The landing obligation in view of different management regimes

Fisheries Research 195 (2017) 202–213

Contents lists available at ScienceDirect

Fisheries Research journal homepage: www.elsevier.com/locate/fishres

Full length article

The landing obligation in view of different management regimes ⁎

MARK

Hans S. Frost, Ayoe Hoff

University of Copenhagen, Department of Food and Resource Economics, Rolighedsvej 25, 1958 Frederiksberg, Denmark

A R T I C L E I N F O

A B S T R A C T

Handled by A.E. Punt

The European Union adopted a landing obligation in 2015 implying that all catches of fish subject to quota management must be landed. We compare and contrast the economic consequences for fisheries of the landing obligation in view of the management system on which it is super-imposed. Four types of management are assessed: open access, shared quota, individual transferable quotas and economically optimal fishery. A standard non-linear programming bio-economic model is applied, providing illustrative numerical examples based on hypothetical parameter values. It is shown that the landing obligation has the strongest influence on both industry profitability and catch of unwanted species in the case of management with shared non-transferable quotas. In addition, the move from management with shared quotas to individual transferable quotas (ITQ) increases industry profitability and reduces unwanted catches. It is concluded that the effects of introducing the landing obligation in ITQ management systems are complex, but small.

Keywords: Fisheries management Economic optimization Discard and by-catches Landing obligation

1. Background and purpose Since the adoption of the Common Fisheries Policy (CFP) of the European Union (EU) in January 1983, the European Commission has repeatedly addressed the issue of discarding fish (EC, 2002, 2007, 2009; Borges, 2015), until the landing obligation (LO) was introduced in 2013 with implementation from 2015 (EC, 2013). This rule of landing is obligatory for all fisheries targeting species subject to TAC/quota management, independently of other management measures adopted by the individual member states. TACs (total allowable catches) are fixed for fish stocks, while quotas are allocations of TACs to Member States. The LO applies to all EU waters, however, the LO is gradually extended to apply also to species that are not subject to TAC but only to minimum landing sizes for the Mediterranean. It must be expected that the effects of the LO, both with regards to fisher economy and to reduction of unwanted catches, will depend on the management system on which the LO is superimposed. This paper investigates how the LO will affect fisher behaviour and profitability under management systems evolving from open access to fully-implemented management with individual transferable quotas (ITQ). Generally, the fisheries of the EU are diverse, with fish ranging from high-value species for human consumption to fish used for fishmeal and fishoil. A number of technological and biological interactions make it difficult for fisheries to be completely selective (Catchpole et al., 2005; Quirijns and Pastoors, 2014). Thus, it can be an economic advantage for fishers to discard fish, for several reasons, such as that: (i) the quota management in place may limit the catch possibilities in mixed ⁎

Corresponding author. E-mail address: [email protected] (A. Hoff).

http://dx.doi.org/10.1016/j.fishres.2017.07.013 Received 7 September 2016; Received in revised form 12 July 2017; Accepted 13 July 2017 0165-7836/ © 2017 Elsevier B.V. All rights reserved.

fisheries, thus creating incentives to discard low quota species to be able to catch a larger part of the quotas of other species and/or to highgrade, i.e. discard small low-value fish in favour of larger highervalue fish, and (ii) market and sorting inconsistencies, e.g. discarding of low-quality or damaged fish (Catchpole et al., 2013). Total allowable catches (TAC) were introduced and allocated to Member States as quotas with the CFP of 1983, using the relative stability principle. Distribution of the quotas between fisheries within a member state was, and is, the responsibility of the individual member state, and this ranges from shared quotas to ITQ systems. However, all member states had to apply a minimum landing size (MLS) of fish for human consumption, introduced with the CFP of 1983, combined with and supported by minimum mesh size regulations, leading to compulsory discard of fish below the minimum size. The technical measures apply to all EU waters. Mesh sizes in fishing gear and minimum size of fish mainly apply to the EU-waters outside the Mediterranean. For the Mediterranean, a range of other technical measures are used to take into account the specific biodiversity of this area (EC, 1998; Reeves et al., 2008). From a conservation point of view, the MLS regulation would help keep stocks at sustainable levels if (i) only fish above MLS was caught, or (ii) if fish discarded below MLS survived. However, in a strict sense, the MLS regulation would only assure sustainability if the MLS was above the age of first maturity. Three aims of the minimum landings size principle were of importance from an economic point of view.

• Protect

the market for human-consumption fish as a lack of

Fisheries Research 195 (2017) 202–213

H.S. Frost, A. Hoff

• •

selectivity in fishing gear caused different species and fish of different sizes to be caught together as is the case for e.g., sole and plaice, cod, haddock and whiting (Quirijns and Pastoors, 2014). Thus, in multi-species, multi-fleet fisheries, temporary large landings of small fish for human consumption would disrupt the market and exhaust the quota. Often, high grading (i.e., discarding fish to make room for more valuable fish) was carried out to make the fishery more profitable. Furthermore, it could be profitable in the industrial fisheries to catch e.g., small haddock and whiting for fishmeal and fishoil which deprives fleets fishing for human consumption of income opportunities. Minimum landing sizes aided in preventing such market disturbances, and it was thought to benefit the growth of fish stocks if catches of small-size fish could be avoided. This process was supported by the introduction of producer organizations (POs), whose aim initially was to stabilize the fish markets and secure fisher’s income through purchasing fish that did not fetch a minimum reference price for later processing and marketing. Later POs evolved in some member states to include also management of the quotas. Impact fishers’ behaviour. The hope was that fishers would avoid targeting certain species and sizes of fish if it was illegal, and thus costly, for then to carry such fish on board. Avoid early stops in yearly fisheries because of choke species. This was a problem particularly for the industrial fisheries, in which large bycatches of fish that otherwise could be used for human consumption and hence exhaust the quota were caught in small mesh gear, and furthermore in mixed demersal fisheries where low quota species could choke a fishery early in the year.

(1997) analysed a value-based ITQ system and found an optimal level of high grading from a welfare-economic point of view. While it seems obvious that fishers high grade to make the best possible use of their quota share in the ITQ management system, it is less clear why they want to high grade under open access or effort management rather than land the whole catch. The reason is high opportunity costs of landing fish, particularly the limited hold and processing capacity on board the vessel, along with the distance between the fishing ground and the port (Vestergaard, 1996). The seminal paper regarding compliance with the regulation is Sutinen and Andersen (1985), and in a recent paper, Hatcher (2014) addresses the incentives to discard and shows how the penalty for discarding/illegal landings, costs of discarding and the quota prices interact. He concludes: “Whether or not a discard ban is potentially welfare improving in any given situation, therefore, will depend on a number of complex factors, of which the regulatory cost of imposing such a ban is but one.” Although compliance is an important topic it is not included in our investigation, in which full compliance is assumed for all scenarios. Several recent papers have addressed the incentives for and effects of the LO. Condie et al. (2013) argue against the opinion of the EU Commission that a discard ban will create strong incentives for more selective fishing practices and a reduction in unmarketable catches of all species. Guillen et al. (2014) address the MSY objectives and say that in certain fisheries biomass at MSY can be significantly different when accounting or not for discard. Prellezo et al. (2016) use a bioeconomic simulation tool (FLBEIA) to anticipate the effects of the landing obligation on the Bay of Biscay Basque trawling fleet and find that there is a negative short term economic effect of the landing obligation and therefore incentives to improve the selectivity and to reduce the discard levels. Simons et al. (2015) investigate two discard prevention strategies for the North Sea saithe fishery where cod is a by-catch species. One was beneficial in protecting the saithe and cod stocks and in increasing net profits while the other had a negative impact on the saithe stock. Batsleer et al. (2016) model the potential effects of a discard ban on the annual fishing strategy of individual fishers in a mixed fishery under individual quota management and apply it to the North Sea beam trawl fishery. It is shown that a discard ban provides an incentive to implement more selective fishing gears. Alzorriz et al. (2016) analyse the selective properties of a bottom trawl fitted with different gear types (mesh sizes) used by Basque bottom otter trawlers and argue that the landing obligation will create an incentive to improve gear selectivity. Garcıa et al. (2017) evaluate the economic impact of the landing obligation policy on the Spanish demersal fleet operating in the Iberian Sea region, and show that the fleet dynamics impacts the result and the landing obligation should be accompanied by a management system with multi-stock reference points. Villasante et al. (2016) investigate the potential social and economic impacts of the discard ban in European small-scale fisheries and critical factors for its successful implementation and argue that compliance with the landing obligation of the small scale fisheries will be difficult to achieve without high economic costs. Finally, Heath et al. (2014) investigate discarding by fisheries in an ecosystem context. Discarded fish are food for a range of scavenging species, thus ending discard practises may have ecological consequences. While most of the papers from the 1990s, discussing discarding from a theoretical and general point of view, address specific management systems, the recent papers directly concerned with the EU LO do not address the underlying management systems explicitly and the impact of the LO on fisher behaviour and profitability subject to these systems. Thus, the aim of this paper is to investigate how different management systems affect economic performance when catches of fish smaller than a reference size must be discarded or landed respectively. The paper specifically addresses management with shared quotas contra ITQ management, and looks at how fishers’ profit-maximizing behaviour encourages the discarding of fish and what the economic repercussions of the landing obligation are for these management cases. The ITQ

However, the possibility for a discard ban was discussed with the revision of the CFP in 2002, given that discard was, and is, seen as waste of possible food resources and as accelerating the already severe decline observed in many fish stocks for human consumption (EC, 2002, 2007; Borges, 2015). Along with the political development, an increasing number of activities took place in terms of international conferences (e.g., FAO, 1996a,b; Pascoe, 1997; NCM, 2003), and purely theoretical work specifically addressing the subject of discard starting in the mid-1990s (Ward, 1994; Boyce, 1996). Ward et al. (2012), and Frost et al. (2013) provided reviews of later results related to the effects of discard. Before 2002, several countries, e.g., Iceland, the Faroe Islands and Norway, had already implemented a discard ban and prohibited fishing in certain areas if the landings of small fish became too high. And also in other parts of the world there has been a growing interest for alleviating discard and misreporting. Thus, the introduction of a LO with the revision of the CFP adopted in 2013 can be seen as a natural and expected development towards a long-term sustainable fishery regulation regarding both decreasing the economic loss from fisheries due to quota collision and increasing the sustainability of fish stocks. However, given that the LO is a command and control measure and not an economic measure, the effects of this on fishers’ behaviour and profitability are still uncertain (Borges, 2015). It must be assumed that these effects will depend on the management system on which the LO is super-imposed, and thus that the LO may affect fishers from different member states, fishing on the same fish stocks, but having different management systems in place, differently. This question is the focus of this paper. Fisher behaviour regarding unwanted catches, i.e., catches of nontarget species in open access and ITQ-managed fisheries, has previously been assessed (Ward, 1994; Boyce, 1996; Turner, 1997). These illustrative economic theoretical models typically include two species (target and non-target) and two fleets exploiting one or both species. The models are used to identify effort levels and effort allocation that leads to maximization of profit (or resource rent). Other theoretical approaches in the 1990s addressed high grading. Seminal papers in this field are Arnason (1994) and Anderson (1994), which show that a traditional ITQ system increases the incentive to high grade. Turner 203

Fisheries Research 195 (2017) 202–213

H.S. Frost, A. Hoff

stocks for each of the target species. The total industry profit π, assuming constant fish prices and constant marginal effort costs, is given by:

management scheme is analysed in both the medium-long run where the capital in terms of fleets but not the fish stocks have adjusted to the ITQ scheme, and in the long run where the fish stocks have also reached their optimal level given the ITQ management. The long run scenario is named OPF in our case. Moreover, the case of open access is included as a baseline scenario. As such, this development in management from open access via shared quotas to ITQ management illustrates the historical development in fisheries management in many European countries during the last 35 years. Our analysis makes it possible to compare and contrast the economic effects of moving from one management system to another, e.g. from shared quotas to ITQ management, with the effects of the introduction of the LO. These aspects are not addressed by the recent papers but rather by the theoretical papers of Ward (1994) and Boyce (1996) of which our paper is an extension with numerical results. As such our paper provides a broader perspective by adding the dimension of changing underlying management systems to the ongoing discussion of the economic and behavioural effects of the LO. A standard bio-economic model (Clark and Munro, 1975; Ward, 1994; Ward et al., 2012; Boyce, 1996) is used, supplemented by illustrative numerical examples with hypothetical, but arbitrarily fixed, parameter values. Non-linear programming is used to solve the model. In many respects, the model resembles large numerical management assessment models (e.g. Fishrent cf. Frost et al., 2013; FLBEIA cf. Garcia et al., 2013), which are calibrated based on available statistical data and is constructed to handle assessments of real-life fisheries. However, the output of such large real-life models is extensive, which may, in some cases, blur the relative results. Contrary to this, the small model used in the present context provides simple indicative results that are more straightforward to interpret and understand than more complex models. The model is developed in general terms, and as such, the qualitative solutions provided by the model produce generally applicable results. Our contribution the bio-economic literature is not that we apply a different model, but rather that different assumptions about the discarding behaviour are introduced by testing the effects of different restrictions in a conventional bio-economic model. This implies that the economic gains moving from open access to an economically optimally planned fishery are in accordance with conventional bio-economic theory. However, the effects of changes in discard rules are the new contribution to the literature.

π=

hi,j is the harvest of target species i above the reference landing size taken by fleet j. Additionally, fleet j may harvest a number of species i below the reference landing size. This amount is assumed to be proportional to hi,j with proportionality factor ki,j. Moreover, unintended bycatch may occur, which is again assumed to be proportional to hi,j by factor li,j. This assumption implicitly entails that large and small fish, together with the unintended bycatch, live among each other and cannot be separated by the fishing gear technology. The price of target species i above the reference landing size taken by fleet j is given by pi,j, while the net price (landing price minus extra landing costs of these species, given that other landings channels are often necessary for undersized and unwanted landings) of species i below the reference landing size is oi,j, and the net price of unintended bycatch is ui,j. cj is the variable cost per day at sea, and fj is the fixed cost per vessel in fleet segment j. Implicitly, the model uses opportunity costs, which are profits foregone by choosing one option at the expense of others. The fishery is, initially, in open access equilibrium as it is assumed that normal profit is included in the costs. Labour and capital are remunerated according to opportunity costs, and normal profit is defined as the remuneration to the owner of the vessel according to foregone earnings elsewhere. Section 3 discusses further aspects of industry and vessel adaptation, including the effects of normal and supranormal profit. Supranormal profit is defined as the profit exceeding the above-defined remunerations and includes the remuneration of the fish stocks. Discard will take place in the fishery if unit opportunity profits oi,j and ui,j are smaller than zero, and landings will take place if they are larger than zero. In practice, it is difficult to estimate the opportunity costs and benefits of fish; it requires estimating the loss of landing instantly rather than waiting until the fish have grown Moreover, long term growth requires that the fish survive, which is not always the case. Three control variables are assumed in the model evaluations: days at sea per vessel, number of vessels in each fleet segment, and stock size for each of the target species. The model initially represents an open access fishery (the baseline case) in which gross revenue equals total costs, i.e. where total profit is zero. The control variables that can vary in this case are the days at sea per vessel and the number of vessels. Thus, in this open access case, the total industry profit is given by:

The model includes two target and one non-target species, which do not interact (i.e., there is no prey-predator relationship). The species are caught simultaneously by two fleets that interact as one fleet’s catches impact the other fleet’s catches. That is, technical, but not biological, interactions are assumed. Each target species is divided into two classes: fish above and below the reference size. The model includes a non-target species, which allows investigations of the impact on the fishery if there is a discard ban on non-target species. The two fleets are assumed to reflect a trawl fleet (fleet 1) and a gillnet fleet (fleet 2) catching, for example, cod (species 1) and plaice (species 2) in different proportions. The fleets catch fish below the reference size in a fixed proportion to the fish above the reference size, and they catch nontarget species in a fixed proportion to their primary target-species. In the model the harvest hi,j of target species i (i = 1,…,M, M = 2 in the present case) taken by fleet j (j = 1,…,N, N = 2 in the present case) is evaluated using the Cobb-Douglas (CD) production function:

h i, j =

⎧⎡ M ∑i =1 pi,j hi,j + oi,j ki,j hi,j + ui,j li,j hi,j⎤⎥ − (cj dj + f j ) Vj ⎫⎬ ⎨⎢ ⎦ ⎩⎣ ⎭ (2)

2. The model

β qi, j (dj ·Vj )αi, j ·Xi i, j

N

∑j = 1

π=

N

∑j = 1

⎧⎡ M ∑i =1 pi,j hi,j + oi,j ki,j hi,j + ui,j li,j hi,j⎤⎥ − (cj dj + f j ) Vj ⎫⎬ = 0 ⎨⎢ ⎦ ⎩⎣ ⎭ (3)

The model considers management with shared quotas. In this case, the behaviour of the fishers is to maximize their catches (race to fish) of the target species by varying their days at sea: N

max d ⎡∑ ⎢ j=1 ⎣

M

∑i =1

h i, j ⎤ ⎥ ⎦

(4)

Finally, the model considers ITQ management in which case, the model will maximize the industry profit of the fishery:

⎡ N ⎡ M Π = max d , V ,(X )π = max ⎢∑ ∑i =1 pi,j hi,j + oi,j ki,j hi,j + ui,j li,j hi,j⎤⎥ j=1 ⎢ ⎦ ⎣ ⎣

(1)

⎤ − (cj dj + f j ) Vj⎥ ⎦

where qi,j, αi,j and βi,j are CD parameters, d = (d1, d2, …, dN ) is the vector of days at sea per vessel, and V = (V1, V2, …, VN ) is the vector of number of vessels in each segment, and X = (X1 , X2 , …, XM ) is the vector of

(5)

In this case, the control variables are days at sea per vessel and the 204

Fisheries Research 195 (2017) 202–213

H.S. Frost, A. Hoff

number of vessels in the medium long run and days at sea, number of vessels and stock sizes in the long run. This will be discussed further in Section 5. In the ITQ cases, quotas of the target species will be traded. Thus, each fleet, including inactive quota holders, has an extra income/ cost component ej, given by

ej =

M

∑i =1

si, j (hi, j − Qi0, j )

Table 1 Costs and fleet parameters.

Variable costs (1000 €/day at sea) Fixed costs (1000 €/vessel) Initial number of vessels Initial number of days per vessel per year Maximum number of days per vessel per year

(6)

where Qi0, j is the initially allocated quota, and si,j is the model’s estimated net price of the quota, i.e. sales revenue of the extra quota minus the opportunity costs of harvesting the extra quota. Thus, the final economic profit of the fishery is

ΠFinal = π +

N

∑j = 1

ej for all j

N

∑j = 1

(1 + ki, j )·hi, j = 0

N

(1 + lo·ki, j )·hi, j ≤ Qi ; Qi = ai Xi (1 − bi Xi )

(7)

(8)

(9)

where lo is a switch given by

1; landing obligation implemented lo = ⎧ ⎨ 0; discards allowed ⎩

0.80 70 40 97 180

Table 2 Cobb-Douglas parameters used in the model.

(10)

The number of days at sea per vessel cannot exceed a maximum number.

0 < dj ≤ Dj

1.38 120 20 217 250

The model has been calibrated with a set of parameter values leading to open access equilibrium, cf. Eq. (3) (see Table 1 for the resulting economic input parameters values for the two model fleets together with the initial number of days at sea per vessel per year). No distinction is made between days at sea and fishing days where the latter define pure fishing time and the former include steaming and search time. Table 1 also shows the initial number of vessels in each fleet. These parameter values, together with fish prices and catch rates, entail that the supranormal profit in the industry is zero and that the total revenue is equal to the total costs of the fleets. Note that this is only one set of parameter values among many that would cause the gross revenue to be equal to the total costs. In the present case, where the number of vessels is fixed, the number of days at sea is 217 for fleet 1 and 97 for fleet 2. Table 2 displays the parameters used in the Cobb-Douglas production functions in the model (cf. Eq. (1)). The parameter q indicates that fleet 1 targets species 1 as the primary species, while fleet 2 targets species 2 as the primary species. The parameter α is the exponent for effort (catch-effort flexibility rate) and shows that for fleet 1, effort has a larger impact on catch than the stock, denoted by parameter β. Fleet 1 is assumed to use mobile gear and to be more flexible than fleet 2, which is assumed to use stationary gear. Therefore, the parameter values for fleet 2, which is more dependent on stock size, are assumed to be the opposite of those of fleet 1. Table 3 displays the sales prices of target species above the reference size applied in the model. The prices reflect the assumption that the gillnetters land fish of larger size and better quality than the trawlers. It is implicitly assumed that the price minus the landings (opportunity) costs of undersized target species and the prices of bycatch species are below zero in the scenarios before the landing obligation is introduced and that undersized fish are therefore discarded before the introduction of the landing obligation. Table 4 shows the proportion of the target species below the reference size and unintended bycatch species caught per kg of target species above the reference size. This is the estimate of the discard fraction. For each kg of species 1 above the reference landing size caught by fleet 1, 0.1 kg below the reference size is caught, while fleet 1 catches 0.5 kg of species 2 below the reference landing size for each kg of species 2 above the reference landing size. For each kg of species 1 above the reference landing size caught by fleet 1, 0.1 kg of bycatch is caught, while 0 kg of bycatch is caught for each kg of species 2 above the reference landing size (as fleet 1 targets species 1). As such, imperfect selectivity is assumed. It is assumed that fleet 1 has a relatively large “bycatch” of species 2

Here it is assumed that the stock growth is described by a logistic surplus production function (Gordon, 1954), where the stock size is Xi, while ai and bi are growth parameters. It is furthermore assumed that the fish stocks are in a steady-state equilibrium ( x˙ ι = 0 ), i.e. that the growth is equal to the harvest implying that the model is comparative static. In the model, it is assumed that the allowed yearly landing (quota) of species i is assumed to be given by the surplus growth, i.e. by ai Xi (1 − bi Xi ) on the left-hand side of (8). In case of no landing obligation, the landings, which must be less than the quotas, are harvest of fish above the reference size, while the total harvest of fish both above and below the reference size must be less than or equal to the quotas given that the landing obligation is assumed implemented. Thus, the following constraint is included in the model:

∑j = 1

Fleet 2

3. Model parameters and data

In a fishery with complete transparency, certainty, divisibility and no transaction costs, the quota prices are determined by the model. When the profit of the whole harvesting sector is maximized in the long run, it is assumed that the exchange of quotas between vessels will take place in an optimal way. These assumptions imply that the economic solution from society’s point of view is optimal, but disregards distributional consequences, i.e., that when quotas are sold, fishers leave the industry, but still acquire income from the fishery. As such, the model does not include a market for ITQ; rather, the model determines the optimal distribution and quota prices, when the profit for the whole industry is maximized. Hence, when the maximum profit, ΠFinal, is determined by (5), the optimal quota price is determined by (6) and (7), and the quota price is affected by the landing obligation, as these are determined by the change in profit. Note that quota price si,j derived in the model is the unit profit in optimum. The actual purchase/selling price is the discounted value over infinity si,j/r, where r is the discount rate. In practice, such a price is not obtained, but a lower price will occur because of a shorter time horizon, uncertainty, a lack of transparency, divisibility and transaction costs. The profit maximization (5) is performed subject to a resource restriction:

x˙ ι = ai Xi (1 − bi Xi ) −

Fleet 1

(11) 205

Fleet no.

Species no.

Q

α

β

1

1 2

0.7 0.1

0.6 0.6

0.4 0.4

2

1 2

0.1 0.52

0.4 0.4

0.6 0.6

Fisheries Research 195 (2017) 202–213

H.S. Frost, A. Hoff

access scenario (baseline). If the fishers consider that the number of vessels in the fleet is constant and they disregard the resource restrictions, they are still able to control the number of days at sea per vessel. If the fishers cooperate (Ostrom, 1999), they will benefit from reducing the number of days at sea per vessel to the point where the distance between the gross revenue and the cost functions are highest (Fig. 3). If they do not cooperate, which is usually assumed in the literature (the ‘race to fish’), the equilibrium is at the upper intersection of the curves (cf. Eq. (3)). Restrictions imposed from outside, e.g., TACs or days at sea limitations, could make cooperation happen as fisher’s first reaction could be to reduce the number of days at sea per vessel, leading to higher profit. How a fisher reacts to known information is shown in Fig. 4. Assuming that the fisher wants to maximize his profit, knowing own fish prices and costs, the fisher will choose the number of days for which the short-run total revenue covers the short-run total costs. Therefore, the fisher will expand the number of days at sea to the point, where the revenue curve intersects the cost curve and even further.

Table 3 Prices of target species above reference landing size. Fleet no.

Species no.

Price (€/kg)

1

1 2

2 1.5

2

1 2

3 2

Table 4 Catch fractions of target species below reference size and bycatch species.

Species 1 below reference size Species 2 below reference size Non-target species fraction of species 1 Non-target species fraction of species 2

Fleet 1

Fleet 2

0.1 0.5 0.1 0

0.1 0.2 0 0.1

below the reference size in proportion of the catches of species 2 above the reference size because the mesh size in trawl is assumed to be smaller than in the gill net used by fleet 2. For fleet 2, the catches of undersized fish are therefore relatively smaller than the catches of the fish above the reference size. Finally, the parameter values for the initial stock sizes in the baseline case and the two fish stocks’ growth functions are displayed in Table 5.

5. Analysed scenarios for various management regimes The scenarios analysed in the present context examine the fishery transitioning from open access (the base case), over management with shared quotas, to management with ITQ, in each case without and with the LO introduced. ITQ regulation, broadly speaking, applies today in various formats for several fisheries within the EU (Quirijns and Pastoors, 2014). However, shared quotas, and the case of open access, both creating the incentive to race to fish, are included in the analyses to create perspective and to assess the possible effects of the landing obligation in view of different management schemes as these will change the behaviour of the fishers towards the LO. As such, a succession of management cases is considered, starting at open access, then moving to quota management with shared quotas, and finishing at ITQ management. Table 6 outlines the assumptions of each case with regards to control variables, constraints, fisher behaviour and constant variables. The TACs are divided among the fleets segments in fixed shares in the model applied here. Therefore, the TAC scenario reflects an individual non-transferable quota (IQ) system. Race to fish as assumed in the TAC case would entail higher costs compared to the IQ case in which race to fish is eliminated. It is assumed that the cost structure of the individual vessel is the same in all cases. In the shared quota cases, named the TAC scenario, the number of days at sea per vessel can change, but the number of vessels are kept constant, as the race to fish induced by shared quotas does not incite fishers to reduce the number of vessels if the short run and long run profit is positive. Conventional theory (e.g., Clark and Munroe, 1975; Anderson, 1986), does not distinguish between vessels (fixed capital) and days at sea (how capital is used variably), but rather between short run, where capital is fixed but the use is not, and long run where also capital is variable. In the ITQ case on the other hand, incentives are created to change both the number of vessels in the long run and the number of days at sea per vessel. In both these cases, the stock abundances are kept constant at the baseline (open access) level. This makes it possible to compare changes in behaviour between the three scenarios directly without having to consider the long-run effects on stock changes on behaviour. However, it is clear that in the long run stock changes will influence on catches of both over- and under-sized fish and on fisher behaviour, and therefore stock abundance is included as a control variable in the OPF scenario; this is the ultimate outcome of an ITQ management regime carried out for a sufficiently long time, in which fleets and stocks are fully adjusted, and the quotas are estimated taking both stocks and fleet technologies into account. In the TAC case, TACs of each species are set equal to natural growth minus estimated discard, i.e., the TAC is based on the landings, and the baseline (open access) days at sea of the two fleets are scaled down proportionally by an equal amount for both fleets until the total

4. Properties of the model To illustrate the impact on fishers’ behaviour of various restrictions, the properties of the model are shown in Figs. 1–4. In the open access equilibrium (Eq. (3)), fishers disregard the resource restriction and react only to economic information available and known to them. That means that other fishers’ impact on their fishery is not considered either. As soon as the government introduces restrictions, fishers react to them. In this case profit-maximizing behaviour must be assumed (Eq. (5)), i.e., the fishers will adjust their effort according to the imposed restrictions (Eqs. (8) and (9)). What happens to the fishers and the effects on the industry are outlined by Figs. 2–4, assuming constant stocks. Fig. 2 displays behaviour when the number of vessels changes. Fig. 3 assumes constant number of vessels, but number of days at sea per vessel may change, i.e. collaboration among fishers is assumed. Finally, Fig. 4 displays the individual fisher’s behaviour when no consideration is given to the fish stocks and other fisher’s behaviour. The resource restriction is defined by the logistic surplus production function (cf. Eq. (8)), as shown in Fig. 1. Note that the solid lines represent the open access stock sizes and growth for the two species, while the dotted lines show the optimal solution (OPF), representing maximum economic yield for the industry. In this solution, the growth of species 1 is smaller in the optimal solution than in open access but the stock size is higher. The production function (Eq. (1)) is a CobbDouglas function, and normal profit is included in the cost function. The cost functions are linear in effort (days at sea and number of vessels), cf. Eq. (2). Fig. 2 displays the total revenue and costs for a change in the number of vessels in the two fleets, assuming that the stock sizes are constant and that they use a fixed maximum number of days at sea per vessel, i.e. where the marginal profit for days at sea is zero in the open Table 5 Initial fish stock parameters.

Species 1 Species 2

Initial biomass

Growth

a

b

8000 4000

4800 2720

1 1

0.00005 0.00008

206

Fisheries Research 195 (2017) 202–213

H.S. Frost, A. Hoff

Fig. 1. The surplus production function for the two species included in the model. The solid lines represent the open access equilibrium case, while the dotted lines represent the stock biomass and yield in the profit maximization case for the whole fishery.

of over-quota catches to exhaust all quotas are not assumed in the present context. This assumption is made to allow a more direct comparison with the introduction of the LO, as the analyses is in this way focused on the effect of having to land previously undersized/unwanted catch and not previously highgraded/over-quota catch. For each management case, four landing obligation scenarios are carried out. These scenarios are outlined in Table 7. When the LO is assumed implemented three scenarios are considered: (i) landed undersized fish are sold at fishmeal/oil prices, but there are no added costs of landing undersized species, thus the net price of undersized fish is positive (‘Price > 0’), (ii) the costs of landing undersized species are equal to the fishmeal/oil price and thus the net price of undersized fish is zero (‘Price = 0’), and (iii) the costs of landing undersized species are higher than the fishmeal/oil prices and thus the net price of undersized fish is negative (‘Price < 0’).

landings are below the TACs for both species. Firstly, this means that the fishery is stopped once the most binding quota for the two species is reached, and secondly that each fleet maximizes their landings. Under these assumptions profit, catches and effort are underestimated in the case without the landing obligation as effort may, legally, be expanded beyond the quota for species two if the profit for continuing fishing only species one is positive (Hoff et al., 2010). Contrary to this, the industry profit is maximized in the ITQ and OPF scenarios. A first rank optimum, i.e., an ideal ITQ system (Frost et al., 2013), is obtained as the model computes the quota prices rather than uses ITQ prices determined in a market. Thus, full transparency, divisibility and zero transaction costs are assumed for the quota market. In the ITQ scenarios, profit maximisation is performed by varying days at sea and number of vessels subject to constant stock abundances, i.e., constant TAC. In the OPF scenario, profit is maximized by varying sea days, number of vessels and fish stocks. In the ITQ scenarios, the general TACs are the same as in the TAC scenarios because the fish stocks, and thus growth, are not changed. Contrary to this, the TACs vary in the OPF scenarios as the stocks vary. In all management cases, except open access, total landings are constrained below the quotas (cf. Eqs. (9) and (10)). ‘Landings’ are in the present context the catches above reference size in the no LO scenarios, and the sum of catches above and below the reference size in the LO scenarios. The fact that catches above reference size are constrained below the quotas in the no LO case implies an assumption of full compliance with all quota restrictions, i.e., highgrading and/or discard

6. Scenario results Table 8 shows the outcome of the baseline scenario, i.e. the open access case. The left-hand side of the table illustrates the economic equilibrium, i.e. that the profit is zero for both fleets. The right-hand side shows that the natural growth of both stocks is higher than the catches of fish above the reference size but lower than the landings plus the discard of fish below the reference size. Thus, the open-access case is in the present context in economic but not biological equilibrium, and will in the long run move towards lower stocks and lower growth until Fig. 2. The long-run economic equilibrium for an open access fishery; total costs and revenues for the whole fishery.

207

Fisheries Research 195 (2017) 202–213

H.S. Frost, A. Hoff

Fig. 3. Adjustment of number of days at sea per vessel subject to constant number of vessels and constant stocks; total costs and revenues for the whole fishery.

Fig. 5 further shows that compared to the shared quota (TAC) case, the profit increases significantly at the industry and vessel level by introducing ITQ management (using both days at sea and number of vessels as control variables and constraining total landings below total TACs, cf. Table 6), both when the stock is kept fixed at the baseline level (ITQ) and when the stock is used as control variable (OPF), and both when discards of undersized fish are allowed (no LO) and when all must be landed (LO implemented). On the average, the total profit is 17 times higher in the ITQ case and 30 times higher in the OPF case compared to the TAC case. The profit per vessel for fleets 1 and 2 are 52 and 117 times higher respectively in the ITQ case and 87 and 232 times higher respectively in the OPF case, compared with the TAC case. The reason is that the two fleets maximize their landings until the most binding quota is reached (i.e., maximise total landings, cf. Table 6) in the TAC case, irrespectively of the cost of doing so (or in other words ‘race to fish’ as discussed above), while the total industry profit is maximized in the ITQ and OPF cases. In the ITQ case, with fixed stocks, this means the quotas are not fully utilised, but this may in the short and medium long term be more profitable than fishing until the quotas stop the fishery. Thus, as is well-known, a considerable gain can be obtained when moving from shared quotas to ITQ management, both before (ITQ) and after (OPF) the stocks have adjusted to the new management scheme. Table 9 shows the profits in the LO scenarios relative to the no LO scenario for each management case. The table shows that compared with the transfer from shared quotas (TAC) to ITQ management, the introduction of the LO under ITQ management has a much smaller influence on the industry and individual vessel profitability. However,

also the biological equilibrium is reached. Finally, the two columns named ‘Natural growth – discard’ and ‘Quotas of fish above ref. size’ in Table 8 reflect how quota restrictions will be set when introduced; the quotas for landings above the minimum size should be equal to natural growth minus discard. However, in practice both the natural growth and the discard must be estimated based on recorded or non-recorded catches of fish below the minimum size. Fig. 5 displays the industry profit together with the average profits of each fleet for the TAC, ITQ and OPF management cases and scenarios (cf. Table 7). The industry profit is the total profit for the fishery, i.e., the sum of the total profit for each fleet segment, which can be calculated as the average profit per vessel in the segment times the number of vessels in the segment. Thus, the industry profit reflects the supranormal profit of the fishery, i.e., the resource rent, while the profit per vessel reflects the average surplus for the individual vessels after all costs have been covered (Anderson, 1986, ch. 3). Comparison of Fig. 5 and Table 8 firstly shows that the profits on industry and vessel levels are higher in the TAC than in the baseline (open access) case. This is caused by the days at sea being reduced in the TAC case, relative to the open access case (the number of vessels and fish stocks are kept constant, cf. Table 6), when the total species landings (catches above ref size in the no LO case and total catches in the LO case) are constrained below the quotas. Thus, with reference to Fig. 3, the costs are reduced linearly (cf. Eq. (2)) while the revenues are reduced along the convex Cobb-Douglas form (cf. Eq. (1)) and thus supranormal profit is created when moving from open access to management with fixed quota shares.

Fig. 4. Short-term total revenue and costs for the single vessel in the two fleets.

208

Fisheries Research 195 (2017) 202–213

H.S. Frost, A. Hoff

Table 6 Assumptions made in each modelled management scenario. Management Case

Control variables

Constant variables

Fisher behaviour

Base Case: Open Access

-Sea days -Number of vessels -Sea days

-Stock size

Total profit equals zero

-Stock size equal to base case -Number of vessels -Stock size equal to base case

Fleet landings maximized

Total landings constrained below TACs

-Industry profit maximized -Quota prices computed by the model -Industry profit maximizes -Quota prices computed by the model

Total landings constrained below TACs

TAC: Shared quotas ITQ: Individually transferable quotas OPF: Optimal Fishery

-Sea days -Number of vessels -Sea days -Number of vessels -Stock size

Restrictions

Total landings constrained below TACs

smaller percentage of undersized fish than fleet 1 and is thus less affected by the LO. In all it is concluded that for the case presented in this paper the LO may in the short run (the ITQ case) have a positive or negative effect on the profitability of the fishing industry, depending on the net price of undersized/unwanted fish, while the LO will have a negative effect on the fishing industry in the long run. However, individual vessels may still gain from the LO, given the species composition of their landings. Fig. 6 displays the total catches of the two species above and below reference size in each of the management cases and Fig. 7 displays the quota utilisations. The two figures firstly show that moving from management with shared quotas (the TAC case) to ITQ management, but with unchanged stock size (the ITQ case, cf. Table 6), leads to reduced catches of both species (Fig. 6) and thus reduced quota utilisations (Fig. 7). This corresponds with the discussion of the profitability above, i.e., that when the fishery moves from shared quotas (the TAC case) to ITQ management the behaviour changes from maximisation of landings weight to profit maximisation, leading to cost minimisation and thus reduced effort. Thus, even without the introduction of the LO there is in the present context a lower pressure on the fish stocks and thus corresponding less catch of undersized/unwanted fish in a fishery managed with ITQs relative to a fishery managed with shared quotas. However, in the long run, when the stocks have reached their optimal level (the OPF case), the catches of fish both above and below the reference size are, with our parameter choice, approximately the same as in the TAC case. However, this result happens with stocks that are almost doubled in size relative to the TAC case, and thus still with an overall lower relative pressure on the stocks, see Fig. 1. Fig. 6 also shows that in the shared quota (TAC) case the catches of both species are reduced when the LO is introduced. This occurs because species 2 chokes in the no LO case (cf. Fig. 7) given the assumption of full compliance with all quotas discussed in Section 5, and thus will also choke in the LO case, where the total catches of fish both above and below the reference size must be less than the quota, thus leading to overall reduced catches. For the ITQ case, the landings of both species increase slightly when the LO is introduced and the net price of undersized fish is positive when there are no extra costs of landing the undersized fish (‘Price > 0’). The reason is that the quotas are not binding for this management case in the no LO scenario (cf. Fig. 7) and thus when undersized fish cab be landed and sold at fishmeal/oil prices, it is possible for fleet 1 to invest in an extra vessel, thus

the profit will approximately double on both the industry and individual vessel level when the LO is introduced for the shared quota (TAC) case if the costs of having to land previously discarded undersized fish can be kept as low as possible (the ‘Price > 0’ scenario). On the other hand, the gain/loss is approximately negligible in the ITQ and OPF case. The reason for the profit increase in the TAC case is that to land undersized fish within the quota restrictions the fleet segments must reduce their effort and thus again move left in Fig. 3, hereby increasing their profits. In the ITQ case, where the stocks are kept fixed at the levels used for the TAC case (cf. Table 6), there is not full quota utilisation of the two species and, therefore, there is room in the quotas to land the undersized fish. Thus, there is a small gain in industry profit when the net price of undersized fish is greater than zero (‘Price > 0), caused by a small increase in number of vessels given the increased earning possibilities, but the individual vessel profits are unchanged. Correspondingly, the industry profit decreases a small amount when the net price of the undersized fish is less than zero. Table 9 further shows that in the OPF case, where the stock is now also a control variable alongside days at sea and number of vessels, the industry profit is generally lower when the LO is introduced, relative to the no LO case, and the profit further decreases with decreasing net prices on undersized fish. When the LO is introduced the stocks (that are control variables in the OPF case) will decrease which will create higher natural growth rate (cf. Fig. 1, the stocks are to the left of the MSY point) and thus quotas to accommodate what it is now total catch, and not only catches above reference size, that are constrained below the quotas. However, the two fleets catch less oversized fish and thus on the average their landings value and effort goes down, the effort being reduced through reduction in number of vessels. This creates profit increase for the individual vessels in the two fleets when net price of undersized fish is positive (“Price > 0) (Table 9), but on the industry level the reduction in both landings value and effort leads to an overall decrease in profit. The industry profit is reduced further when the net price of undersized fish is reduced (‘Price = 0” and “Price < 0”), while the picture is more varied for the individual vessels. Especially for fleet 2 the individual vessels have increasing profits given the reduction in net price for the unwanted species, while fleet 1 have decreasing profits when the price of undersized species is lowered. This difference is caused by the different landings compositions and amount of undersized fish taken by the two fleets; fleet 2 on the average takes a Table 7 Scenarios analysed in the paper. Management scenario

Open access TAC ITQ OPF

No landing obligation

Landing obligation (LO)

No cost at discard

Price = 0.2 € equal to fish for fishmeal and −oil, no extra landings costs.

Price=extra landings costs

Price < extra landings costs equal to 0.5 €

+ TAC – No LO ITQ – No LO OPF – No LO

Na TAC – LO, Price > 0 ITQ – LO, Price > 0 OPF – LO, Price > 0

Na TAC – LO, Price = 0 ITQ – LO, Price = 0 OPF – LO, Price = 0

Na TAC – LO, Price > 0 ITQ – LO, Price > 0 OPF – LO, Price > 0

209

Fisheries Research 195 (2017) 202–213

H.S. Frost, A. Hoff

Table 8 Output of the baseline scenario, i.e. the open access scenario.

No. of vessels No. of days per vessel per year Species 1 (landing tonnes) Species 2 (landing tonnes) Species 1 below ref. size (tonnes) Species 2 below ref. size (tonnes) Total landing value (1000 €) Total profit (1000 €) Profit per vessel (1000 €)

Fleet 1

Fleet 2

Total

20 217 3882 420 388 210 8395 0 0

40 97 600 2057 60 411 5914 0 0

60 4482 2478 448 622 14308 0

Natural growth

Landings + discard

Natural growth-discard

Quota of fish above ref. size

4800 2720

4930 3099

4352 2098

4352 2098

Table 9 Relative change (%) of the profit in each management scenario relative to the no LO case. TAC

ITQ

OPF

% change in total profit relative to no LO

LO, Price > 0 LO, Price = 0 LO, Price < 0

92.06 0.79 −136.51

4.74 0.00 −6.89

−3.16 −6.64 −11.52

% change in profit/vessel for Fleet 1 relative to no LO

LO, Price > 0 LO, Price = 0 LO, Price < 0

85.71 0.00 −142.86

0.00 0.00 0.00

6.59 −2.33 1.74

% change in profit/vessel for Fleet 2 relative to no LO

LO, Price > 0 LO, Price = 0 LO, Price < 0

100.00 0.00 −133.33

0.00 0.00 0.00

20.95 13.56 14.44

leading to increasing overall catches. However, when it becomes costly to land undersized/unwanted fish (‘Price = 0’ and ‘Price < 0’) the catches fall again. The picture is more complicated in the OPF case as the stock will now in each case adjust to its optimal size and the days at sea/number of vessels adjust to give as high a quota utilisation as possible (cf. Fig. 7). When moving from no LO to introduction of the LO with no extra costs of landing the undersized fish (‘Price > 0’) the two fleets reacts by reducing their number of vessels to accommodate the lower average landings prices. This leads to lower overall catches of the two species, but to higher landings and thus higher vessel profits for the two fleets. When the price of unwanted fish falls through introduction of landings costs on these (‘Price = 0’, ‘Price < 0’), both fleets reacts by increasing their sizes to be able to catch more fish above the reference size. This leads to marginally higher catches, but lower vessel profits for the two fleets. In the no LO scenario and in the ‘LO – Price > 0’ and ‘LO – Price = 0’ scenarios, the stocks of both species are continually decreasing, accommodating higher quotas (given the logistic shape of the growth function determining the TACs, cf. Fig. 1). However, when the costs of landing undersized/unwanted fish decrease further (‘LO – Price < 0’), the stock starts to increase as it becomes even less profitable for the fishery to land unwanted species. Thus, the size of both fleets decreases, leading to lower catches. The quotas are no longer binding in the ITQ case. In this case, the fishery stops when the marginal profit becomes negative. In our model, this effect is clear, but empirical evidence suggests that this effect may also occur in actual fisheries, e.g., in Denmark, where ITQ management is applied and where there is a tendency not to fully exhaust the quotas compared to the period where ITQ was not applied. This situation should not be confused with the TAC/quota management cases where the TACs/quotas are not fully exploited because of quota collision or very large TACs e.g. sand eel in certain periods. The stock abundances will grow if the quotas are not binding. In reality, many countries and fleets share the same stocks. Therefore, all countries and fleets – and not only the fleets for which the quotas are not exhausted – will benefit. The equilibrium is unstable, but in practice quota may not be reallocated, irrespective of the possibilities for quota swaps, because of the principle of relative stability between countries in

Fig. 5. Profit at industry level and at vessel level for each management scenario. ‘No LO’ = No Landing obligation, ‘LO, Price > 0’ = Landing obligation and price of undersized fish > 0, ‘LO, Price = 0’ = Landing obligation and price of undersized fish = 0, ‘LO, Price < 0’ = Landing obligation and price of undersized fish < 0.

210

Fisheries Research 195 (2017) 202–213

H.S. Frost, A. Hoff

Fig. 6. Total catch above and below reference size of species 1 and 2 in each of the management scenarios. ‘No LO’ = No Landing obligation, ‘LO, Price > 0’ = Landing obligation and price of undersized fish > 0, ‘LO, Price = 0’ = Landing obligation and price of undersized fish=0, ‘LO, Price < 0’ = Landing obligation and price of undersized fish < 0.

present open access case, serving as baseline, we assume economic, but not biological, equilibrium because this point is reached only by accident. However, complete bioeconomic equilibrium is reached in the long run (the OPF case) where both fishing effort and fish stocks adjust. Generally, if the objective is to maximize profit, this requires interference in terms of management, which is based on ex ante estimates of stock growth and fishing technology. This situation reflects a welfare economic equilibrium (MEY). Stability requires that there be full knowledge about stock growth, prices, costs, species composition catches and how to transform this information into quotas. If the quotas are not set correctly, the equilibrium will not be stable. Moreover, this situation also requires that prices and costs are constant, as any variation in these can lead to changes in the TACs/quotas. The model evaluations firstly show that there is a significant increase in both industry and vessel profitability when moving from shared quota (TAC) to ITQ management, accompanied by a reduction in catch of fish both above and below the reference size. A further gain in profit occurs (both without and with the LO implemented) and a corresponding increase in catches of fish both above and below the reference size when ITQ management are brought to long term equilibrium (the OPF case). However, in this case the stocks have almost doubled relative to the short term ITQ and TAC cases and thus the relative pressure on the stocks are lower than in the TAC case. Thus, both without and with the landing obligation a change in management system from shared quotas to ITQ management will improve industry and vessel profitability and at the same time reduce the pressure on the fish stocks and the relative uptake of undersized/unwanted fish. Introduction of the LO in the shared quota (TAC) case leads to a considerable decrease in catches and thus especially in catches of undersized fish. Depending on the net prices obtained for undersized fish

the EU. The change in the stocks in the long run will continue to the points where the supranormal profit (resource rent) is maximized. In the model, this leads to a stock abundance of both species 1 and 2 that more than doubles. However, this does not necessarily mean the natural growth determining the quotas changes in the same amount, as the growth functions are assumed to be logistic, cf. Fig. 1. 7. Discussion and conclusion The model is used to investigate the economic repercussions of the introduction of the landing obligation in view of various management measures. Three management cases have been considered; shared quotas resulting in maximization of landings weight (the TAC case), ITQ management in the short run (assuming the same fish stock sizes as in the shared quota case), and ITQ management, named OPF, in the long run (assuming that the fish stocks and fleets have reached a long run equilibrium). Moreover, a baseline, open access case, has been included, and this case has been used to calibrate the model. Each of the three management cases has been modelled without and with the landing obligation (LO) implemented. This makes it possible to evaluate how the management systems have dealt with discards before the introduction of the LO, and to evaluate the economic impacts of the LO on fishing fleets subject to shared quotas respectively ITQ management. Two criteria must be fulfilled to reach complete bioeconomic equilibrium in an open access fishery. Firstly, the marginal profit for the whole industry must be zero, based on opportunity costs including normal profit. Secondly, this solution must correspond to a point on the yield function of the fish stock biomasses, i.e., both economic and biological equilibrium must be reached, (Anderson, 1986). This is rarely the case in practice and the equilibrium is not stable. In the 211

Fisheries Research 195 (2017) 202–213

H.S. Frost, A. Hoff

fishing gear designs, grids and panels in the gear, closed areas and periods. Standard investment theory implies that if the marginal investment costs in gear changes to avoid unwanted species are lower than the marginal costs of landing unwanted species, it will pay to implement such gear changes. Secondly, the study analysed in this paper is based on fictional and not real-life data and is as such only indicative of possible tendencies rather than yielding absolute results. However, the model parameters are chosen somewhat realistically. The catch production function (Cobb Douglas) and cost structure used in the model are based on conventional fisheries economics theory and are therefore considered appropriate for the analysis (Clark and Munro, 1975; Eide et al., 2003; GarzaGil et al., 2003). However, the results are sensitive to the assumption about the catch production function. In the present context, it is assumed that the exponents of the function reflect constant returns to scale, which leads to underutilisation of quotas when the industry profit is maximized. If, on the other hand, the parameters reflected increasing returns to scale in effort, marginal profit would be positive for all efforts, and it would pay to increase effort as much as possible to obtain profit maximization and as such to full quota utilisation. Thirdly, the results are sensitive to the choice of the surplus production function (Subbey et al., 2014). The logistic Schaefer shape used here is only one type of surplus production relationship. In biological research, other types are often used, such as e.g. the more general Pella and Tomlison form that is skewed compared to the logistic function. The optimal economic solution will be affected by the actual choice of surplus production function, and again it is emphasized that the results presented in this paper provides tendencies, rather than absolute results. The analysis shows that the type of management regime is much more important economically than the introduction of the landing obligation compared to no landing obligation. In theory, introduction of ITQ leads to higher discards compared to open access because the opportunity costs of the fish stocks increase under ITQ. Hence, the fisher will discard to preserve the individual quota. On the other hand, the ITQ reduces effort, which tends to increase fish stock sizes and reduces the impact of the landing obligation in proportion to the stock size. These results are generally found in the research of the 1990s. Very recent studies (later than 2010) of the impact of the landing obligation do not distinguish between these management regimes but investigate the impact of the landing obligation compared to no landing obligation in view of the contemporary management regime and how fishers adapt to that. The results from the recent literature are ambiguous ranging from no improvement of fish stock preservation to changes in fishing patterns and choice of gear because of the introduction of the landing obligation. Despite these precautions that must be considered when interpreting the results some interesting tendencies have been highlighted in the present analyses: Generally, the adjustment of the industry is complex, but the results suggest that a higher reduction in the catch of undersized fish can be obtained through change of management system from shared quotas to ITQs, than through introduction of the LO. However, given fixed management systems, the LO will have a greater influence on a fishery managed with shared quotas than on a fishery managed with ITQs. This forms basis for future research using larger and more complex bio-economic models based on real life data.

Fig. 7. Quota utilization for each of the two species in each of the management scenarios. ‘No LO’ = No Landing obligation, ‘LO, Price > 0’ = Landing obligation and price of undersized fish > 0, ‘LO, Price = 0’ = Landing obligation and price of undersized fish = 0, ‘LO, Price < 0’ = Landing obligation and price of undersized fish < 0.

the LO may even improve both industry and individual vessel profitability, but may also have the adverse effects if it is costly to land undersized fish. Relative to this case the introduction of the LO leads to only small changes in catches in the two ITQ cases, and to insignificant profitability changes in the medium long run. In the long run, the LO may lead to reduced industry profitability but individual vessels may still profit. Thus, if a fishery is managed by shared quotas the LO may reduce the pressure on fish stocks and reduce the catch of undersized/ unwanted fish. However, depending on the cost of landing unwanted species, the LO may have both positive and negative influence on industry and vessel profitability. On the other hand, if a fishery is managed by ITQs our study suggests that it will be less affected by the introduction of the LO even though small reductions in profitability may be observed. However, in the long run the pressure on fish stocks may be reduced also under ITQ management. The strength of the model applied in the present context is that it provides simple indicative results that are more straightforward to interpret and understand than the outputs of more complex models. However, there are some things that need to be taken into consideration when reviewing the results. Firstly, it is assumed in the model that the industry complies with the quota regulation, as was also discussed in Section 5. If not, the probability of being detected and the size of the penalty will also influence the adaptation of the industry (Sutinen and Andersen, 1985). The model moreover assumes imperfect selectivity, i.e., that catch of undersized and undesirable species occurs. Measures that can be used to increase selectivity are increased mesh sizes of the fishing gear, other

Acknowledgments The authors are indebted to anonymous reviewers and to our colleague, Frank Jensen for valuable comments and suggestions. References Alzorriz, N., Arregi, L., Herrmann, B., Sistiag, M., Casey, J., Poos, J.J., 2016. Questioning the effectiveness of technical measures implemented by the Basque bottom otter

212

Fisheries Research 195 (2017) 202–213

H.S. Frost, A. Hoff trawl fleet: implications under the EU landing obligation. Fish. Res. 175, 116–126. Anderson, L.G., 1986. The Economics of Fisheries Management. John Hopkins University Press, Baltimore and London. Anderson, L.G., 1994. An economic analysis of high grading in ITQ fisheries regulation programs. Mar. Res. Econ. 9 (3), 209–226. Arnason, R., 1994. On catch discarding in fisheries. Mar. Res. Econ. 9 (3), 189–207. Batsleer, J., Rijnsdorp, A.D., Hamon, K.G., Overzeea van, H.M.J., Poos, J.J., 2016. Mixed fisheries management: is the ban on discarding likely to promote more selective and fuel efficient fishing in the Dutch flatfish fishery? Fish. Res. 174, 118–128. Borges, L., 2015. The evolution of a discard policy in Europe. Fish Fish. 16, 534–540. Boyce, J.R., 1996. An economic analysis of the fisheries bycatch problem. J. Environ. Econ. Manag. 31, 314–336. Catchpole, T.L., Frid, C.L.J., Gray, T.S., 2005. Discards in North Sea fisheries: causes, consequences and solutions. Mar. Policy 29, 421–430. Catchpole, T.L., Feekings, J.P., Madsen, N., Palialexis, A., Vassilopoulou, V., Valeiras, J., Garcia, T., Nikolic, N., Rochet, M.-J., 2013. Using inferred drivers of discarding behaviour to evaluate discard mitigation measures. ICES J. Mar. Sci. 71, 1277–1285. Clark, C.W., Munro, G.R., 1975. The economics of fishing and modern capital theory: a simplified approach. J. Environ. Econ. Manag. 2, 92–106. Condie, H.M., Grant, A., Catchpole, T.L., 2013. Does banning discards in an otter trawler fishery create incentives for more selective fishing? Fish. Res. 148, 137–146. EC, 1998. Council regulation (EC) No 850/98 of 30 March 1998 27.4.98. Off. J. Eur. Commun L 125/1. http://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri= CELEX:31998R0850&from=EN. EC, 2002. Communication from the Commission to the Council and the European Parliament: On a Community Action Plan to Reduce Discards of Fish. Commission of the European Communities, COM, Brussels (2002)656 final 26.11.2002. http://ec. europa.eu/transparency/regdoc/rep/1/2002/EN/1-2002-656-EN-F1-1.Pdf. EC, 2013. Regulation (EU) No 1380/2013 of the European Parliament and the Council of 11 December 2013 on the Common Fisheries Policy, amending Council Regulation (EC) No 1954/2003 and (EC) No 1224/2009 and replacing Council Regulation (EC) No 2371/2002 and (EC) No 639/2004 and Council Decision 2004/585/EC. Off. J. Eur. Union (L 534/22-61). http://eur-lex.europa.eu/legal-content/EN/TXT/PDF/? uri=CELEX:32013R1380&from=EN. EC, 2007. Communication from the Commission to the Council and the European Parliament—A Policy to Reduce Unwanted By-Catches and Eliminate Discards in European Fisheries. Com(2007) 136 final. SEC(2007) 380, SEC(2007) 381. http:// eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:52007DC0136& from=DA. EC, 2009. Green Paper on the Reform of the Common Fisheries Policy. Brussels, 22.4.2009, Com(2009) 163 final. http://eur-lex.europa.eu/LexUriServ/LexUriServ. do?uri=COM:2009:0163:FIN:EN:PDF. Eide, A., Skjold, F., Olsen, F., Flåten, O., 2003. Harvest functions: the Norwegian bottom trawl cod fisheries. Mar. Res. Econ. 18, 81–94. FAO, 1996a. Global Assessment of Fisheries Bycatch and Discards and the Technical Consultation on Reduction of Wastage in Fisheries (Tokyo, Japan, 28 October-1 November 1996). Fish. Tech. Pap. 339, Rome. http://www.fao.org/docrep/003/ T4890E/T4890E00.HTM. FAO, 1996b. Report of the Technical Consultation on Reduction of Wastage in Fisheries. Tokyo, Japan, 28 October-1 November 1996. FAO Fish. Rep. 547, Rome. http:// www.fao.org/docrep/009/w3696e/w3696e00.htm. Frost, H., Andersen, P., Hoff, A., 2013. Management of complex fisheries: lessons learned from a simulation model. Can. J. Agric. Econ. 61, 283–307. Garcıa, D., Prellezo, R., Sampedro, P., Da-Rocha, J.M., Castro, J., Cervinõ, S., GarcıaCutrın, J., Gutierrez, M.J., 2017. Bioeconomic multistock reference points as a tool for overcoming the drawbacks of the landing obligation. ICES J. Mar. Sci. 74 (2), 511–524. Garcia, D., Urtizberea, A., Diez, G., Giland, J., Marchal, P., 2013. Bio-economic management strategy evaluation of deep water stocks using the FLBEIA model. Aquat.

Living Resour. 26, 365–379. Garza-Gil, M.D., Varela-Lafuente, M.M., Suris-Regueiro, J.C., 2003. European hake fishery bioeconomic management (southern stock) applying an effort tax. Fish. Res. 60, 199–206. Gordon, H.S., 1954. The economic theory of a common property resource: the fishery. J. Polit. Econ. 62, 124–142. Guillen, J., Macher, C., Merzéréaud, M., Fifas, S., Guyader, O., 2014. The effect of discards and survival rate on the maximum sustainable yield estimation based on landings or catches maximisation: application to the nephrops fishery in the Bay of Biscay. Mar. Policy 50, 207–214. Hatcher, A., 2014. Implications of a discard ban in multispecies quota fisheries. Environ. Res. Econ. 58 (3), 463–472. Heath, M.R., Cook, R.M., Cameron, A.I., Morris, D.J., Speirs, D.C., 2014. Cascading ecological effects of eliminating fishery discards. Nat. Commun. https://www.ncbi. nlm.nih.gov/pubmed/24820200. Hoff, A., Frost, H., Ulrich, C., Damalas, D., Maravelias, C.D., Goti, L., Santurtún, M., 2010. Economic effort management in multispecies fisheries: the FcubEcon model. ICES J. Mar. Sci. 67, 1802–1810. NCM, 2003. Report from a workshop on discarding in Nordic fisheries. In: Valdemarsen, John Willy (Ed.), Fangstseksjonen, Havforskningsinstituttet, Bergen. Sophienberg Slot, København, 18–20 November 2002. TemaNord 2003:537. Nordic Council of Ministers, Copenhagen. Ostrom, E., 1999. Coping with tragedies of the commons. Annu. Rev. Polit. Sci. 2, 493–535. Pascoe, S., 1997. Bycatch Management and the Economics of Discarding. FAO Fish. Tech. Pap. 370. FAO, Rome. http://www.fao.org/docrep/010/w6929e/w6929e00.htm. Prellezo, R., Carmona, I., García, D., 2016. The bad, the good and the very good of the landing obligation implementation in the Bay of Biscay: a case study of Basque trawlers. Fish. Res. 181, 172–185. Quirijns, F., Pastoors, M. (Eds.), 2014. Discard Atlas of North Sea Fisheries. IMARES Wageningen UR, Wageningen. http://www.nsrac.org/wp-content/uploads/2014/ 11/discardatlas_northsea_demersalfisheries_2014.pdf. Reeves, S.A., Revill, A.S., Hutton, T.P., Pinnegar, J.K., 2008. Technical Measures in the CFP and the Reform of Council Regulation 850/98. Directorate General for Internal Policies of the Union, Policy Department B: Structural and Cohesion Polices Study IP/ B/PECH/IC/2007/090. http://www.europarl.europa.eu/RegData/etudes/etudes/ join/2008/408930/IPOL-PECH_ET(2008)408930_EN.pdf. Simons, S.L., Döring, R., Temming, A., 2015. Modelling fishers’ response to discard prevention strategies: the case of the North Sea saithe fishery. ICES J. Mar. Sci. 72 (5), 1530–1544. Subbey, S., Devine, J.A., Schaarschmidt, U., Nash, R.D.M., 2014. Modelling and forecasting stock?recruitment: current and future perspectives. ICES J. Mar. Sci 71, 2307–2322. Sutinen, J.G., Andersen, P., 1985. The economics of fisheries law and enforcement. Land Econ. 61 (4), 387–397. Turner, M.A., 1997. Quota-induced discarding in heterogeneous fisheries. J. Environ. Econ. Manag. 33, 186–195. Vestergaard, N., 1996. Discard behavior, highgrading and regulation: the case of the Greenland shrimp fishery. Mar. Res. Econ. 11 (4), 247–266. Villasante, S., Pita, C., Pierce, G.J., Guimeráns, C.P., Rodrigues, J.G., Antelo, M., Da Rocha, J.M., Cutrínj, J.G., Hastie, L., Sumaila, U.R., Coll, M., 2016. To land or not to land: how do stakeholders perceive the zero discard policy in European small-scale fisheries? Mar. Policy 71, 166–174. Ward, J.M., Benaka Moore, L.R.C.M., Meyers, S., 2012. Bycatch in marine fisheries. Mar. Fish. Rev. 74 (2) United States Department of Commerce. http://spo.nmfs.noaa.gov/ mfr742/mfr7422.pdf. Ward, J.M., 1994. The bioeconomic implications of bycatch reduction devise as a stock conservation management measure. Mar. Res. Econ. 9 (3), 227–240.

213