>1 indicates country i has a comparative advantage in production of j; the greater the index, the stronger the advantage. RCA 1 ij

Transcription

1 69 APPENDIX 1 RCA Indces In the followng we present some maor RCA ndces reported n the lterature. For addtonal varants and other RCA ndces, Memedovc (1994) and Vollrath (1991) provde more thorough revews. RCA 1 where represents country s export of product. RCA 1 >1 ndcates country has a comparatve advantage n producton of ; the greater the ndex, the stronger the advantage. RCA 1 <1 ndcates that country has a comparatve dsadvantage n producton of ; the smaller the ndex, the greater the dsadvantage. RCA 2 = M + M M + M where and M represent country s export and mport of product. RCA 2 >1 ndcates country has a comparatve advantage n producton of ; the greater the ndex, the stronger the advantage. RCA 2 <1 ndcates that country has a comparatve dsadvantage n producton of ; the smaller the ndex, the greater the dsadvantage. T RCA 3 = Y Q Y w where Q and t represent country s producton and net trade (.e. producton mnus consumpton) of product whle Y and Y w represent country s GNP and the world GNP, respectvely. RCA 3 > 0 ndcates country has a comparatve advantage n producton of ; the greater the ndex, the stronger the advantage. RCA 3 < 0 ndcates

2 70 Assessment of comparatve advantage n aquaculture that country has a comparatve dsadvantage n producton of ; the smaller the ndex, the greater the dsadvantage. RCA 4 = / X k X n / X nk M / M k M n / M nk RCA 5 = ln X / X k X n / X nk RCA 6 = ln M / M k M n / M nk where and X k represent country s exports of product and ts total exports of other products; X n and X nk represent the exports of product and the total exports of other products by the rest of the world; M and M k represent country s mport of product and ts total mports of other products; fnally, M n and M nk represent mports of product and total mports of other products by the rest of the world. Accordng to Vollrath (1991, p. 276), a postve RCA 4, RCA 5, or RCA 6 reveals a comparatve advantage, whle a negatve value reveals a comparatve dsadvantage.

3 71 APPENDIX 2 A proper measure for comparatve advantage varatons Snce Balassa s RCA ndex s a measure of comparatve advantage at a pont n tme, t seems natural to use the dfference between RCA ndces at the begnnng and end of a perod to measure the change of comparatve advantage durng the perod. Although ths has been a common practce, 1 ts theoretcal ustfcaton has not been establshed. In the followng we derve a measure of comparatve advantage varaton wth theoretcal foundaton; the result shows that the drect use of the dfference between RCA ndces at dfferent tmes to measure revealed comparatve advantage varaton s not generally approprate. Methodologcally, we frst derve what country s RCA ndex for product would have been at tme t+1 f t mantans ts comparatve advantage n the product durng the perod between tme t and t+1. Then we can use the devaton of ts actual RCA,t+1 from ths benchmark RCA ndex to measure ts comparatve advantage varaton durng the perod. Accordng to equaton (3), country s revealed comparatve advantage n any product s measured by the rato between ts share n export market and ts world market share. Thus, country s comparatve advantage n market relatve to market k can be measured by the rato between ts shares n the two markets,.e. s,t /s k,t. Therefore, country would experence no change n ts comparatve advantage pattern between tme t and t+1 f ts market share ratos for any two markets reman constant durng the perod,.e. s,t +1 = s,t,,k, (A.1) s k,t +1 s k,t where s,t +1 represents what country s share n market would have been under no comparatve advantage varatons. Accordng to equaton (A.1), a country would experence no comparatve advantage varaton when ts market share n every market grows at the same rate,.e., s,t 1, (A.2) s,t where α s a postve constant. Gven the total export of product at tme t+1 (.e E,t+1 ), had country experenced no comparatve advantage varaton between tme t and t+1, ts export of product would have been E,t +1 = s,t +1 E,t +1, whch, accordng to equaton (A.2), gves, E,t 1 s,t E,t 1 (A.3) 1 For example, Bonec (2001); Hley (1999); Havrla and Gunawardana (2003); and Yeats (1992).

4 72 Assessment of comparatve advantage n aquaculture Accordngly, country s specalzaton n producton would have been c,t +1 = E,t +1 E,,t +1 k k whch, substtuted n equaton (A.3), gves c,t +1 = k s,t E,t +1 s k,t E k,t +1 (A.4) Then, gven ts actual total export E,t+1, country s constant-comparatve-advantage benchmark export of product would be E,t +1 = c,t +1 E,t +1, whch, substtuted n equaton (A.4), gves E,t +1 = s E E,t,t +1,t +1 s E = (1 + g )c E,t,t +1. (A.5) k,t k,t +1 c (1 + g k,t k ) k k Snce E,t +1 represents what country s export of product would have been under no comparatve advantage varaton, the devaton of ts actual export of product (.e., E,t+1 ) from ths constant-comparatve-advantage benchmark would provde a measure of the country s comparatve advantage varaton between tme t and t+1. Smlar to how the RCA ndex s defned, a revealed comparatve advantage varaton (RCAV) ndex can be defned as RCAV = E,t +1 / E,t +1 E,t +1 / E t +1 E,t +1 / E,t +1 = RCA,t +1 βrca (A.6),t E,t +1 / E t +1 1 g where, 1 c,t g g = (E,t+1 E,t ) /E,t represents the growth rate of world exports of product between tme t and t+1, and g = (E +1 E t ) /E t represents the growth rate of total world exports of all products. The RCA ndex can be expressed n two equvalent forms (see equatons 3 and 4). Lkewse, the RCAV ndex defned n equaton (A.6) can be manfested as or RCAV = s,t +1 s,t +1 s,t +1 s,t +1 RCAV = c,t +1 c,t +1. c,t +1 c,t +1 (A.6.1) (A.6.2)

5 Appendx 2 A proper measure for comparatve advantage varatons 73 A postve RCAV ndex mples that country has ncreased ts comparatve advantage n product ; the hgher the ndex s, the greater the advantage gan s. A negatve RCAV ndex would have the exact opposte mplcaton. It s not dffcult to see that c,t g = g, where c,t = E,t /E t represents the proporton of world cultured shrmp exports sold to market. Thus, β would be unty when c,t s dentcal to c,t for every market,.e. accordng to equaton (4), when country s RCA ndex for every market s equal to unty (.e. g g, ). Otherwse, β would generally be dfferent from unty. Therefore, when the szes of markets are changed dsproportonately, drect use of the varaton of RCA ndces to measure comparatve advantage varaton would not be approprate n general. For example, LAC country Uruguay had catfsh RCA ndces of and respectvely durng and , whch seemngly ndcates that t has reduced ts comparatve advantage n catfsh between the frst and second half of the 1990s. However, the country s specalzaton n catfsh has actually ncreased from 69 percent durng to 77 percent durng The correspondng RCAV ndex, whch s postve at 4.65, properly reflects ths comparatve advantage gan.

6

7