 5 per cent and I per cent level of significance.
Based on the largest Eigen value we considered the first cointegrating equation
and normalised long-run colntegration equation on liisfex variable. The cointegrating
vector corresponding to maximal Eigen value (i e., dominant long run relationship) is
presented as follows.

-
IAOLC 5 JOllANSrN COINTt(iKAIl0N TESl RtSIJLTS
I ra~c 5 per rmt I pcr cca Mux-I IEC~ 5 per ccnl
II~PULCEIC~ Llgon rlaltrlrc crtt~irl crbllcal \lul~slic sr~llc~l l lvr csnt
no ol ('I($1 vnlac h., valuc valur A,. valvr crlllcal valllc
(11 I21 (3) I4l
1% (bl 17) IBl -
Nunc 0937 13q6X V4 15 103 18 58 I2 39 17 45 111
Atinnst 1 0171 77 56 b8q2 7607 309( 3146 3871
Atnli,rl2 II hh3 46 61 4721 $4 46 22 84 2707 32 24
A1moql3 0 66 23 76 QOX 21 65 17 56 21197 2( 52
Almon4 0242 621 1541 2004 qR1 1407 1863
- M -. nmo,l - 5 .. - - 0 -- 018 I -
~
1~7h I1 3811 3 76
6 65 .
Ilh/ex = -15.462' Icc I 4.902' /yct0.03'l,!/grlp t 0.774' l,infgd/~ I 0.610' I,r,\lrpr
(2.214) (I 728) (0 102) (0 093) (0077) .. (10)
lo thc abobe long-rut1 model all coefficients liave the ant~cipated signs indicattng
yc. !i/pf/i,. ~rnfydp arrd mhla. have positwe effect on the tnstahility in cxpons and cc
have negative effect Thus one untt increase in the geographic concentration lndicates
- -
4.9 ltntts incrcasc 111 the tnstahil~t! index of lnd~an seafood exports. 1 he lnipact of
instahilttv doe to non-fiqheries GIN' is more comoared to ~nstah!l~tv In fisher~es GDP
as reflected by the magnitude of coeficients. Similarly. instahllity in shrimp
production also has a postttvc Impact on instahtlity tndex
Ilowever, commod~ty concentration appears to have a negative effect on the
instabilih. in seafood exports indtcarina that tf commoditv concentration decreases bv
one unit the instability of exports wtll increase by 15.46 tirnea. In this conlext this
may be interpreted as a fall in the proportion of high value frozen shrimp in favour of
lower value items in the composition of Indian seafood exports ovcr ttme Table 6
provides the shon run coefficiietits of the instability equation including errorcorrection
tern.
TAHL.1 h I STtMATl'S 01 SHOKI-RlIN INDIAN ~kAF001~ tXI'OHIS INSIAHII I1Y MOIICI
The international seafood market is governed by both price and non-price factors.
In the post-WTO scheme of globalised scenario, product improvement and
diflerentiation is more of a norm than an exception. Therefore, the short-term
coefiicients are smaller than their long run counter parts. This suggests that the
impact of these variables causing instability in Indian seafood exports requires tlme

250 INIIIAN JOIJRNAL Or AGKICUl.nlRAL ECONOMICS
for adjustment. The est~mated coeffic~ents show that all the var~ables except
instability in cultured shrimp production did not show sign~ficant short-run impact on
~nstab~lity 111 seafood exports. The significance of this short run effect is minimised
hy thc error-correction term, whlch is s~pn~ficant with the expected sign and of a
fairly larger magniludc. 'l'h~s findlng not only supports the validity of long-run
equilibrium relationsh~p among the variables but also indicates that instab~l~ty
scafood exports is sensit~ve and tends to depart from the equilibrium value in the
previous period. Its magnitude indicates that deviation from the long-run 8s adjusted
fairly quickly when 54.6 per cent of disequilibrium is recovered in each period. The
results also substantiate the fact that the seafood expon sector I5 responsive to the fast
changing profile of the market. Conforming to I-lazard Analysis and Critical Control
Points (HACCP) standards, improvcmcnts in processing and packaging standards and
develop~neiit and marketing of niche products to select the markets are enabled
quickly in order to take advantage of the gains from trade even in the short run (Dey
el ul. 2005).
VI

INDIAN SEAF1NJll kXMKTS ISSULS OF INSTABIL.ITY, COMMODITY CONCCNTRArlON 211
AYYRPPB~. A S and M Krlshnon (2005). "Ofthe111 Attempts Nrcded. Thr Ihndu Sume,, d Ayrirrirurr.
pp 124-136
Hhal K Sham and V Ntrmala 12001). "lnnah~l~r) of Agncuruial Exponr and Econom~i grodh I"
Ind~a"lndron Fionomic Joc,rnol. Vol 8. No 3. pp 87-91
Campa. Jort Mnnucl and (i bcmande, Nuna (20041. Sourcrr 01 Gamr /om lnrernorionai Po,,/i~lto
I)~vrrrrjicorrun. CEPR D~rcurs~oli Papcr Nu 4390 hIt~,i:s~r~~ cum'g~ctrj~ct 556712
Chln. I. and ti I-imp 120111). "Culnlegre~on Annlysls ol Singapore ller~dcnt~al Market". 10"' I.uropesl
Real Crolc Soctcty Conlerencc. 2001
1)evkota. S C (20041. (bust. ofhport lnrlvbliin rn hepol. lntcrnationul lrndc 0410002. Economlcr
Workkng Papcr Archtvc k.conWPA
IDcy Madan Mohan. Ruehlano M flrluncs and Maflur~udln Ahnlcd l2OOS). "D~saggregnted Analys~s of
Flih Supply. Ilemand and 1 rdc ~n Arm llascl~n~ Modcl and I rllmation Ssalegy'. Aqzmcuirurr
Economir~s undMonopamenr. Vol 9. No 1-2. pp 111-1411
t.nglc. RF and C WJ (irangcr (1987). "('cl-lntcgratlon and I.rror ('orrection Rcprescntst~on.
Csumiluon and Tcrt~ng". Lcononresrco. Vul 55, pp 251-276
Englc. R I aid C W J Granger (1991). Long-Kun Lconomic Krlation,~h~ps Keadmpr ~n Co~nrc$m,ron
Oxford IIn~vers~t) Press, OxRlrd
Eriat. (i, and Akyur (2001). "Countv Crrnccntrvt8on 01 lurlt~h thponr and lmpons over I~me".
Econum~ Wch Innltutc. www economic wehlnurnuidbcorioni,i. rhoomicr ond
Control. \'a1 12. pp 231-254
Johansen. S. (1995). Ltbbhood-Hosed Inlrrencr ,n (biaeprared t'scror Autorqqrurrirr d4odelr. Oxlard
linsvcrs~l) Press, Oxford
Johansen. S, and K. Juscl

211 INDIAN JOURNAL 01 AGRICUL.TURA1. EC ONOMICS
Shrimp and Gorp Form~ng in Asca. A Coliecl,on o/Rerrorch Popers hared on the ADlJ/,+'ACA
Ihrm Per/wmoncr Suwey. Network of Aquaculture Centcr~ ~n Ana-Pa~ific (NACAl (20011
I%sngkok. 1 halland. pp 33.46
Slnha. I' Ilipcndra (1999). Exporl Inrlobilib: lnve,~ln~enl ond Eronnmrr Growth tn A~ian Colinlr!es A
Time Scrtel A~O/YEIS. Cenler Dlscusslon I'npcr No 799. Economic Growth Ccnler, Yalc Untrcnlty.
Connrcr!cur
Stock. J I1 and M W Watson (1993). "A Simple hstirnator nl'Coir,agiat~ng Vcclon ~n Higher-Ordci
lntcgralcd Sysrems". Cronomrlrico. Vol 61. pp 7X3-820
lcgugnc. A (2000). "('ommotllt) Conc~.ntr"laln and Lkpon l'nrnlnps In\rahlll@ Cv~duncc from Airlcso
rounlrte\." Cenl~r hlr Ivannric Rnr\s. Moniclair Svslr
Llslvcrslty llppcr Moorclair Accc\scd Scprcmher. 2005
Ihup, ~1ph.1 tt>~tnl~I,~br \,