FISHERY BULLETIN: VOL. 70, NO. 3 



up our yellowfin tuna resource and have reduced 

 the unused portion of the skipjack resource from 

 47,489 to 30,769 tons. In addition, we have 2,749 

 tons of empty capacity in the small boats. This 

 excess capacity is enforced, to a large extent, 

 by the technological constraints, and we can see 

 that these modifications would enable utilization 

 of the empty space with skipjack tuna. Thus we 

 can formulate, in a programming context, the 

 relation between the inputs and outputs of the 

 fishing process. If we agree that the manage- 

 ment process requires the kinds of information 

 that are required in the programming problem, 

 then we can see that we have been collecting the 

 wrong kinds of information on our fisheries. 



To sum up, then, we have discussed the pro- 

 duction function from a linear-programming 

 point of view. We have picked two possible ex- 

 amples out of an infinitude of possible examples. 

 The particular examples we have chosen may be 

 criticized from the point of view of their imme- 

 diate applicability to real situations. This criti- 

 cism is correct and indeed it is quite an important 

 criticism which simply reflects that in these tuna 

 fisheries and most of the other fisheries in the 

 world, we simply neither have nor collect the 

 kinds of data that we need to enter into an anal- 

 ytic evaluation of what is perhaps the most 

 critical of fishery management problems, the al- 

 location of fishery resources among various user 

 groups throughout the time stream. This is not 

 because these data do not exist; it is because, 

 in general, explicit attempts have not been made 

 to gather these sorts of data. It is a contradic- 

 tion to deny the usefulness of utilizing the phy- 

 sical metric for managing fisheries and to not 

 provide mechanisms for obtaining the kinds of 

 data that are required to manage the fisheries 

 in the appropriate way, in the value metric. 



The point, then, of demonstrating the linear- 

 programming technique is to (1) call attention 

 to a powerful allocation tool which can be used 

 for guidance in, for example, a serious contemp- 

 orary tuna problem, the allocation of the tuna 

 catch among the nations; (2) highlight the im- 

 portant diflference between the inputs of the 

 fishing process and the fishing eff"ort used in pop- 

 ulation dynamics; (3) point out the nature of 

 sensitivity in a programming context which can 



show, for example, that when we examine the 

 entire productive process that, given the right 

 kinds of economic data, we can think of man- 

 aging stocks in terms of, say, an upper and lower 

 bound on catch which could free research eff"ort, 

 for example, to other productive endeavors; and 

 (4) finally, because of recent confusion on the 

 subject, suggest that the term fishing eflfort be 

 utilized only in the context in which it is defined 

 in the population dynamics literature and that 

 the term fishing inputs be reserved for the more 

 general connotation of "fishing effort." 



INTERPRETATION OF FISHING SKILL 



Now let us look at the input process in a little 

 more detail. When we do this we have to admit 

 that we can, having established the definitions 

 of fishing effort and fishing inputs, especially if 

 we restrict our consideration of management to 

 manipulating physical quantities of the catch, 

 relate at least some but, in general, not all of 

 the fishing inputs to fishing effort through the 

 appropriate catchability coefficient. This en- 

 ables the dynamicist to have comparable mea- 

 sures of the abundance of fish from time-space 

 point to time-space point. Again the adjustment 

 of estimates of abundance to common units 

 through the computation of fishing power is well 

 treated in the literature, and we will not belabor 

 it here, except to note that fishing power is al- 

 most always calculated on the basis of some usu- 

 ally single physical feature of the fishing vessel 

 such as engine horsepower, etc., or simply on 

 em])irical differences in the catch-per-nominal- 

 effort that is obtained by the fleet. Differences 

 in fishing power are certainly more complicated 

 than comparisons among the physical attributes 

 of the fishing vessels would indicate. A consid- 

 erable portion of the variability in fishing power 

 among fishing units can be attributed to varia- 

 bility in the skill of the fishing skipper. This 

 assertion is subsumed in Figure 1. 



Figure 1 is hypothetical and shows that the 

 quality of fishing skippers could be a more im- 

 portant determinant of the "ciuality" of a fishing 

 operation than the physical characteristics of the 

 boats. We might guess that boats that are phy- 



676 



