Back to list of documents

Western Central Atlantic Fishery Commission

Ninth Session

WECAFC - Lesser Antilles Fisheries Committee

Sixth Session

Castries, Saint Lucia, 27-30 September 1999

COSTS AND BENEFITS OF COOPERATION AND CAPACITY BUILDING FOR RESPONSIBLE FISHERIES MANAGEMENT

SUMMARY

This technical document is intended to identify the potential benefits and costs of regional cooperation in capacity building and responsible fisheries management in the WECAFC region. It is being tabled for information purposes and will be presented and discussed using a workshop format. The aim is to assist the Commission in taking a decision on its future.

Introduction

1. The multispecies and transboundary nature of stocks distribution impose complexities and challenges for intelligent use of these resources over time which requires research and intense cooperation to address critical management questions at national, subregional, and regional levels in the WECAFC area.

2. The living resources of the WECAFC area are recognised as being highly diverse. (Cervigón et al., 1993) pointed out that in northern Atlantic coast of South America from northern Brazil to Colombia, there are about 680 fish species, 49 species of shark, and a number of cephalopods species, gasteropods and bivalves of economic importance. They are included in the harvest of 1.8 to 2.2 million tons landed yearly in the area (FAO, 1999). It should also be recognised that management capacity for sustainable harvest is also diverse in the WECAFC region.

3. The purpose of this paper is to present the regional application of the precautionary approach to fisheries management (FAO, 1995) to deal with the dynamics, risk and uncertainty inherent to relevant industrial and artisanal fisheries of WECAFC. This paper has also the purpose of providing a simple framework for identifying the potential benefits and costs of cooperation and capacity building aimed at assessing and managing responsibly the most important fisheries of the West Central Atlantic region.

A framework for cost-benefit analysis of training and cooperation

4. To undertake cost-benefit analysis of training and cooperation in the region several steps must be conducted. The following structure provides a guide to the essential steps (Hanley and Spash, 1993; Johansson, 1993; Schmid, 1989; Pearce and Nash, 1981):

• Defining the management cooperation project;
• Identifying bio-economic and social impacts which are relevant to the region;
• Physically quantifying the impacts;
• Calculating monetary values over time;
• Discounting to estimate the present value of the dynamic flow of costs and benefits of alternative management strategies under consideration;
• Specifying distributional criteria;
• Using decision tables to facilitate decision-making in an environment of risk and uncertainty.

Defining the management cooperation project

5. The definition of the fisheries management project will include the change in management strategy being proposed and, the population of gainers and losers within the coastal state or among participating countries. This definitional step is also used to determine boundaries of the analysis. For a coastal state for instance, certain management strategy oriented to improve the performance of an specific fishery may generate positive and/or negative externalities to fishermen targeting at technologically or ecologically interdependent fisheries. From the perspective of the coastal State the analysis should include the private and external marginal benefits and costs resulting from considering a change in fishery management strategy.

Identifying bio-economic impacts of the fisheries cooperation project

6. Once the fishery management cooperation project is specified, the next step is to identify all those impacts resulting from its implementation. In this process two important concepts are additionality and displacement. Additionality refers to the net impacts of the fishery management strategy. If a coastal state is appraising the introduction of a licence limit to avoid overexploitation and elimination of economic rent in an existing open access fishery, the benefits should be net of the benefits that could have been generated over time with the prevailing open access regime. Displacement refers to the reduction of fishermen opportunities in situations where reductions in fishing effort, and consequently direct employment, are not offset by possibilities of expanding effort in an alternative fishery that may still be in developmental stages.

7. In cases where the fishery is targeting at a transboundary resource, the distributional impacts of alternative management strategies should be accounted for. In fact, this type of situation calls for joint bio-economic analysis of the fishery in which scientists of coastal states involved should participate in the process of assessing the distributional impacts of the fishery management scheme under consideration.

Physical quantification of relevant impacts

8. This stage involves determining the physical amounts of costs and benefits flows for the fisheries cooperation project, and identifying when in time they will occur. Calculations made at this stage should make explicit the sources of uncertainty in dynamic fishery performance and conduct the corresponding risk analysis.

Monetary valuation of relevant effects

9. In order for physical measures to be co-measurable they must be valued in common units. The common unit in cost-benefit analysis is money. This is merely a devise of convenience rather than an implicit statement that money is all that matters. Inherent in training and cooperation programmes aiming at sustainable use of fisheries resources is the presence of values related to benefits or cost that do not have market expression and consequently they constitute unpriced values that should be made explicit (Sinden and Worrel, 1979).

Discounting of costs and benefits of the training and/or cooperation programme

10. Once all relevant costs and benefits that can be expressed in monetary amounts have been so expressed, it is necessary to convert them into present value (PV) terms. Selection of the appropriate rate of discount (price of time) in the estimation of the net present value or present value of the benefit/cost ratio of training and cooperation is a critical decision. Inherent in the price of time there are intertemporal preferences in the use of resources. The selection of the rate of discount also has implications on the time horizon considered as relevant for sustainable use of fishery resources. A high rate of discount will tend to favour fishery management decisions concerned with short run impacts rather than with intergenerational equity and sustainable harvest of marine living resources.

Distributional criteria in cooperation for responsible fisheries management

11. Under open access conditions and decreasing stock abundance, some fishers will be more affected than others. Generally those vessels with high autonomy and economic rent, will tend to increase harvest rates by exerting their fishing effort in distant fishing grounds. This context could produce conflicts among fishers of the countries targeting at a transboundary resource. Therefore, one of the major tasks in the selection of fishery management alternatives resides in how to favour a fishers group without damaging others and, in fact, how to increase at least proportionately the net benefits of all fishers involved. In the context of cooperation for sustainable fisheries management, economic damage is defined as the reduction in incomes or satisfaction of fishers of a coastal state as a result of an adopted fishery management strategy. In this context, equity is an important consideration in the process of cooperation for responsible fisheries. Nevertheless, it is difficult to implement, as equal opportunities do not guarantee an equal distribution of a marine renewable resource.

12. The main postulate of welfare economic theory as applied to fisheries (Hannesson, 1978), establishes that the maximisation of the social welfare is the final purpose of fishery policy. The main underlying concept is Pareto Optimality. A shared stock fishery management strategy is Pareto optimal, if changes generated by the joint fishery management strategy allow improving the welfare of one or more country fishers without worsening others. The neoclassical welfare theory applied to natural resource use and conservation (Randall, 1981; Hannesson, 1978; Schmid, 1989) considers several welfare approaches that have different distributional implications of costs and resulting benefits of alternative management strategies. These are summarised as follows:

• Pareto efficiency. This approach consists in eliminating inefficient fishery management solutions for the shared stock. It should be mentioned however, that more than one efficient solution can exist. The concept of Pareto efficiency is therefore defined as a situation in which it is impossible to improve the welfare of fishers of one country without simultaneously damaging another. In order to achieve this goal, all the possibilities of voluntary exchange that allow reallocating fish resources in a more efficient manner have been exhausted. The efficiency approaches permit economic damage, as total efficiency must be reached through exchange, an initial distribution of fishery rights will affect the distribution of benefits and costs.
• Pareto safety. This approach searches for improvements in the fishery sector, with the condition that the selected management strategy for the transboundary resource must increase the income of one country fishers without reducing the income of the other countries involved in the harvest of the same species. In welfare theory, this concept is defined as Paretian improvement, where economic damage is not permitted. However, a relative economic damage could exist, since although the reduction in the income of any fisher is not allowable, the relative distribution of the wealth generated by the transboundary resource could change.
• Maximum value of social welfare. This approach allows for real economic damage in one of the countries involved as long as there is a social agreement between the countries sharing the resource where the selected management strategy represents an improvement in social welfare of the transboundary resource fishery as a whole. As long as the sum of the changes in economic welfare of all countries involved is greater than zero and maximum, economic damage of one of the countries would be permitted.
• Proportional constant parts. This approach selects the shared stock fishery management strategy that generates an improvement that results in proportional increments in the income of fishers of co-operating countries. Therefore, all fishers, are benefited from the new management strategy in proportion of their initial income and historical rights. As a result, economic damage is not allowed. It should be mentioned however, that the distribution of the absolute amount of rent generated by the fishery is not the same for each fisher, but it is in proportion to the relative rent obtained previous to the implementation of the management strategy. The main characteristics of these criteria are presented in Table 1 (Seijo et al.1998).

Table 1. Welfare criteria applied to fisheries of transboundary resources

Welfare criteria of shared stock fishery	Distributional characteristics of welfare criteria	Distributional Impact
Pareto efficiency	Economic damage is permissible	Neutral for the less efficient fishers
Pareto safety	Economic damage is not allowed for participant countries	Neutral for all fishers
Maximum value of the social product	Permissible economic damage if sum of changes in countries� welfare > 0 and maximum	Neutral
Proportional parts to countries sharing the stock	Absolute or relative economic damage between fishers is not allowed	Proportional
Maximum social welfare	Economic damage is allowed.	Neutral

Costs of organising a workshop aimed at answering management questions

13. The organisation of meetings of regional or sub-regional assessment working groups, and annual meetings of decision-makers from each country involve: planning and preparation costs and implementation costs. The minimum requirements for effective cooperation for the two types of meetings and possible sources of financing are presented in a simple format in Tables 2 and 3.

14. Alternative cost sharing strategies and the role of FAO in facilitating the organisation of courses and workshops for stock assessment, bio-economic analysis, and the development of precautionary fisheries management plans are also suggested Tables 2 and 3.

Regional assessment working groups

15. It was estimated that a regional working group meeting of 26 participants, 3 FAO/CFRAMP Staff members, and 4 consultants could cost US$ 128,500 for a three-week workshop.

16. It should be pointed out that a three-day meeting with 10 decision-makers (Fisheries Directors) to present and discuss workshop results, which could have an additional cost of US$ 12,000.0, seems fundamental.

Table 2. Costs of organising meetings for regional working groups (e.g., spiny lobster, Panulirus argus, fisheries of the whole WECAFC area

Planning and implementation costs	Cost composition (% of total cost)	Possible cost sharing strategies
Planning costs	3.9	FAO, CFRAMP Host country
Travel costs of participants	15.6	FAO, CFRAMP
Travel costs of experts and consultants	5.8	FAO, CFRAMP
Hotel and daily subsistence allowance	47.9	Host country FAO, CFRAMP
Rental of equipment	5.6	FAO, CFRAMP
Supplies, materials and other operating costs	4.8	Host country
Honorarium of consultants	11.7	FAO, CFRAMP
Proceedings	3.9	FAO
Follow-up costs	0.8	FAO, CFRAMP
Total	100.0

Sub-regional assessment working group

17. Working group meetings for specific fisheries of geographic areas and ecosystems shared by two or more countries are also of great importance at sub-regional level. It was estimated that a sub-regional working group meeting of 12 participants, 2 FAO/CFRAMP Staff members, and 2 consultants could cost US$ 69,300 for a three-week workshop with the corresponding proceedings and follow-up costs. It should be pointed out that a three-day meeting with 5 decision-makers (Fisheries Directors) to present and discuss workshop results seems fundamental. It could have an additional cost of US$ 6,000.

Table 3. Costs of organising meetings for sub-regional working groups (e.g., shrimp and groundfish fisheries of the Brazil-Guianas shelf)

Planning and implementation costs	Cost composition (% of total cost)	Possible cost sharing strategies
Planning costs	2.2	FAO, CFRAMP Host country
Travel costs of participants	13.4	FAO, CFRAMP
Travel costs of experts and consultants	7.2	FAO, CFRAMP
Hotel and daily subsistence allowance	43.2	Host country FAO, CFRAMP
Rental of equipment	7.0	FAO, CFRAMP
Supplies, materials and other operating costs	4.0	Host country
Honorarium of consultants	14.4	FAO, CFRAMP
Proceedings	7.2	FAO
Follow-up costs	1.4	FAO, CFRAMP
Total	100.0

Cost sharing strategies

18. In order for the capacity building process to be sustainable over time, it is necessary that participating countries receiving the benefits from courses and workshops on fisheries assessment and management, contribute through sharing the costs of planning and organising them. There seems to be three types of possible arrangements for cooperation where the corresponding cost sharing strategy could take place.

• Bilateral agreements for joint assessment and management of shared resources such as the Trinidad & Venezuela agreement.
• Sub-regional working groups for training and cooperation for responsible management of a geographic area such as the shrimp and groundfish fisheries of the Brazil-Guianas shelf, or a specific species such as the flying fish.
• Regional working groups to deal with research and management questions relevant to highly migratory species, shared throughout the WECAFC area, in different stages of their life cycle (e.g. Spiny lobster (Panulirus argus) fishery).

19. With the participation of FAO as facilitator, the first of these possibilities may require a cost sharing arrangement between the two countries involved. The second, may need sustained contribution of participating countries, FAO, and other regional bodies such as OLDEPESCA and CARICOM. For resources shared throughout the WECAFC area, in addition to FAO and participating countries, other sources of financing could be explored to facilitate continuity in research, assessment and management efforts.

Net benefits from cooperation

20. After outlining the costs involved in planning and organising training courses and workshops for stock assessment and bio-economics of fisheries management, an estimation is needed for the present value of expected benefits net of the benefits that could have been generated over time with the prevailing harvest regime. These netted out benefits resulting from answering specific fishery management questions, should be greater than the costs incurred for arriving at the management strategy plus the change in corresponding enforcement costs.

Training and cooperation for responsible fisheries management

21. Responsible fisheries management (Caddy, 1996) requires regional cooperation through working groups for the application of the precautionary approach to fisheries of the WECAFC region. The inter-temporal use and management of living marine resources of the region require a systematic integration of the resource biology and ecology with the economic and social factors that determine resource and fishers behaviour over time. The approach suggested for the development of intelligent management strategies for the most important fisheries of this region, involves the following steps:

• To identify the set of management questions needed to be addressed by the working group. These management questions are to be specified by fisheries managers;
• To undertake biological, economic and social assessment of the fishery, i.e., estimate size and dynamics of the population structure, age structure of the harvest, costs and revenues of alternative fishing methods, direct employment and export earnings;
• To select the performance variables for the fishery;
• To establish limit and target reference points for the selected performance variables;
• To identify alternative management strategies for the fishery with the specific policy instruments;
• To identify different states of nature for those fishery variables and parameters (i.e. recruitment seasonality, natural mortality, unit costs of effort, catchability, etc.) that involve high levels of uncertainty;
• To determine if mathematical probabilities can be assigned for the occurrence of the identified states of nature;
• To build decision tables with and/or without mathematical probabilities;
• To apply different decision criteria reflecting different degrees of caution or risk aversion to select the optimum management strategy;
• To estimate the probabilities of exceeding the limit and target reference points of performance variables for the alternative management strategies under consideration; and
• To re-evaluate periodically the fishery to establish new reference points and management strategies.

22. The use of reference points (Caddy, 1998; Caddy and McGarvey, 1996, Caddy and Mahon, 1995; Die and Caddy, 1997) as objectives for resource administration represents an important step in the management process. Also, the recognition of the uncertainty present in various parts of the fishery system is fundamental for a precautionary approach to the decision making process. To aid this process, the use of fisheries specific mathematical models allow researchers, managers and resource users to experiment with different management options in order to observe the possible dynamic consequences on different parts of the system and corresponding performance variables.

23. To illustrate this process, two case studies developed by stock assessment and bio-economics working groups are presented. These cases involve regional and sub-regional management strategies for specific fisheries involving shared resources in one or more stages of their life cycles (e.g. shrimp fisheries in the Guiana-Brazil region, and spiny lobster, Panulirus argus, fisheries of the WECAFC).

Case study for a multi-species multi-fleet shared stock fishery: preliminary bio-economic analysis of the Trinidad & Tobago/Venezuela trawl fishery

24. To illustrate the above described precautionary approach, a summary of the working group cooperation resulting from the Shrimp and Groundfish Workshop which was held in Belem, Brazil from May 25 to 11, June 1999, is presented. The working group was formed by L. Ferreiera from Trinidad and J. Alió, and L. Marcano from Venezuela.

Shared Stock Fishery Description

The Trinidad and Tobago Fishery

25. Based on a trawl vessel census conducted in 1997 by the Fisheries Division of Trinidad, the trawl fleet comprises a total of 126 vessels: 30 artisanal Type I (7 to 10 m with outboard engines) and 66 artisanal Type II vessels (8 to 12 m with inboard diesel engines); 11 semi-industrial Type III vessels (10 to 12 m with inboard diesel engines); and 19 industrial Type IV vessels (17 to 23 m Gulf of Mexico double-rigged vessels) (Fabres et al 1995). All trawlers operate in the Gulf of Paria. In addition the industrial fleet operates in the Columbus Channel, as well as on the north coast of Trinidad. Five species of shrimp are exploited: Penaeus subtilis; P. schmitti; P. notialis; P. brasiliensis; and Xiphopenaeus kroyeri. Several species of demersal finfish from families such as Sciaenidae, Serranidae, Haemulidae and Lutjanidae are caught incidentally or may be targeted (Ferreira, 1998).

The Venezuela Fishery

26. This fishery is composed of two fleets, the industrial trawl fleet comprises 88 vessels (mostly metal vessels with 24 to 30 m in lentgh) that targets shrimp (P. subtilis and P. schmitti) and finfish of the families Sciaenidae, Carangidae, Haemulidae, Trichiuridae, Lutjanidae, Arridae and Mustelidae. This fleet operates in the southern Gulf of Paria and in front of the Orinoco river delta. Landings of this fleet during 1998 reached 6178 t of finfish and 636 t of shrimp (436 t of P. subtilis and 200 t of P. schmitti). The artisanal fleet of trawlers is composed of 28 wood vessels (8 m in length with out board engines) and operates in the northern sector of the Orinoco river delta. This fleet only targets juvenile P. schmitti, and during 1998 landed 131 t. The artisanal fleet was not included in the simulation modelling.

Methods

Seasonal model used to represent resource and fisher dynamics

27. A multi-species multi-fleet dynamic bio-economic model with seasonality was developed to represent the nature of this shared stock fishery. The model covered a four-year period. Data were included for 1995 to 1998 for Venezuela, and 1995 to July 1996 for Trinidad. The input parameters used for the model are given in Table 4. The model incorporates four of the shrimp species exploited in the Gulf of Paria-Columbus Channel region: Penaeus subtilis; P. schmitti; P. notialis; and Xiphopenaeus kroyeri. Data from Venezuela are included for only the first two species since the landings of the latter two are negligible. P. brasiliensis was excluded from the model since, although the species is important for the Trinidad industrial fleet, the data are not available. In the case of Venezuela the landings of P. brasiliensis are also negligible. Bycatch landed by the Trinidad trawl fleet was taken to be 0.4 times that of the shrimp landed by the fleet, while the bycatch landed by the industrial Venezuelan fleet was taken to be 10.1 times the shrimp landed by that fleet. Two recruitment peaks (February/March and July/August were assumed for all shrimp species except X. kroyeri based on an assessment of P. schmitti conducted by Altuve et al (1998), as well as a post larval abundance study conducted by Alio et al (1990).

Limit reference points

28. The limit reference points (LRP) specified were 0.3 of the standing biomass at the end of the four- year period (Bmax) of P. subtilis, i.e. 566 tonnes, and 0.3 of the Maximum Economic Yield (MEY) for the two fleets, i.e. US$2.9 million for Trinidad and US$5.0 million for Venezuela. The former LRP, 0.3 Bmax, was calculated using the formula CPUEmax/q where CPUEmax was taken as that of P. subtilis (0.226 tonnes/fishing day) in 1977 from the Venezuelan industrial fleet (Marcano et al. 1997) and q as the catchability coefficient (generated by the bio-economic model) of P. subtilis at 18 months old when the individuals are fully recruited. Limit reference point 0.3 MEY was determined for the entire Trinidad fleet and the Venezuelan industrial fleet separately where the MEY was derived from the bio-economic model.

Major sources of uncertainty

29. Natural mortality (M) was identified as a parameter representing a major source of uncertainty and hence the M for each of the four species was allowed to vary randomly while running Monte Carlo with varying levels of fishing effort for each fleet. The levels of effort used for the analysis were 4000 to 13000 days per fleet at 1000-day intervals, or 8000 to 26000 days for both fleets combined. The objective here was to observe performance variables (biomass of P. subtilis and the present value of rent of the Trinidadian and Venezuelan fleets) and the probabilities of exceeding the LRPs. The optimum effort at which the present value of rent of the fishery is maximised was then determined.

Bio-economic parameters

30. To illustrate data requirements to undertake the bio-economic analysis of a WECAFC fishery, the bio-economic parameter set used to model the dynamics of the Trinidad � Venezuela shrimp fishery are presented in Table 4.

Table 4. Bio-economic parameters of the Trinidad & Venezuela shrimp fisheries used in the bio-economic model

	P. schmitti	P. notialis	P.subtilis	X. kroyeri
Recruitment, R (individuals)	17000000	20000000	220000000	9000000
Growth parameter, k (1/month)	0,35	0,25	0,0927	0,09
Natural mortality coeficient, M (1/month)	0,2	0,2	0,146	0,2
Maximum weight, Wmax (g)	162,296333	145,589	64,9323044	27,7893088
Maximum Total length Lmax (mm)	250	220	176	155
Parameter t0 of growth equation	0	-0,017	0	0
Rate of discount	0,0045	0,0045	0,0045	0,0045
Effort dynamics parameter (fleet 1)	0,000003	0,00003	0,000003	0,000003
Effort dynamics parameter (fleet 2)	0,000003	0,000003	0,000003	0,000003
Average price of species 1 ($/ton)	7400	7400	3700	1380
Unit cost of effort fleet 1 ($/vessel/day)	186	186	186	186
Unit cost of effort fleet 2 ($/vessel/day)	1113	1113	1113	1113
Catchability coeficient fleet 1	0,002	0,000694	0,000694	0,000694
Catchability coeficient fleet 2	0,002	0	0,000694	0,000694
Length 50% retention fleet 1 (mm)	107	107	107	107
Length 75% retention fleet 1 (mm)	120	120	120	120
Selectivity parameter S1 fleet1	3,92707494	3,9270749	3,92707494	3,92707494
Selectivity parameter S2 fleet 1	0,03670163	0,0367016	0,03670163	0,03670163
Area swept fleet 1 (km2/day)	1,08	1,08	1,08	1,08
Total area of resource distribution (km2)	5000	3000	6000	1500
Selectivity & catchability parameters of the seabob fleet 2
Length 50% retention fleet 2 (mm)	107	107	107	107
Length 75% retention fleet 2 (mm):	120	120	120	120
Selectivity parameter S1 fleet 2	3,92707494	3,9270749	3,92707494	3,92707494
Selectivity parameter S2 fleet 2	0,03670163	0,0367016	0,03670163	0,03670163
Area swept fleet 2 (km2/month)	2,1	2,1	2,1	2,1
Parameter "a" of the growth equation	0,00000111	0,00001	0,000029	0,00000346
Parameter "b" of the growth equation	3,405	3,058	2,82789	3,1524
	Marcano et al. 1997	Fabres et al. 1995	Marcano et al. 1997	Lum Young et al. 1992

Results

31. Figure 1 provides the probability of exceeding the limit reference points (0.3 Bmax and 0.3 MEY) for the Trinidad and Venezuela fleets for the range of fishing efforts. In addition, the present value of the rent for the two fleets is given for the range of efforts. Figures 1 illustrates that as the total effort of the two fleets increases, the probability of exceeding 0.3 Bmax increases.

Figure 1. Probability of exceeding LRP=0.3Bmax of P. subtilis

32. For the entire range of efforts examined, there is 0% probability that profits will be less than 0.3 MEY for Trinidad, while for Venezuela the probability of exceeding this LRP increases fairly rapidly with increasing effort from about 8000 days (Figure 2).

Figure 2. Present value of the shared stock fishery rent over a range of fishing effort of the Trinidad and Venezuela fleets

33. Figure 2 shows that for the entire range of effort examined the present value of the Trinidad fleet ranges from approximately US$9 to 11 million, while the present value for the Venezuelan fleet decreases quite rapidly from 5000 days until the rent is completely eliminated by a fishing effort of that fleet of 13000 days. The current fishing effort (taken as that from 1995) is 8175 days (in Type IV units) for the Trinidad fleet, and 9348 days for the Venezuelan fleet. The optimum effort for the two fleets interacting dynamically (Díaz de León and Seijo, 1992), is estimated to be 5000 days for the Trinidad fleet and 6462 days for the Venezuelan fleet (Table 5).

Table 5. Current and optimum effort levels for the Trinidad and Venezuela fleets

Effort levels	Trinidad fleet (Type IV units)	Venezuela fleet Industrial vessels	Total effort (TT & Venezuela)
Current effort (1995 )	8175	9348	17523
Optimum effort1	5000	6460	11460

1. Effort generating maximum net present value of the fishery

Relevant conclusions of the working group

34. The preliminary bio-economic analysis conducted revealed that the fishing effort of both Trinidad and Venezuela fleets should not be increased beyond the current effort. In addition, the optimum allocation of fishing effort between the two fleets which would yield maximum profits is 61% of the current effort of the Trinidad fleet, and 69% of the current effort of the Venezuelan fleet which is equivalent to the effort of each fleet being reduced by approximately 3 000 days.

35. It is also evident that the effort reduction strategy towards the bio-economic optimum for the shared shrimp fishery of the Gulf of Paria, could potentially generate net benefits substantially above the cost of planning and organising the shrimp and groundfish fisheries workshop of the Brazil-Guianas shelf. Answering, through scientific cooperation, the management question of optimum effort levels for the Trinidad and Venezuela fleets sharing the transboundary stocks, provides a good example for future WECAFC efforts in the region to be facilitated by FAO.

Recommendations for Future Research

36. The input parameters to the model need to be refined. In addition, the costs should be determined as a function of yield, effort and number of boats in accordance with Sparre and Willmann (1993). Recruitment should also be included as a random function, with a probability distribution that approximates that of environmental parameters known to affect this process, like mean Orinoco river discharge during the more rainy months (August) or maximum yearly wind speed in the Gulf of Paria (Güiria or Port of Spain airports) (Alió et al. 1999b).

Fisheries management in the WECAFC region: dealing with risk and uncertainty

37. Butterworth et al. (1993), Hilborn and Peterman (1996) among others have identified a set of sources of uncertainty associated with stock assessment and management procedures, including:

• Uncertainty in resource abundance in space and time;
• Uncertainty in model structure and parameters;
• Uncertainty concerning behaviour of resource users;
• Uncertainty in future environmental conditions; and
• Uncertainty in future economic, political and social conditions.

38. To deal with these variety of uncertainties using a precautionary approach, it was suggested, in the Lysekil meeting (FAO, 1995), the use of Bayesian and non-Bayesian decision theory (Francis, 1992; Perez and Defeo, 1996; Defeo & Seijo 1999), and the incorporation of limit and target reference points to manage fisheries (Caddy and Mahon, 1995). Under this approach, decision-makers of the WECAFC region are expected to select one management strategy, d, out of a set of D alternative strategies. When selecting a strategy, the fishery manager should be aware of the corresponding consequences. These consequences are likely to be a function of the cause-effect relationships specified in the fishery model, the estimated bio-economic parameters, and the possible states of nature (Seijo et al. 1998). There is a probability that a target reference point (i.e. resource biomass, yield, rent, direct employment, export earnings, contribution to food security in coastal areas, etc.) may not be achieved because of inherent randomness of natural systems, incomplete knowledge of the fishery system, changes in economic and biological/ecological exogenous variables (Garcia, 1996a).

Decision tables with and without mathematical probabilities

39. In decision theory, it is important to be able to estimate a loss of opportunities function, L(d,q), which reflects the resulting losses of having selected strategy d when the state of nature occurring is q.

40. If prior or posterior probabilities are available to build decision tables to shrimp fishery managers, the expected values (EV) and their corresponding variance (VAR) should be estimated for the selected fishery performance variable (e.g. net present value of the fishery, biomass, yield, direct employment, export earning, among others) as follows:

EV_d= å Pq PVq _d (1)

 

VAR_d = S Pq (PVq _d - EV_d)² (2)

 

where Pq are the probabilities associated to the different states of nature, PVq _d are the values of the performance variable resulting from management decision d when state of nature q occurs. A risk neutral fisheries manager will select the management strategy that generates the maximum expected value with no consideration of the corresponding variance. A risk averse decision maker will tend to select the fisheries management strategy that generates the minimum variance. There are however different degrees of risk aversion, and therefore the decision theory provides alternative criteria for increasing degrees of caution in decision making (Shotton, 1995; Shotton & Francis, 1997). To apply these concepts to the precautionary approach to fisheries we will describe in the following section decision criteria with and without mathematical probabilities.

Bayesian criterion

41. The Bayesian criterion is a procedure that uses prior or posterior probabilities to aid the selection of a management strategy. For instance, it indicates the shrimp fishery manager to select the decision that minimises the expected loss of opportunities. Decisions without experimentation use prior distributions estimated out of experiences that are translated subjectively into numerical probabilities. Shrimp fishery decisions that are based on experimentation can use posterior probabilities. Posterior probabilities are the conditional probability of state of nature q, given the experimental data.

Decision criteria without mathematical probabilities

42. In the absence of sufficient observations to assign probabilities to possible states of nature, there are three decision criteria reflecting different degrees of precaution concerning selection of management strategies (Schmid, 1989; Seijo et. al.1998; Defeo & Seijo, 1999).

Minimax Criterion

43. The Minimax criterion estimates the maximum loss of opportunities of each management strategy and selects the one that provides the minimum of the maximum losses. This criterion proceeds as is nature would select the probability distribution, defined for all possible states of nature, that is least favourable for the decision-maker.

Maximin Criterion

44. This criterion uses the performance variable decision table that estimates the resulting values for a set of combinations of alternative decisions and states of nature. The criterion calculates a vector of the minimum values for the performance variable resulting from each alternative management decision. Then, the shrimp fishery manager proceeds to select the maximum of the minimum of those values. This is the most cautious of the decision theory approaches.

Maximax Criterion

45. A risk prone fishery manager would tend to apply the Maximax decision criterion when selecting the management strategy. The criterion calculates a vector of the maximum values for the performance variable resulting from each alternative management decision. Then, the shrimp fishery manager proceeds to select the maximum of the maximum of those values and the corresponding decision that generates it.

Case study of the Brazilian spiny lobster fishery

46. To illustrate the use of decision tables for the precautionary approach to fisheries management, a summary of the working group cooperation resulting from the Second Workshop on the Spiny Lobster, Panulirus argus, fisheries in the WECAFC area whichwas held in Merida, Mexico1-12 June, 1998, is presented. The group was was composed of R.C.A. Carvalho, C.A.S. Rocha, J.A.N. Aragao and R.N.L. Conceicao.

47. The exploitation of the spiny lobster in Brazil begun in the mid 50's in Recife and Fortaleza. A rapid expansion of the fishery took place since 1965, after the technological and economical feasibility was demonstrated. Two fishing methods have coexisted since then: traps and gill nets. Three fleets exercise their fishing effort on this resource: two artisanal (with and without outboard engine), and an industrial fleet. The former operate in near-shore areas while the latter apply their effort in deeper and more distant ecosystems. The fishing intensity has traditionally concentrated in north eastern Brazil. In the last few years there has been exploration of new fishing areas, mainly in the north-western and south-western fishing banks. The highest fishing intensity is applied in north-eastern Brazil.

Mathematical model and bio-economic parameters

48. The mathematical model describing the Brazilian fishery is a sex specific dynamic age structured bio-economic model. Effort was standardised in one effort unit called trap-day, the dominant fishing method. It should be pointed out that, as effort data by fishing method becomes available, it would be desirable to estimate effort dynamics for the different methods and corresponding fleet that undertake them. Specified prices correspond to x-vessel price. The unit cost of effort estimated during this workshop should carefully be reviewed to account for the number of effective fishing days.

Decision tables

49. Decision tables for fishery performance for net present value (NPV) of alternative management strategies were built for different states of nature and management options. States of nature refer to possible states of annual recruitment.

Table 6. Decision tables with and without mathematical probabilities

	Net present value of the fishery (US ' 000000)
Decision	R1=3000000	R2=6000000	R3=9000000	VE	VAR
(103 trap-days)	P1=0.3	P2=0.5	P3=0.2
Open access	-18.9	136.9	299.5	122.7	12367.5
15000	149.0	322.4	495.8	305.1	14731.4
28000	140.8	328.9	517.0	310.1	17336.5

Bayesian Criterion
Decision analysis with mathematical probabilities
(Loss of opportunities matrix)

Decision	R1=3000000	R=6000000	R=9000000
	P1=0.3	P2=0.5	P3=0.2	VE
D1	168.0	192.1	217.5	189.9
D2	0.0	6.5	21.2	7.5
D3	8.2	0.0	0.0	2.5
The Bayesian criterion is a procedure that indicates the lobster fishery manager to select the decision that minimises the expected loss of opportunities, in this case decision D3.

Minimax Criterion
Decision analysis without mathematical probabilities
(Loss of opportunities matrix)

Decision	State of Nature 1	State of Nature 2	State of Nature 3	Max
	R1=3000000	R2=6000000	R3=9000000
D1	168.0	192.1	217.5	217.5
D2	0.0	6.5	21.2	21.2
D3	8.2	0.0	0.0	8.2
The Minimax criterion estimates the maximum loss of opportunities of each management strategy and selects the one that provides the minimum of the maximum losses, in this case D3.

Maximin Criteria
Decision analysis without mathematical probabilities

Decision	State of Nature 1	State of Nature 2	State of Nature 1	Minimum
	R1=3000000	R2=6000000	R3=9000000
D1	-18.9	136.9	299.5	-18.9
D2	149.0	322.4	495.8	149.0
D3	140.8	328.9	517.0	140.8

50. The Maximin criterion uses the performance variable decision table that estimates the resulting values for a set of combinations of alternative decisions and states of nature. The criterion calculates a vector of the minimum values for the performance variable resulting from each alternative management decision. Then, the lobster fishery manager proceeds to select the maximum of the minimum of those values, in this case decision D₂. This is the most precautionary of the decision approaches.

Maximax Criteria
Decision analysis without mathematical probabilities

Decision	State of Nature 1	State of Nature 2	State of Nature 3	Maximum
	R1=3000000	R2=6000000	R3=9000000
D1	-18.9	136.9	299.5	299.5
D2	149.0	322.4	495.8	495.8
D3	140.8	328.9	517.0	517.0

Maximax Criterion

51. The criterion calculates a vector of the maximum values for the performance variable resulting from each alternative management decision. Then, the lobster fishery manager proceeds to select the maximum of the maximum of those values and the corresponding decision that generates it, in this case decision D3.

Concluding remarks

52. This simple framework for considering costs and benefits of training and cooperation for responsible fisheries management, is an exercise that provides the basis for future discussions when planning and organising training courses and workshops aiming at managing responsibly the most important fisheries of the WECAFC region. A cost-benefit rationale of this nature allows for evaluating the impacts of cooperation in assessment and bio-economic analysis of fisheries dealing with transboundary resources. The Trinidad-Venezuela joint bio-economic analysis applying the precautionary approach to shrimp fisheries management in the Gulf of Paria, provides an excellent case study where the potential net benefits from cooperation exceed substantially the costs of planning and organising workshops aiming at answering specific management questions. Sustainable efforts in training and cooperation facilitated by FAO in the region, will require establishing a reasonable cost sharing strategy agreed upon within WECAFC.

References

Alió, J.J.; M. Boada; L. Granadillo; D. Altuve; L. Briceño; J. Marcano & N. Tineo 1990. Disponibilidad de postlarvas de camarones peneidos en la región oriental de Venezuela. Informe Técnico. FONAIAP, Cumaná, Venezuela.

Alió, J.J.; L.Marcano; J.Páez C.; D.E. Altuve & N. Ehrhardt 1999a. Evaluation of the Penaeus schmitti stock within the Orinoco delta and Gulf of Paria region. Pp:157-172 in: National Reports and selected assessment reports presented at CFRAMP/FAO/DANIDA I Stock Assessment Workshop on the Shrimp and Groundfish Resources of the Guianas � Brazil Shelf. Port of Spain, Trinidad-Tobago 7 to 18 April 1997. FAO Fish. Rep. (Supplement).

Alió, J.J.; D. Die; L. Ferreira; K. Goriesigh; S. Kuruvilla; L. Maharaj; L.Marcano; I. Ramnarine & A. Richardson-Drakes 1999b. Penaeus subtilis stock within the Orinoco and Gulf of Paria region. Pp:173-189 in: National Reports and selected assessment reports presented at CFRAMP/FAO/DANIDA I Stock Assessment Workshop on the Shrimp and Groundfish Resources of the Guianas � Brazil Shelf. Port of Spain, Trinidad-Tobago 7 to 18 April 1997. FAO Fish. Rep. (Supplement).

Altuve, D.E.; R. Alvarez; L. Majaraj; J. Páez C.; N. Ehrhardt; L. Marcano & J.J. Alió 1998. Analysis of industrial and artisanal trawl fisheries of white srhimp, Penaeus schmitti, of the Gulf of Paria and Atlantic margin of Venezuela. Preliminary results. CFRAMP/FAO/DANIDA II Stock Assessment Workshop on the Shrimp and Groundfish Resources of the Guianas � Brazil Shelf. Liliendall, Guyana, May, 18-29, 1998.

Butterworth, D.S.; J.A.A. De Oliveira and K. Cochrane. 1993. Current initiatives in refining the management procedure for the South African anchovy resource. Proceedings of the International Symposium on Management Strategies for Exploited Fish Populations, Alaska Sea Grant College programme, 439-473.

Caddy, J.F. 1998. Deciding on precautionary management measures for stock and appropriate limit reference points (LRPs) as a basis for a multi-LRP harvest law. Paper presented in NAFO Scientific Council Meeting, March 1998

Caddy, J.F. 1996. A checklist for fisheries resource management issues seen from the perspective of the FAO Code of Conduct for Responsible Fisheries. FAO Fish. Circ., 917, 22p.

Caddy, J.F., and R. Mahon. 1995. Reference points for fisheries management. FAO Fish. Tech. Pap., 347, 83p.

Caddy, J.F. and McGarvey. 1996. Targets or limits for management of fisheries. N. Amer. J. Fish. Manage. 16(3):479-487.

Cervigón, F. et al. 1993. Field guide to commercial marine and brackish water resources of the northern coast of south America. FAO, Rome. 513 pp.

Defeo, O., and J.C. Seijo. 1999. Yield-mortality models: a precautionary bio-economic approach. Fisheries Research,840:1-10

Díaz de León, A., and J.C. Seijo. 1992. Multiple criterion optimisation for the control and management of a tropical fishery. Mar. Res. Econ., 7:23-40.

Die, D.J., and J.F. Caddy. 1997. Sustainable yield indicators from biomass: are there appropriate reference points for use in trpocal fisheries? Fish. Res., 32:69-78.

FAO, 1993. Marine fishery resources of the Antilles. FAO Fish. Tech. Pap. 326. 325 pp.

FAO, 1995. Precautionary approach to fisheries. FAO Fish. Tech. Pap. 350(1): 1-47

FAO, 1999. The status of resources and fisheries of the WECAFC area. WECAFC/99/E. 19p.

Fabres, B.; L, Maharaj & L. Ferreira 1995. The shrimp fishery of Trinidad and Tobago. National Report of Trinidad and Tobago. Pp: 30-51 in: National reports and selected papers presented at the Third Workshop on the Biological and Economic Modeling of the Shrimp Resources of the Guianas � Brazil shelf. FAO Fish. Rep. No. 526 (Supplement).

Ferreira, L. 1998. Economic analysis of the shrimp trawl fishery of Trinidad and Tobago with management implications. MMM Graduate Project, Dalhousie University, Nova Scotia, Canada.

García, S. M. 1996a. The precautionary approach to fisheries and its implication to fisheries research, technology and management: an updated review. In: FAO( 1996): precautionary approach to fisheries. Part 2. Scientific papers. FAO Fish. Tech. Pap., (350/2): 1/76.

García. S. M. 1996b. Indicators for sustainable development in fisheries. Paper presented at the 2^nd World Fisheries Congress. Workshop on Fisheries Sustainability Indicators. Brisbane, Australia, August 1996. 28p.

Hanley, N. and C.L. Spash. 1993. Cost-Benefit Analysis and the Environment. Edward Elgar Publishing Limited. USA. 278 p.

Hannesson, R. 1978. An introduction to fisheries economics.

Hilborn, R. and C.J. Walters. 1992. Quantitative fisheries stock assessment. Routledge, Chapman and Hall Inc.: 570p.

Hilborn, R., and R. Peterman. 1996. The development of scientific advice with incomplete information in the context of the precautionary approach. In: Precautionary approach to fisheries. Part 2. Scientific papers. FAO Fish. Tech. Pap. (350/2): 77-101.

Johansson, P. 1993. Cost-benefit analysis of environmental change. Cambridge University Press, Cambridge, UK. 232p.

Lum Young, P.; L. Ferreira & L. Maharaj 1992. Morphometric relationships for five species of western Atlantic shrimp occurring in the Trinidad and Tobago trawl fishery. Technical report of the Project for the Establishment of Data Collection Systems and Assessment of the Fisheries Resources. FAO/UNDP:TRI/91/001/TR12. Port of Spain (Trinidad and Tobago). July, 1992: 31 p.

Marcano, L.A.; J.J. Alió & D.E. Altuve 1997. Technical report of Venezuela: Venezuela shrimp fisheries on the Atlantic margin of Guayana (revised version 1994). Pp: 77-92 in CEFRAMP. National reports and selected papers presented at the Joint Meeting of the CFRAMP. National reports and selected papers presented at the Joint meeting of the CFRAMP Shrimp and Groundfish Subproject Specification Workshop and the Fourth Meeting of teh WECAFC Ad Hoc Shrimp and Groundfish Working Group of the Guianas-Brazil Continental Shelf. Port of Spain, Trinidad and Tobago, 8-12 January 1996. CARICOM Fish. Res. Doc. No. 22:281 p.

Pearce, D.W. and C.A. Nash, 1981. The social appraisal of projects. A text on cost-benefit analysis. A Halsted Press Book. John Wiley and Sons. New York. 225p.

Pérez, E. 1996. Análisis de riesgo e incertidumbre en la pesquería de Mesodesma donacium en Bahía de Coquimbo (Chile, IV región). M. Sc. Dissertation. CINVESTAV-IPN, México. 82 pp.

Perez, E. and O. Defeo. 1996. Estimación de riesgo e incertidumbre en modelos de producción captura-mortalidad. Biol. Pesq. (Chile)25: 3-15.

Seijo, J.C.; J.F. Caddy and J. Euán. 1994. SPATIAL: space-time dynamics in marine fisheries, a bio-economic software package for sedentary species. FAO Comp. Inf. Ser. Fish. (6): 116pp

Seijo, J.C., O. Defeo, and S. Salas. 1998. Fisheries bio-economics: theory, modelling and management. FAO Fish. Tech. Pap, 368, 108p.

Seijo, J.C. 1993. Individual transferable grounds in a community managed artisanal fishery. Mar. Res. Econ. 8:78-81.

Schmid, A.A. 1989. Benefit-Cost Analysis. A political economy approach. Westview Press. USA. 427 p.

Shotton, R. 1995. Attitudes to risk relative to decisions on levels of fish harvest. In: Precautionaryapproach to fisheries. Part 2. Scientific papers. FAO Fish. Tech. Pap. (350/2).

Shotton, R. and Francis, R.I.C.C. 1997. "Risk" in fisheries management: a review. Can. J. Fish. Aquat. Sci. 54: 1699-1715.

Sinden, A.J. and A.C. Worrell. 1979. Unpriced Values. Decisions without Market Prices. John Wiley and Sons. USA.