FISHERY BULLETIN: VOL. 84, NO. 2 



tated by widely held industrial perceptions is re- 

 jected. The results indicate that feedback exists 

 between the monthly fresh round ex-vessel prices 

 of pollock in both Gloucester and Boston. Both ports 

 dominate pollock landings and are close to one 

 another. 



A complete time series of prices for sea scallops 

 exists only for New Bedford. Since New Bedford 

 greatly dominates this fishery by both volume and 

 value of landings, it may be safely concluded that 

 monthly round ex-vessel sea scallop prices are 

 formed in New Bedford. Finally, Gloucester is the 

 only one of these ports to possess a complete time 

 series of prices and landings of ocean perch or red 

 fish. Since Gloucester dominates this fishery, 

 monthly fresh round ex-vessel ocean perch prices 

 appear to be formed in this port, at least among 

 these three. 



CONCLUDING COMMENTS 



The within-sample bivariate direct Granger 

 causality tests of monthly round ex-vessel price 

 linkages for the three most important New England 

 ports (Boston, New Bedford, and Gloucester) and 

 the most important groundfish species lead to unex- 

 pected results. Conventional wisdom considers the 

 round ex-vessel cod and haddock prices formed in 

 the Boston auction market to lead the comparable 

 prices of the other New England ports. However, 

 the empirical results indicate that New Bedford's 

 prices lead those of the other ports, although in cer- 

 tain cases Boston's cod prices may lead those of 

 Gloucester as well. 



The common industry perception also holds that 

 the yellowtail and winter flounder round ex-vessel 

 prices are first formed in New Bedford and lead 

 those of Boston and Gloucester. Instead, the em- 

 pirical findings suggest that feedback and simul- 

 taneous price formation occur among all three ports 

 for both species. Since flounder landings in Boston 

 and Gloucester are negligible in comparison to those 

 of New Bedford, a spurious relationship due to the 

 leading wholesale prices formed in the even earlier 

 and more flounder-important Fulton Fish Market 

 of New York City is suggested. Feedback is likely 

 for fresh round pollock ex-vessel price formation in 

 Boston and Gloucester. Finally, it is suggested that 

 the New Bedford auction market dominates fresh 

 ex-vessel sea scallop price formation and that 

 Gloucester dominates among these three ports for 

 ocean perch. New Bedford thus generally dominates 

 ex-vessel price formation among the major New 



England ports for the most important species 

 harvested. 



ACKNOWLEDGMENTS 



Helpful comments from David Bessler, Joseph 

 Mueller, Robert Reidman, Patricia Kurkul, and 

 DanieJ Huppert and an anonymous reviewer are 

 gratefully acknowledged. 



LITERATURE CITED 



Box, G. E. P., and D. Pierce. 



1970. Distribution of residual autocorrelations in autoregres- 



sive-integrated moving average time series models. J. Am. 



Stat. Assoc. 65:1509-1526. 

 Feige, E. L., and D. K. Pearce. 



1980. The casual causal relationship between money and in- 

 come: some caveats for time series analysis. SSRI Pap. No. 

 7809, 39 p. Social Syst. Res. Inst., Univ. Wisconsin, 

 Madison. 



Geweke, J. 



1981. The approximate slopes of econometric tests. Econo- 

 metrica 49:1427-1442. 



Geweke, J., R. Meese, and W. Dent. 



1983. Comparing alternative tests of causality in temporal 

 systems: analytic results and experimental evidence. J. 

 Econ. 21:161-194. 

 Granger, C. W. J. 



1977. Investigating causal relations by econometric models 

 and cross-spectral methods. Econometrica 37:424-438. 

 Guilkey, D. K., and M. K. Salemi. 



1982. Small sample properties of three tests for granger- 

 causal ordering in a bivariate stochastic system. Rev. Econ. 

 Stat. 64:668-680. 



Haugh, L. D. 



1976. Checking the independence of two covariance-station- 

 ary time series: a univariate cross correlation approach. J. 

 Am. Stat. Assoc. 71:378-385. 



Pierce, D. A. 



1977. Relationships - and the lack of thereof - between 

 economic time series, with special reference to money and 

 interest rates. J. Am. Stat. Assoc. 72:11-21. 



Sargent, R. J. 



1976. A classical macroeconomic model for the United States. 

 J. Polit. Econ. 84:207-238. 



Sims, C. 



1972. Money, income, and causality. Am. Econ. Rev. 62: 

 540-552. 



1974. Distribution lags. In M. D. Intriligator and D. A. Ken- 

 dricks (editors), Frontiers of quantitative economics, Vol. II, 

 p. 289-336. Amsterdam: North-Holland Publ. Co. 



1977. Comment to Pierce. J. Am. Stat. Soc. 72:23-24. 

 Snedecor, G., and W. Cochran. 



1976. Statistical methods. 6th ed. State Univ. Press, 



Ames, 593 p. 

 Wilson, J. A. 



1980. Adaption to uncertainty in small numbers exchange: 



the New England fresh fish market. Bell J. Econ. 11, p. 



491-504. 



442 



