Legitimacy, local participation, and compliance in the Galápagos Marine Reserve

ARTICLE IN PRESS

Ocean & Coastal Management 50 (2007) 253–274 www.elsevier.com/locate/ocecoaman

Legitimacy, local participation, and compliance in the Gala´pagos Marine Reserve Ce´sar Viteria, Carlos Cha´vezb, a Natural Resources Program, Mentefactura, P.O. Box 17-22-20503, Quito, Ecuador Department of Economics, Universidad de Concepcio´n, P.O. Box 1987, Concepcio´n, Chile

b

Available online 11 September 2006

Abstract We analyze the compliance behavior of artisanal fishermen in the Gala´pagos Marine Reserve. Our empirical analysis explores the role of the reserve’s participatory management system as a determinant factor in decisions to violate regulations. The results indicate that, along with traditional enforcement tools (detection and penalties), the perception of legitimacy that regulations and local organizations have among the boat-owners, their individual sense of belonging, as well as their participation levels in their related organizations are also relevant to the compliance/violation behavior. Policy implications to improve compliance with fisheries regulations in the reserve are discussed. r 2006 Elsevier Ltd. All rights reserved.

1. Introduction The ‘‘Law of the Special Regime for the Conservation and Sustainable Development of the Province of Gala´pagos’’, enacted in 1998 provides a new frame for fishery activity management within the Gala´pagos Marine Reserve. When the special regime was enacted, its designers expected the local-based decision-making process to foster the commitment of local users to abide by the decisions and regulations designed to improve the management and conservation of natural resources. However, the results of fisheries monitoring and population analysis are far from being optimistic. Sea Cucumber densities in some zones have declined for three consecutive years, as has the catch per unit of effort of Spiny Lobster [1,2]. The reasons for this decreased performance have not yet been identified; Corresponding author. Tel.: +56 41 203 067; fax: +56 41 254 591.

E-mail addresses: [email protected] (C. Viteri), [email protected] (C. Cha´vez). 0964-5691/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.ocecoaman.2006.05.002

ARTICLE IN PRESS 254

C. Viteri, C. Cha´vez / Ocean & Coastal Management 50 (2007) 253–274

however, enforcement and compliance problems might have contributed to these undesired outcomes. In a peer review of the strategies and activities to promote compliance with regulations among the boat-owners of the reserve’s artisanal fleet, preliminary factors have been identified that impact an individual fisherman’s decision to comply with the regulations [3]. Although this descriptive study concludes that surveillance and patrolling activities detect 6% of the fleet committing transgressions of fishery regulations, the incidence of violations could be larger. According to responses obtained from a field survey directed at artisanal boat-owners operating in the reserve, the self-declared rate of regulation transgressions is close to 30% of all boat-owners. This suggests that the objective of reducing noncompliance with the agreed regulations is not limited to improved logistical or technical aspects for increased monitoring, but also requires addressing social and economic matters. The objective of this paper is to develop an empirical study of the boat-owners’ violation behavior regarding management regulations in the context of the artisanal fishery of the Gala´pagos Marine Reserve. The analysis is devoted to identifying the determinant factors motivating decisions to violate regulations, and to estimating the effects these factors have on this decision. This study places special emphasis on estimating how violation decisions are impacted by the following: the participation of boat-owners in the institutions that comprise the reserve’s participatory management system, the perception of boat-owners regarding the legitimacy of regulations, and the traditional enforcement instruments available to the authorities (detection and penalties). We found indications that, along with traditional enforcement tools (detection and penalties), the boat-owners’ perceptions of the legitimacy of regulations and local organizations, the sense of belonging of individual boat-owners, and their participation levels in their organizations are relevant to compliance/violation behavior. Two of the problems faced by the regulatory and enforcement authorities responsible for fisheries management in developing countries are the limited availability of specialized staff and restricted budgets for monitoring and regulation enforcement. Learning what determines a fisherman’s compliance/violation behavior might help reserve authorities refine their current enforcement strategies to induce adequate, affordable compliance levels. The conceptual analysis of enforcement and compliance behavior is well established in the existing fisheries economics literature [4–8]. Following the development of compliance behavior theories in the fisheries regulation context, interest in empirically analyzing the determinant factors explaining compliance decisions of individual fishermen has been growing. Unfortunately, most of the existing literature has been conducted in the context of developed countries [9–12].1 The study of an individual fisherman’s compliance behavior in the context of developing countries or within protected marine areas is significantly less documented. Interestingly, to our knowledge, only one contribution among the existing literature is devoted to the analysis of artisanal fishermen’s non-compliance in the context of a developing country [16]. The empirical analysis of the present paper is based on theoretical contributions to an individual fisherman’s compliance decision [7]. In that work, an extended compliance/ violation model is developed. The authors found that the decision to violate a regulation is 1

Other recent analyses of enforcement and compliance behavior in fisheries relate to Danish [13], Swedish [14], and North American [15] fisheries.

ARTICLE IN PRESS C. Viteri, C. Cha´vez / Ocean & Coastal Management 50 (2007) 253–274

255

not only motivated by pure traditional economic incentives and disincentives, but also by additional factors, currently emphasized by other social sciences’ disciplines. Similarly, recent literature analyzing governing systems for common pool resources and the performance of regulatory management systems for fisheries suggests that individual compliance/violation decisions do not depend only on variables such as the expected earnings of illegal activity or monitoring efforts and penalties faced by the offender, but also on social variables such as legitimacy of the regulations, local control, community membership, participation levels, degree of individual moral development, and social or peer pressure [11–13,16–19]. The paper is organized in five sections. In Section 2, we briefly describe the reserve’s current management system and artisanal fishery. Section 3 contains a description of the theoretical framework used in our empirical analysis on fisheries regulation violation decisions in Gala´pagos. This section also includes a description of the data used in the empirical analysis. In Section 4, we present the results of the econometric estimations of the boat-owners’ violation decision model. Conclusions and implications from our work to improve compliance with fishery regulations in the reserve are offered in Section 5. 2. Management system in the Gala´pagos Marine Reserve The management system in Gala´pagos is an innovative experience in the management of natural resources and one of the few examples—in Ecuador and the world—in which the management scheme incorporates the rights and responsibilities of local users in the decision-making process of the administration of a protected area [20]. The management system was established as part of the Law of the Special Regime for the Conservation and Sustainable Development of the Province of Gala´pagos (Special Regime for Gala´pagos)2 in 1998, to deal with the existing conflicts over the use of natural resources among antagonistic groups of the marine reserve and as a strategy for obtaining commitments from local users regarding the decisions and regulations designed to improve the management and conservation of natural resources, as well as to promote compliance of these regulations.3 The Special Regime for Gala´pagos contains a series of instru ctions that seek to delegate decision-making of natural-resource management, especially fishery resources, to a local coordination body constituted by representatives of the tourism sector, the artisanal fishery sector, and the conservation sector. In addition, this regime grants the Gala´pagos National Park the authority and instruments required to promote the implementation of regulations and decisions agreed to by the mentioned sectors. As for the artisanal fishery, the Special Regime for Gala´pagos establishes two main aspects aimed at regulating the use of fishery resources in the marine reserve. First, the Law considers a local management regime of common property resources with an 2

Congreso Nacional del Ecuador, Plenario de las Comisiones Legislativas. Ley de Re´gimen Especial para la Conservacio´n y Desarrollo Sustentable de la Provincia de Gala´pagos, Registro Oficial No. 278. Quito, Ecuador, 1998. 3 These aims are reflected in the vision of the new management regime elaborated by the Grupo Nu´cleo (group made up of representatives of the fishery and tourism sectors in the Gala´pagos, the Charles Darwin Foundation, the National Park and the Fisheries Bureau) in June 1997. The vision highlights effective participation of local sectors in decision-making as a key element to promote local commitment towards the compliance of the regulations [21].

ARTICLE IN PRESS 256

C. Viteri, C. Cha´vez / Ocean & Coastal Management 50 (2007) 253–274

identified and limited group of users. In this sense, the regime restricts fishery activities in the area of the reserve to small-scale (artisanal) fisherman, as defined in the Management Plan and applies entrance barriers to the fishery through the imposition of a series of requirements that affect both fishermen and boats.4 Second, the Law establishes a decision-making scheme to control artisanal fishery activities.5 The scheme considers the involvement of local fishermen in the decision-making process through the Participatory Management Board. The Board analyzes several aspects related to the development of fishery activities and subsequently makes a proposal to the Inter-institutional Management Authority. The decisions made in this regard are summarized in the fishery calendar, an instrument that contains information about which kind of species are allowed to be caught, their quotas, and fishing seasons. Once approved by the management authorities, the fishery calendar becomes the ‘‘rules of the game’’ which the artisanal fishermen must obey. Third, the Special Regime for Gala´pagos designates the Gala´pagos National Park Direction as the organism in charge of the administration and control of the reserve. The direction of the park should keep records of and regulate the number of individuals involved in the artisanal fishery, control the activities that are carried out in the reserve through a surveillance and patrol program, and establish administrative procedures against offenders and apply the penalties described in the Special Regime for Gala´pagos, should an infraction be detected. The artisanal fishery, along with tourism, is among the most important economic activities developed in the reserve. Currently, fishing generates approximately US$4–6 million annually (2002) and sustains a population of about 1000 fishermen and their families living on the islands of the Gala´pagos archipelago [22]. Fishing activity is focused on three main fisheries: (i) Sea cucumber (Isostichopus fuscus), the most important source of income for local fishermen; in 2002, this fishery generated a gross income close to 1.7 million USD (larger than the income generated by the Spiny lobster or whitefish fisheries), with a total catch of 8.3 million individuals [22]. The sea cucumber is harvested to a depth of approximately 20 m in eight specific macro-zones of five different islands. The fishermen cook, salt, and dry the harvested sea cucumbers. The most important 4 Servicio Parque Nacional Gala´pagos. Plan de Manejo de Conservacio´n y Usos Sustentable para la Reserva Marina de Gala´pagos. Parque Nacional Reserva Marina de Gala´pagos Ecuador. Puerto Ayora, Ecuador, 1998. Small-scale fishing is characterized by restrictions imposed on boats, including banning certain fishing gear and prohibitions against using mechanical equipment. Entrance barriers include a moratorium for issuing new boat permits and a restriction only granting new fishing permits to fishermen’s kin. 5 The decisions agreed upon to regulate the small-scale fishery stem from the participatory management system established by the Special Regime for Gala´pagos and the Management Plan, which works at three levels: (1) the National Park, which is responsible for the administration of the area and the control of natural-resource management; (2) the Inter-Institutional Management authority, which is comprised of institutions which have competence and control over the area, and is also the top level agency for policy making regarding reserve management; and (3) the Participatory Management Board, which channels the responsible participation of local reserve users, including small-scale fishermen. Decision-making begins with a participatory process in which local users of the reserve submit their proposals to the board to be discussed. The decisions agreed upon by consensus in the Participatory Management Board are sent to the management authority for approval by voting. Within the Board, the National Park behaves as any other stakeholder seeking a consensus; in the management authority, the National Park serves as a Technical Secretariat. Then, the National Park is responsible for implementing the decisions made, with the support of the co-responsible stakeholders in the system [20].

ARTICLE IN PRESS C. Viteri, C. Cha´vez / Ocean & Coastal Management 50 (2007) 253–274

257

management tools applied in this fishery include no-take zones, restricted fishing seasons (2 months, April–May), size restrictions, and a total allowable catch (TAC) quota. (ii) Spiny Lobster, the second largest in terms of economic relevance; in 2002, the fishery yielded revenues of 1.2 million USD with a catch of 51.4 metric tons (mt) [22]. Three lobster species are caught: Spiny Red Lobsters (Panulirus penicillatus), Green Lobsters (Panulirus gracilis), and Slipper Lobsters (Scyllarides astori). Lobsters are handharvested by divers; only the tails of the lobsters are marketed and most of them are frozen and exported. The management of this fishery comprises a fishing season from September to December, a minimum legal size of 15 cm tail length, gear restrictions (lance- and pistol-type harpoons are prohibited), and a ban on the harvest of eggbearing females. (iii) Whitefish, less important in economic terms than the previous fisheries; in 2000 the total catch was 374.2 mt, in economic terms, approx. 0.7 US million [22]. Landings comprise a wide range of species and are year-round. Management instruments include gear restrictions, such as a longline ban and shark catch prohibition. Most potential fishing violations include infringement of one or several of the management regulations mentioned for the three fisheries. According to the National Park patrolling reports, the more frequent violations include fishing in no-take zones and illegal catching and trading of shark. Another violation is the employment of individuals who are not in the park records. Fishing activity in the reserve has generated great international attention, primarily for its potential impact on the biodiversity of the marine reserve. 3. Behavior of the fishery boat-owner In this section, we describe the theoretical model, on which our empirical analysis is based, for the individual decisions to infringe the regulations. We briefly discuss a deterrence model, where the infraction decision is motivated by purely economic variables (expected costs and benefits of violating regulations) as well as social variables, which we consider to be decisive for the violation decision. Then, we specify an econometric model to carry out the empirical analysis. 3.1. Individual regulation violation decisions Our conceptual understanding of the individual decision to violate a regulation is based primarily on the contribution of Sutinen and Kuperan [7]. The model attempts to explain an individual fisherman’s behavior facing the non-compliance decision, and it merges the basic deterrence model with intrinsic and extrinsic motivations that influence the individual’s decision of whether to obey the law. Specifically, to evaluate the influence of social factors on the decision of whether or not to comply, this model extends the classical deterrence model of Becker [23] by explicitly considering two additional factors: the moral and social reputation of the individual. The first factor describes how the compliance decision is influenced by individual’s moral development, personal values, and perception of the legitimacy of the norms.

ARTICLE IN PRESS 258

C. Viteri, C. Cha´vez / Ocean & Coastal Management 50 (2007) 253–274

The second factor attempts to reflect how social pressures inside the community affect the individual’s decision.6 Likewise, it has been suggested [24] that the compliance with social norms will depend on, among other aspects, the level of legitimacy of the norm, the institutions which promote that norm, and the level of membership the individual feels to the community. It has also been underlined [11] that current fishery management literature with a local focus on ‘‘co-management’’ frequently suggests that broader participation of fishermen in the management process results in an increment of compliance levels due to the regulations being agreed upon and having legitimacy. This can be understood as the individual feeling a moral commitment towards the compliance and success of the regulations, under the premise that when an individual identifies with a regulation and does not feel it to be an obstacle to his/her freedoms imposed by some external body, this regulation is more likely to be successful.7 Including an individual’s social reputation and commitment to the institutional framework has two effects [7]. First, the number of offenders decreases because the condition for the decision to violate is stricter than in the simple case of the classic deterrence model. Second, since the marginal costs of infringing are larger in the extended model than the marginal costs of infringing in the context of the classic deterrence model, the magnitude of the infraction decreases because fishermen allocate less effort to illegal activities. Thus, following the analysis of the fisherman’s compliance decision in other studies [7,11,12,16,18], and considering the participatory management system of the reserve and its potential effects on the compliance decision, we consider a set of social factors that we presume influence the individual’s behavior: legitimacy of regulations and institutions related to fishery management; the boat-owner’s sense of belonging to the community; and the boat-owner’s participation in local organizations. 3.2. Hypothesis The empirical analysis is guided by the theoretical response of a boat-owner’s compliance behavior when changes are perceived in independent variables. We denote V* as the difference between the maximum expected utility transgressing the regulations and the maximum utility the individual gains by obeying the regulations [7]; that difference is likely to depend on variables such as the price of landings (p), the surveillance effort from the regulatory authority (probability of detection and conviction) (v), the magnitude of the fine ðf¯ Þ, the cost structure linked with the characteristics of the vessel (size, power, ¯ the legitimacy of the current regulations (l), the sense employees, storage capacity, etc.) ðkÞ, of belonging of the individual in the community (s), the participation level that the individual has in the institutions involved in the creation of the norm (u), and a set of 6

Along the same lines, the non-monetary factors that affect the compliance of a catch quota among fishermen in the UK were analyzed [11]. This model, as with others [7,18], is an extension of the basic deterrence model to evaluate the influence of moral obligation, peer attitudes, and perceived legitimacy of the norm on compliance. The authors find some evidence that a better level of involvement of individuals in the management process, leads to better compliance levels since the regulations will be accepted with more legitimacy. 7 A similar proposition has been concluded elsewhere [25]. In addition, the issue of lost freedom due to external regulations and their effect on an individual’s behavior in the framework of common property resources has been addressed [17].

ARTICLE IN PRESS C. Viteri, C. Cha´vez / Ocean & Coastal Management 50 (2007) 253–274

259

characteristics for each individual ð¯aÞ, i.e., ¯ l; s; u; a¯ Þ. V  ðp; v; f¯ ; k;

(1)

We interpret V* as the magnitude of the incentive that individuals have to not obey the rules and to allocate effort to illegal fishing activities; a positive difference indicates incentives to violate. It can be shown that the magnitude of the incentive to violate increases with the financial incentive (p) and diminishes as result of an increase in the surveillance effort from the regulatory authority (detection and conviction probability) as well as for positive changes in the magnitude of the penalty. The same effect is observed when we consider social variables; thus, the incentive to violate decreases with the legitimacy of the regulations, the sense of belonging of the individual, and the individual’s participation levels in the institutions that generate the regulations.8 3.3. Econometric model specification and data description In this section, we are interested in evaluating the violation probability of an agreed regulation by a boat-owner of the artisanal fishery fleet of Gala´pagos. The general equation we use to evaluate the decision to violate regulations in the reserve is9 Violaltion ¼ Violation ½Detection and Penalty; Price of Landings; Costs; Individual Characteristics; Legitimacy; Belonging to the Community; and Participation,

ð2Þ

where Violation is equal to one if the fisherman violates the regulations; otherwise the variable has a value of zero. Considering the dichotomous nature of the dependent variable, we estimated the parameters of interest by using a probit model. Assuming a normal distribution of errors, the probability of observing Violation ¼ 1 is given by the value of the normal distribution function evaluated with the estimated parameters from the estimation procedures for Eq. (2). The independent variables considered are classified in seven groups. Detection and Penalty includes a set of variables intended to capture the individuals’ perception of the likelihood of being detected in violation and if so, being punished, and the individual fisherman’s perception about the magnitude of the penalty. Price of Landings is a variable that attempts to capture the financial incentive to violate the regulations. Costs is a set of variables through which we control for the influence of the operational cost of fishing activity on the violation decision. Individual Characteristics includes variables for several individual characteristics of each fisherman. Legitimacy includes a set of variables for the perception of the legitimacy of a regulation and the authorities that promote it. Finally, Belonging to the Community and Participation is a group of variables that represents the individual sense of belonging to the local community as well as the participation of the fisherman in her/his organization (cooperative). The data used in the estimations comes from a field survey applied to a sample of boatowners from the artisanal fleet operating in the reserve. The survey was conducted by one of the co-authors of this work during September 2002, spending about 1 week in each of the three main inhabited islands. Boat-owners were selected randomly at landing sites. 8

Comparative static results [7]. We avoid the use of a sub-index to simplify notation. The list of specific variables and their definitions are presented in Table 1a. 9

ARTICLE IN PRESS 260

C. Viteri, C. Cha´vez / Ocean & Coastal Management 50 (2007) 253–274

The survey was carried out thanks to the collaboration of the Gala´pagos Artisanal Fishermen Cooperatives: COPESPROMAR (all acronyms in this paragraph are in Spanish), Cooperative of Fishery Production ‘‘San Cristo´bal’’(COPESAN), Cooperative of Fishery Production Horizons of Isabela, and Cooperative of Gala´pagos Artisanal Fishery Production (COPROPAG), and the Participatory Management Board. The survey was confidential; as such, no data that could reveal the identity of the boat-owner was requested. In addition, the survey design allowed the boat-owner to fill out the survey independently,10 even though most interviewed fishermen preferred a diallog. We obtained 155 observations from a population of 426 boat-owners registered by the Unit of Marine Resources of the Gala´pagos National Park.11 Our empirical analysis is based on the fishermen’s responses to the survey questions; therefore, non-compliance status is reported as non-compliance. We acknowledge that this implies that we are assuming that fishermen provided truthful answers. Although the assumption is unavoidable, the experience doing field work resulted in the perception that the responses are reliable. This is the perception of the person who conducted all the surveys, using face-to-face interviews/diallog. The survey included several sections addressing different topics. The first section requested information about the characteristics of the boat-owners and their families. The second section asked for data about technical characteristics of the boat and details of the fishery activity. The third section inquired about the boat-owners’ perception of the surveillance activities performed by the authority. The fourth section asked the boatowners about their behavior regarding compliance with regulations; the last section asked boat-owners about their perception regarding the fishery regulations, the management authorities, the fishermen cooperatives, and participation in these organizations. In the compliance section, the survey asked boat-owners to choose a sentence that best described their fishery activity in the last season (2001 was the reference year); it also asked them directly if, in the past season, the boat-owner had committed some of the most frequent infractions according to the records of the National Park’s patrolling reports. Thus, with the information collected, we proceeded to classify the boat-owner by her/his self-reported compliance status. We defined a violator to be any individual who stated that, in the last season, their fishery activities ‘‘rarely fulfill the regulations, because there are too many restrictions’’, or who stated directly that they have carried out some activity judged as an infraction. Table 1a presents the list of variables considered in the econometric model. The table is divided in two sections. The first section shows the dependent variable (Violation), representing the violation decision. The second section shows, separated in blocks, the explanatory variables used in the estimations. Each block represents an element of the ¯ p; a¯ ; l; s; u. vector of independent variables specified in the theoretical model: v; f¯ ; k; Block I ‘‘Probability of Detection and Punishment (v)’’ contains the variables ZONE, MOTORHP, AGE, SEEN, SEA_INSPECTION, INSPECTED_PUNISHED, and PUNISHED; and attempts to represent the perception of each boat-owner regarding 10

To maintain confidentiality, the results of this paper do not refer to any specific island or town. We employed the stratified sampling technique [26]. The sample was divided into three strata according to boat type (bote, fibra and panga); to make this division we used the detection variances observed in each boat type group of boat-owners. Additionally, the selected strata were distributed proportionally among the three towns. The sample size resulted in 148 observations of a population of 426 boat-owners, with a theoretical error of 3% and a confidence level of 95%. 11

ARTICLE IN PRESS C. Viteri, C. Cha´vez / Ocean & Coastal Management 50 (2007) 253–274

261

Table 1(a) Description of the variables used in the econometric model Name of the variable and descriptiona Dependent variable VIOLATION: 1: If the boat-owner acknowledges: to have carried out an activity considered infraction; rarely fulfilling regulations or having made some infraction. 0: otherwise. Independent variables I. Probability of detection and punishment (v) ZONE: 1: If the boat-owner fishes in North, North West, or West zones or declares to operate in the whole archipelago. 0: otherwise. MOTORHP: Defined as the power of the boat ‘s motor; this variable is expressed in horse power (HP). AGE: Age of the boat-owner in years. SEEN: 1: If the boat-owner states having seen the surveillance of the GNP or the National Army: always (A), almost always (AA) or sometimes (ST). 0: otherwise. SEA_INSPECTION: 1: If the boat-owner states having been boarded for inspection A, AA or ST by personal of the GNP or the National Army. 0: otherwise. INSPECTED_PUNISHED: 1: If the boat-owner considers that ships detained for infractions are punished A, AA or ST. 0: otherwise. PUNISHED: 1: If the boat-owner indicates having been punished at some time. 0: otherwise. II. Magnitude of the fine. (f¯ ) FINE: 1: If the boat-owner considers that the fines applied to offenders are high. 0: otherwise. ¯ III. Structure of costs kÞ STORAGE: Indicates the storage capacity of the boat in metric tons (mt). DAYCOST: Indicates the daily cost in American dollars of a fishery trip. CAPITAL: Indicates the commercial value of the ship in American dollars. INDEBT: 1: If the boat-owner declares having a debt related to fishery activity. 0: otherwise. IV. Structure of prices (p) PRICEIND: Price Index based on the market prices of the species the boat-owner caught, weighted by the share of each species in the total income of boat-owners. The index is expressed in American dollars. V. Individual characteristics (¯a) MARRIED: 1:If the boat-owner is married or cohabiting. 0: otherwise (single, separate, widow). FAMILY: The number of boat-owner family members. EDUCA: Number of years of formal education completed. ISLAND1: 1: If the boat-owner lives on the Islands. 0: otherwise. VI. Legitimacy of the regulations and authorities that promote them (l). REGULA_GRL: 1: If the boat-owner strongly agrees (SA), agrees (A) or neither agrees nor disagrees (NA/ ND) with the fishery regulations. 0: otherwise REGULA_SHARK: 1: If the boat-owner strongly disagrees (SD), disagrees (D) or NA/ND, with allowing shark fishery. 0: otherwise. REGULA_SEASON: 1: if the boat-owner strongly agrees (SA), agrees (A) or NA/ND with the statement: ‘‘nocatch seasons benefits the fisherman’’. 0: otherwise. REGULA_SEACUCUMBER: 1: If the boat-owner strongly disagrees (SD), disagrees (D) or NA/ND with allowing unregulated fishery of sea cucumber. 0:otherwise. OTHER: 1: If the boat-owner declares that other fishermen usually make some infraction of those commonly detected by the GNP. 0: otherwise. JUNTA_AUTHORITY: 1: If the boat-owner considers that the JMP/AIM takes into account the sector fishery little time, most of the time or every time. 0: otherwise. LEAD_REPRESENT:1: If the boat-owner considers that his leaders represent the interests of the sector appropriately or at least they represent them fairly. 0: otherwise.

ARTICLE IN PRESS 262

C. Viteri, C. Cha´vez / Ocean & Coastal Management 50 (2007) 253–274

Table 1(a) (continued ) VII. Sense of membership to the community (s). NATIVO: 1: The boat-owner was born in the Gala´pagos. 0: otherwise MOVECONT: 1: If the boat-owner declares that he could change his residence to the continent. 0: otherwise YEARSMIGR: 1/Number of years of immigration. FISH_EXC: 1: If the boat-owner is exclusively devoted to the fishery activity. 0: otherwise YEARSEXP: Years of experience that the boat-owner has in the fishery activity. VIII. Participation (or) COOPDIR:1: If the boat-owner is a representative or occupies some directive position in the Cooperatives. 0: otherwise. ATTEND:1: If the boat-owner attends all the meetings of the cooperative. 0: otherwise. Source: Elaborated by the authors. a The variables ZONE, SEEN, SEA_INSPECTION, INSPECTED_PUNISHED, MARRIED, ISLAND1, REGULA_GRL, REGULA_SHARK, REGULA_SEASON, REGULA_SEACUCUMBER, JUNTA_AUTHORITY, and LEAD_REPRESENT that originally were of multiple discreet answers have been transformed to binary discreet variables, with the purpose of increasing the degrees of freedom in the regression and of obtaining a better behaved model.

the probability that an offender is detected and punished. Block II ‘‘Magnitude of the fine (f¯ )’’ contains the variable FINE, which represents the size of the fine perceived by the boat¯ includes STORAGE, DAYCOST, CAPITAL, owners. Block III ‘‘Structure of Costs (k)’’ INDEBT; and indicates the cost structure faced by the boat-owners. Block IV ‘‘Structure of Prices (p)’’ has the variable PRICEIND; an indicator of the market prices of the targeted species as observed by the boat-owner. We used this variable to indicate the financial incentive faced by individual fisherman to violate the regulation.12 Block V ‘‘Individual Characteristics (¯a)’’ contains MARRIED, FAMILY, EDUCA, ISLAND1; these describe the personal features, family characteristics education, and place of residence of each boat-owner. Block VI ‘‘Legitimacy of the Regulations and Authorities that promote them (l)’’ includes: REGULA_GRL, REGULA_SHARK, REGULA_SEASON, REGULA_SEACUCUMBER, OTHER, JUNTA_AUTHORITY, LEAD_REPRESENT; which represents the degree of legitimacy that regulations and relevant authorities have among the boat-owners. Block VII ‘‘Sense of Ownership to the community (s)’’ contains the variables: NATIVO, MOVECONT, YEARSMIGR, FISH_EXC, YEARSEXP; and represents the level of membership of the boat-owner towards his town and union. Finally, block VII ‘‘Participation (u)’’, includes: COOPDIR, ATTEND, attempts to indicate the degree of involvement of the boat-owner in the local organizations (cooperatives). Table 1b presents descriptive statistics (mean, standard deviation, and the coefficient of variation) for the variables considered in the model.

12

The construction of a variable that represents the expected benefit from violating differs among the related empirical literature according to the type of violation analyzed. For example, the financial incentive to violate variable is defined as the difference in the value of catch per unit of effort between inshore and offshore areas [18]; as restriction of earning (in percentage) due to quotas [12]; they also consider a binary variable for leasing in extra quota. Finally, the expected difference in value of catch per crew member between illegal and legal mesh size regulation has also been considered [16].

ARTICLE IN PRESS C. Viteri, C. Cha´vez / Ocean & Coastal Management 50 (2007) 253–274

263

Table 1(b) Description of the variables used in the econometric model Variable Dependent variable VIOLATION

Average

0.305

Standard deviation

0.462

Independent variables I. Probability of detection and punishment (v) ZONE 0.760 0.429 MOTORHP 68.554 42.452 AGE 40.507 10.640 SEEN 0.409 0.493 SEA_INSPECTION 0.305 0.462 INSPECTED_PUNISHED 0.409 0.493 PUNISHED 0.292 0.456 II. Magnitude of the fine (f¯ ) FINE 0.481 0.501 ¯ III. Structure of costs (k) STORAGE 1.841 2.430 DAYCOST 108.920 115.591 CAPITAL 43746.200 103634.000 INDEBT 0.468 0.501 IV. Structure of prices (p) PRICEIND 2.412 1.726 V. Individual characteristics (¯a) MARRIED 0.877 0.330 FAMILIA 3.500 1.958 EDUCA 10.058 3.869 ISLAND1 0.286 0.453 VI. Legitimacy of the regulations and authorities that promote them (l) REGULA_GRL 0.435 0.497 REGULA_SHARK 0.305 0.462 REGULA_SEASON 0.909 0.288 REG_SEACUCUMBER 0.708 0.456 OTHER 0.721 0.450 JUNTA_AUTHORITY 0.552 0.499 LEAD_REPRESENT 0.273 0.447 VII. Sense of ownership to the community (s) NATIVO 0.474 0.501 MOVECONT 0.084 0.279 YEARSMIGR 0.023 0.028 FISH_EXC 0.805 0.397 YEARSEXP 19.604 10.447 VIII. Participation (u) COOPDIR 0.273 0.447 ATTEND 0.370 0.484 Source: Elaborated by the authors based on the survey of September 2002.

Coefficient of variation

1.51

0.56 0.62 0.26 1.21 1.51 1.21 1.56 1.04 1.32 1.06 2.37 1.07 0.72 0.38 0.56 0.38 1.58 1.14 1.51 0.32 0.64 0.62 0.90 1.64 1.06 3.32 1.22 0.49 0.53 1.64 1.31

ARTICLE IN PRESS 264

C. Viteri, C. Cha´vez / Ocean & Coastal Management 50 (2007) 253–274

4. Econometric results In this section, we present the econometric results for the violation decision of fishery management regulations in the Gala´pagos Marine Reserve. The section concludes with the quantification of the impact of relevant independent variables on the individual violation decision.

4.1. Violation decision model results Considering the nature of the independent variable, we estimated the parameters of interest by using a probit model. Table 2 presents the violation decision model results for two alternative specifications. The first model specification (Model 1) is estimated according to the original specification of Eq. (2). The second specification (Model 2) is a restricted version of the previous one. We decided to estimate a more parsimonious specification, which excluded certain variables. The restricted model was obtained after experimenting with several alternative specifications. Highly correlated variables were used as exclusion criteria. Furthermore, we also performed tests for omitted variables on those variables included in the original specification that were not statistically significant. In addition, for the second specification, we also decided to exclude the variable PUNISHED.13 In this way, a simpler specification was obtained.14 From a global perspective, the estimated models are statistically significant. We evaluated the null hypothesis (all parameters equal to zero) against the alternative hypothesis (all parameters different from zero). For this evaluation, we used the criteria of the maximum-likelihood statistics and the percentage of certainty. First, due to the fact that the maximum-likelihood statistic is higher than the critical value for the two estimated models, we reject the null hypothesis that the parameters of the independent variables are all equal to zero; consequently, both models are relevant when explaining the decision to violate by the artisanal fishermen operating in the reserve. Second, the models predict correctly 85.5% and 80.5% of the observations for the first and second estimated specification, respectively. This tells us that both models, based on the information of the independent variables, can correctly forecast the individual’s violation decision in more than 80% of the observations. 13 This variable might not adequately represent the effect that we wish to represent; namely, that the existence of previous experiences with authorities increase the likelihood of being punished, which would thus have an effect on the probability of being detected and punished. Furthermore, even if we are able to capture that effect, there are endogeneity/causality problems with this variable, as being punished in the past depends in part on being a violator. 14 The variables with high correlation were: YEARSMGR with NATIVO; and YEARSEXP with AGE. Highly correlated variables were considered to be those variables with correlation coefficients greater than 0.5. Furthermore, in order to avoid the omitted variable problem, an omitted variable test was carried out before the elimination of each variable; here, the evaluated null hypothesis H0:bi ¼ 0, against the alternative hypothesis H1:bi6¼0. For this, we used the maximum-likelihood MV ¼ 2:ðln LR  ln LNR Þ, which has a w2 distribution; its critical value with 1 degree of freedom is 3.84 (at 5% level of significance). As a result, H0 was not rejected. Furthermore, the Wald-statistic was calculated to evaluate the null hypothesis that the parameters of the 20 variables eliminated as one group are all equal to zero. The value obtained was 22.44, for which the null hypothesis is not rejected (critical value at the 5% significance level is w2ð20Þ ¼ 31:41); that is to say, the 20 parameters of the variables equal zero as a group.

ARTICLE IN PRESS C. Viteri, C. Cha´vez / Ocean & Coastal Management 50 (2007) 253–274

265

Table 2 Estimation results Block

Variables

Model 1

Model 2

Coefficient

‘‘t’’ Std.

Coefficient

‘‘t’’ Std.

CONSTANT

1.077

0.608

0.695

0.807

ZONE MOTORHP AGE SEEN SEA_INSPECTION INSPECTED_PUNISHED PUNISHED

0.402 0.003 0.004 1.003 0.416 0.055 0.759

1.014 0.494 0.158 2.307 1.022 0.151 1.895

0.528

1.793

II

FINE

1.154

2.695

0.917

2.959

III

STORAGE DAYCOST CAPITAL INDEBT

0.177 0.004 0.263 0.356

2.020 1.626 1.162 1.020

0.144

2.590

IV

PRICEIND

0.037

V

MARRIED FAMILIA EDUCA ISLAND1

0.797 0.009 0.125 0.843

1.407 0.099 2.359 1.987

0.127 1.091

2.870 3.253

REGULA_GRL REGULA_SHARK REGULA_SEASON REG_SEACUCUMBER OTHER JUNTA_AUTHORITY LEAD_REPRESENT

1.075 0.368 0.025 1.174 0.723 0.155 1.378

2.548 0.822 0.043 2.831 1.686 0.418 3.174

NATIVO MOVECONT YEARSMIGR FISH_EXC YEARSEXP

0.856 0.115 21.529 0.459 0.047

1.531 0.172 2.184 0.954 1.885

0.033

2.122

COOPDIR ATTEND

0.299 1.048

0.726 2.805

0.864

2.954

No. of observations w2 Critical value at 5% % correct

154 81.11 w2(31) ¼ 44.99 84.4

I

VI

VII

VIII

0.416

0.710

2.352

0.947 0.782

2.969 2.281

1.276

3.418

15.040

2.667

154 62.85 w2(31) ¼ 21.03 80.5

Source: Elaborated by the authors from the econometric estimation results of the models.  Coefficient significant at 10%, two-tail test. Coefficient significant at 5%, two-tail test. Coefficient significant at 1%, two-tail test.

ARTICLE IN PRESS 266

C. Viteri, C. Cha´vez / Ocean & Coastal Management 50 (2007) 253–274

Interestingly, most of the individual parameters were estimated as statistically significant and had the expected signs according to the related literature. If we consider the parameters that turned out to be relevant, taking into account the blocks in which the independent variables were, it can be noted that all blocks had at least one relevant variable, except for Block IV ‘‘Structure of prices’’. That is, factors considered in the theoretical model are relevant to the individual’s decision to violate fisheries regulations in the reserve. In Block I ‘‘Probability of Detection and Punishment’’, the variable SEEN was estimated as significant at the usual levels. For Block II ‘‘Magnitude of the fine (f¯ )’’, FINE was also statistically significant. In principle, it is possible that individuals violate/comply with the management regulations because they think the fine is low/high. However, it is also possible that individuals respect regulations regardless of the existence of the fine. In fact, our data set reveals that not all violators consider the fine to be low, nor do all compliers report the fine to be high. The estimation allows us to compare individual perceptions with respect to the fine and test if those perceptions are relevant to the violation decision. The results indicate that those perceptions are indeed relevant. Additionally, perceptions are important in a person’s decision-making process, as they contain different information handled by, and characteristics of the, individual.15 ¯ In the case of Block III ‘‘Structure of Costs (k)’’, only the variable STORAGE turned out to be relevant for the violation decision; this can be explained by the correlation that exists between the variables of this block, since, for example, a large storage capacity is related to high costs and a higher capital investment. In the case of Block IV, which contains the PRICEIND, we obtained the expected sign for the associated estimated parameter; however, we were not able to estimate it as statistically significant. We think that, due to the way in which the variable was constructed, it probably does not adequately reflect the economic incentives for the boatowner to violate the wide range of regulations considered.16 The results obtained in these first four blocks confirm the relevance of the probability of detection, the amount of the penalties, and some of the economic considerations, as determinant in the decision to violate. The variables ISLAND1 and EDUCA in Block V, ‘‘Individual Characteristics (¯a)’’, were estimated as statistically significant with positive signs. This is to say that individuals that reside on that specific island or those with a higher level of education are more likely to infringe than the rest of the boat-owners who do not display these characteristics. For Block VI, ‘‘Legitimacy of the Regulations and Authorities that promote them’’, the following variables were relevant for the non-compliance decision: REGULA_GRL, REGULA_SEACUCUMBER, OTHER, and LEAD_REPRESENT. Block VII, ‘‘Sense of Membership of the community(s)’’, contains various relevant parameters (YEARSMIGR and 15

We acknowledge; however, that there might be some problems for the comparisons. First, what a high or low fine means might vary among individuals. Second, it might also be that it is only after experiencing the compliance decision that individuals form their perception about the extent of the fine. Survey questions asked about compliance behavior during the past season, therefore, it might be the case that for some individuals the size of the fine was perceived to be high/low depending on their self-reported compliance behavior. 16 As noticed before, the variable PRICEIND (price index) is constructed based on market prices of the target species for boat-owners who operate in the reserve. The prices are weighted by the percentage that each species represents in the total harvest. As mentioned earlier, other authors who analyze the effects of financial incentives on compliance behavior focused on one or two types of specific violations and attempted to reflect the economic incentive for that specific type of violations.

ARTICLE IN PRESS C. Viteri, C. Cha´vez / Ocean & Coastal Management 50 (2007) 253–274

267

YEARSEXP). For the last block ‘‘Participation (u)’’ the variable ATTEND was statistically significant. The results obtained from blocks VI–VIII are especially relevant because they tend to support some of the previously analyzed evidence on the importance of social factors, such as legitimacy, membership, and participation on the decision to violate fisheries regulations. 4.2. Effect of the independent variables on the violation decision Considering the results obtained by the estimation of the violation model, we decided to quantify the impact of the relevant independent variables on the violation probability. Table 3 shows the results of the marginal effects obtained for each independent variable on the violation decision. Because we did not find significative differences between the results obtained for the marginal effects associated with the estimated specifications, we decided to present results only considering Model 2.17 In Block I ‘‘Probability of Detection and Punishment (v)’’, the variable SEEN has a marginal effect over the probability of infringement of 0.14, which indicates that boatowners who report having seen, with some frequency, National Park and the National Navy surveillance have a 0.14 less probability of being a violator. For Block II ‘‘Magnitude of the fine (f¯ )’’, the variable FINE has a marginal effect of 0.25, whereas in Block III ¯ ‘‘Structure of Costs (k)’’, the proxy STORAGE reaches a marginal effect of 0.04. The former result suggests that those individuals that consider the penalties to be high have a 0.25 lower probability of being a violator; the latter indicates that the probability of violating increases by 0.04 for each extra metric-ton increment in boat storage capacity. Considering Block V, ‘‘Individual Characteristics (¯a)’’, for variables EDUCA and ISLAND1, we estimated marginal effects of 0.04 and 0.30, respectively. This indicates that, for each additional year of schooling, the probability of a fisherman being a violator increases by 0.04.18 Also, if the boat-owner is a resident on Island 1, the probability of being a violator increases by 0.30. As for Block VI ‘‘Legitimacy of the Regulations and the Authorities that promote them (l)’’, we present the marginal effects estimated for the variables: REGULA_GRL (0.19), REGULA_SEACUCUMBER (0.26), LEAD_REPRESENT (0.35), and OTHER (0.21). These results indicate that a boat-owner who declares some affinity with the regulations in general (REGULA_GRL), or with the seacucumber regulations (REGULA_SEACUCUMBER), has a probability of being a violator that is 0.19 and 0.26 lower, respectively. The same occurs with a fisherman who considers that the leaders of his/her organization represent his/her interests. In this case, the probability of being a violator was estimated to decrease by 0.35. The opposite occurs 17

We calculated the marginal effect of each continuous independent variable on the probability of violation as the impact of one additional unit of the independent variable with the other explanatory variables held constant (mean levels). In the case of a binary independent variable, say xj, we calculated the marginal effect on the violation probability using the expression Prob (Violation ¼ 1jx, ¯ xj ¼ 1)Prob (Violation ¼ 1jx, ¯ xj ¼ 0); where x¯ represents the vector containing the sample mean of the other explanatory variables in the model, and xj has a value of 1 or 0, for a discrete change, respectively. 18 Perhaps surprisingly, this result is qualitatively similar to that obtained for the effect of education on the probability that a fisherman would have been in violation in the Lake Victoria artisanal fisheries [21]. In their study, they were able to estimate that the probability for a fisherman being a non-violator decreased by 2% for each additional year of schooling.

ARTICLE IN PRESS 268

C. Viteri, C. Cha´vez / Ocean & Coastal Management 50 (2007) 253–274

Table 3 Marginal effects of the independent variables Block

Variables

Model 2 Marginal effect

I

SEEN

0.144

II

FINE

0.250

III

STORAGE

0.039

V

EDUCA ISLAND1

0.035 0.297

VI

REGULA_GRL REG_SEACUCUMBER OTHER

0.194 0.258 0.213

VII

LEAD_REPRESENT YEARSMIGRa YEARSEXP

0.348 0.084 0.010

VIII

ATTEND

0.236

Source: Elaborated by the authors with the econometric estimation results. a The estimated marginal effect for this variable shows the impact on the violation probability of a unitary reduction in the number of years of immigration from the sample mean (see variable definition in Table 1a).  Coefficient significant at 10%, two-tail test.  Coefficient significant at 5%, two-tail test.  Coefficient significant at 1%, two-tail test.

with a fisherman who declares that other fishermen commit infractions (OTHER), which increases the probability of being a violator by 0.21. In Block VII ‘‘Sense of membership within the community (s)’’ the variables YEARSMIGR and YEARSEXP obtained marginal effects of 0.084 and 0.01, respectively. We notice here that YEARSMIGR is an immigration index that is constructed as: 1/ (Number of years with residency in the Gala´pagos). The index adopts the value of zero in the case of natives and one for recent immigrants with a residency of 1 year. This index is not very sensitive to the variance of years of residence on the islands. Thus, we see that for the interval of migration years observed in the sample—1 to 53 years—the average increment of the index is very small (0.0185) for each year less of residency. According to our estimations, a unitary change in the number of years of immigration from the sample mean leads to an increase of 8% in the probability of violation. For example, if we consider the extreme case of a recent arrival, that is to say, an individual with an immigration indicator equal to 1, this individual exhibits a violation probability close to or equal to the upper limit of the probability interval (i.e., 1) which will always be higher than any other fisherman who has resided on the islands for a longer time. In addition, our estimation for the marginal effect of YEARSEXP indicates that a unitary increase in the years of fishery experience reduces the infringement probability by 1%. Finally, in Block VIII ‘‘Participation (u)’’, the variable ATTEND has a marginal effect of 0.24, which suggests that the boat-owners who participate in his/her organization by attending the cooperative meetings reduce the probability of being a violator by about 0.24.

ARTICLE IN PRESS C. Viteri, C. Cha´vez / Ocean & Coastal Management 50 (2007) 253–274

269

Using the coefficients obtained by the econometric estimation of the violation model for the second specification (Model 2, in Table 2) and considering the sample mean of the continuous variables reported in Table 1b, it is possible to write the violation probabilities as ProbabilityðViolation ¼ 1Þ ¼ F½0:695  0:528  SEEN  0:917  FINE þ 0:144  1:841 þ 0:127  10:058 þ 1:091  ISLAND1  0:710  REGULA_GRL  0:947  REGULA_SEACUCUMBER þ 0:782  0:721  1:276  LEAD_REPRESENT þ 15:04  YEARSMIGR  0:033  19:604  0:864  ATTEND,

ð3Þ where F represents the normal standard distribution function. We use expression (3) to evaluate violation probabilities. Table 4 presents the estimated violation probabilities for various categories of boat-owners. Initially, the boat-owners were divided in two groups: recent immigrants (7.5 years of residence19) and natives of the Gala´pagos Islands.20 This first differentiation between individuals is motivated by the fact that the marginal effect of YEARSMIG was estimated to be strong. Additionally, within the immigrant and native groups, individuals were differentiated according to their perception of the authorities’ surveillance efforts, their perception of the size of penalties, their place of residence, their affinity to the regulations and leadership of fishery co-operatives, and their participation in co-operative meetings. These divisions generated 12 profiles with the peculiarity that each individual within the immigrant group has a counterpart within the native group. The results for the estimated probabilities of each individual profile being a violator are presented in Table 4. Within the group of recent immigrants, a probability of being a violator was estimated for individual ‘‘M’’ to be 0.997. The profile for this individual considers someone who has not seen National Park or Navy vigilance; does not perceive the penalties applied by the Park to be high; does not agree with the regulations; does not consider the leaders of the co-operatives to be representative; and does not have high attendance at organization meetings. In contrast, for M’s counterpart within the group of natives, individual ‘‘N’’, we estimated a probability of non-compliance of 0.78. It is interesting to note that if only the variables from Blocks I and II are modified—the individual has seen surveillance and considers the penalties to be high—for the immigrant (AM), we estimated a violation probability of 0.90, whereas the native counterpart (AN) experiences a higher decrease in the violation probability to 0.25. Furthermore, we also explored the effect of modifying variables included in Block VI, related to the legitimacy of the regulations (REGULA_GRL and REGULA_SEACUCUMBER). In this case, the violation probability of individual ‘‘BM’’, from the immigration group, falls to 0.87 and the violation probability of the counterpart, in the native group (BN), decreases to 0.19. Thus, blocks that represent perception of regulations’ legitimacy have a larger effect on the estimated reduction in the violation probability than those representing traditional enforcement tools (monitoring and penalty), suggesting that individuals who show a degree of agreement with the regulations are more likely to respect them. 19 The immigration index (YEARSMIGR) in this case is 0.133. The number of years residency considered (7.5) is consistent with the expansion of migration registered in the 1990s. 20 The immigration index (YEARSMIGR) for natives of the islands is 0.

270

Table 4 Probability of committing an infraction by type of boat-ownera Block Variable

II

FINE

V VI

ISLAND1 REGULA_GRL

REGULA_SEACUCUMBER

LEAD_REPRESENT

VII

YEARSMIGR

VIII

ATTEND Probability of infringementc

Boat-owner has observed surveillance of the GNP or the National Navy almost always or sometimes Boat-owner considers the fines to be high Boat-owner resides on Island 1 Boat-owner strongly agrees, agrees or is indifferent with the regulations of small-scale fishery in general Boat-owner strongly agrees, agrees or is indifferent with the regulations that are applied to the capture of sea-cucumber Boat-owner feels that the leadership of the fishery sector represents its interests or at least moderately represents them Number of years that the boatowner has resided on the islands in the case of not being a native The boat-owner attends all of the co-operative meetings

Natives

M

AM

BM

CM

DM

EM

N

AN

BN

CN

DN

EN

No

Yes

No

No

No

NO

No

Yes

No

No

No

No

No

Yes

No

No

No

No

No

Yes

No

No

No

No

No No

No No

No Yes

No No

No Yes

Yes Yes

No No

No No

No Yes

No No

No Yes

Yes Yes

No

No

Yes

No

Yes

Yes

No

No

Yes

No

Yes

Yes

No

No

No

Yes

Yes

Yes

No

No

No

Yes

Yes

Yes

7.5

7.5

7.5

7.5

7.5

7.5

Native Native Native Native Native Native

No

No

No

Yes

Yes

Yes

No

No

No

Yes

Yes

Yes

0.997

0.908

0.868

0.736

0.153

0.526

0.779

0.249

0.187

0.085

0.001

0.026

Source: Elaboration by the authors with the results obtained from the econometric estimations. a Model 2 is defined as: VIOLATION ¼ VIOLATION(CONSTANT,SEEN,FINE,PUNISHED,STORAGE,EDUCA,ISLAND1,REGULA_GRL,REGULA_ SEACUCUMBER,OTROS,LEAD_REPRESENT,YEARSMIGR,ATTEND). b To calculate the probability of infraction, the specified variables were considered relevant at 5% and 10%. The most prominent results are presented here. c The probability of infraction is calculated with the average of the variables (continuous and discrete), except for the variables presented in this table.

ARTICLE IN PRESS

SEEN

Recent immigrants

C. Viteri, C. Cha´vez / Ocean & Coastal Management 50 (2007) 253–274

I

Characteristics of the boatownerb

ARTICLE IN PRESS C. Viteri, C. Cha´vez / Ocean & Coastal Management 50 (2007) 253–274

271

We then also considered modifying the variables of Block VI, related to attendance at organization meetings and the level of representation of the fishery leadership, and Block VIII, related to attendance at organization meetings. Individual (CM) exhibits a violation probability of 0.74, lower than that of the previous individuals. Similarly, for the counterpart (CN), in the native group, we estimate a violation probability of 0.09, far lower than the previous native individual. This result indicates that fishermen who feel that they are represented by the organizations leaders and also get involved in co-operative meetings tend to have a lower probability of infringement than those who do not feel represented by the leadership and who do not attend the co-operative meetings. Furthermore, the decrease in the probability due to these variables (LEAD_REPRESENT and ATTEND) is higher in magnitude than the decrease in probability due to the variables of Blocks I, II, and those related to the legitimacy of the regulations (see Table 4). Finally, the probability of infraction for each of the profiles for native boat-owners is inferior to the probability of its counterpart within the group of immigrant profiles. This result supports the relevance of the sense of belonging to a community as a determinant factor of the individual violation decision. Furthermore, this is an indication that the problems of fishery regulation transgressions for the recent immigrant group need to be dealt with specifically for this group. 5. Conclusions The results obtained from the econometric analysis indicate that decisions to violate management regulations in the case of boat-owners of the Gala´pagos Marine Reserve are not only determined by deterrence variables, but also by variables related to the perception of boat-owners with respect to the legitimacy of regulations, the level of membership within the community, and fishermen’s participation in their organizations. It is important to emphasize the results we obtained related to variables that represent social aspects, which we found to influence the violation decision, i.e. the level of legitimacy of the norms and the sense of belonging to the community; as well as the moral commitment the individual has to the compliance and success of the regulations, reflected in the level of participation in the institutions that generate the regulations. Specifically, our results indicate that the level of legitimacy that the regulations have within the group of boat-owners, the individual’s sense of belonging, the legitimacy of local organizations, and the levels of participation of fishermen in their cooperatives positively influence decisions to violate the agreed regulations within the reserve. Thus, our results contribute empirical evidence supporting recent theoretical developments to explain regulation compliance/violation for the management of natural resources in fisheries. Additionally, the results are limited in terms of determining a profile based on particular characteristics. Only two characteristics were identified to have a significant positive effect on the violation decision: residency on Island 1 and level of formal education. The result in case of education is perhaps surprising. A more in-depth analysis of the individuals with higher levels of education within the context of participatory management indicates that the majority of these individuals (56%) disagree with the regulations and most of them (81%) consider the representation of the leadership to be inadequate.21 This suggests that 21

Those individuals with 12 or more years of schooling were considered to belong to the higher educated group; they represent 37% of the sample.

ARTICLE IN PRESS 272

C. Viteri, C. Cha´vez / Ocean & Coastal Management 50 (2007) 253–274

there is a relationship between education and adopting a critical position towards the regulations and the leadership and, as we observed in the previous results, these aspects are important in the violation decision. The results also have implications for the proper design of enforcement to achieve adequate levels of compliance with fisheries regulations in the reserve. Monitoring and penalties, traditional deterrence instruments, are limited in their ability to improve the detection capacity and increase the financial cost of non-compliance, appear to have inferior results in terms of deterring violations when compared to those instruments orientated towards promoting the legitimacy of the norms, improving the representation of individuals by the leadership of local organizations, and increasing the participation of individuals within grass roots organizations.22 The results of the role of social variables as determinants of violations are useful for the Gala´pagos National Park authority as well as for conservation organizations that work on the islands. They suggest that improving compliance might require reallocating efforts and resources in their annual programming. How to make users feel participatory in their institutions is perhaps beyond the purposes of this work; however, there is a wide range of potential areas where the reallocation might have positive impacts on compliance results. These range from actions that provide more information about regulations and their scientific basis, which can directly contribute to improving the legitimacy of the regulations, to actions such as supporting the operation of the participatory management system of the reserve, improving co-operative leader representation, and strengthening cooperative organizations. According to our results, these actions should have more paybacks in terms of respecting the regulations than traditional enforcement strategies such as increased vigilance or penalties. Our suggestions to reallocate efforts and resources to improve compliance do not mean abandoning traditional strategies of enforcement, control, monitoring, and the imposition of penalties, since all groups are likely to include the presence of chronic violators, who can only be controlled by traditional methods such as tougher sanctions and more surveillance. The control of these individuals is important because their illegal behavior could erode the level of legitimacy of norms and increase the probability of infractions by their fellow fishermen.23 Our work can be extended in a number of ways. We would like to mention two of them. First, it would be desirable to develop an investigation that further explores the differences that exist between violation decisions of native individuals with deeper island roots and recent immigrants. Immigration might be associated with more rule-breaking behavior for a number of reasons. For example, the immigrants might have different views on what rules should be used, different attitudes toward social pressures, or perhaps different intertemporal preferences. In addition, because we use a cross-section data set on compliance behavior, the existence of a potential assimilation process of immigrants to the 22

This result is observed independently if the boat-owner is a recent immigrant or native, and also independently if the individual resides on Island 1. 23 It is important to underline that the surveillance effort of the National Park is not devoted only to the control of the activities of the artisanal fisherman operating in the reserve, but also to the potential illegal incursions of continental and foreign vessels. Weakening the control or sanctions of illegal behavior for potential chronic violators or vessels outside of the community might send two messages to the rest of the boat-owners who normally comply; namely, that the regulations are unfair, and that the regulations neither adequately protect the fishery resources nor the local fishermen [18].

ARTICLE IN PRESS C. Viteri, C. Cha´vez / Ocean & Coastal Management 50 (2007) 253–274

273

native population behavior is difficult to uncover. Second, we also feel that exploring alternative management regulations in the reserve is worth pursuing. It is possible that both the type of management tools used in the reserve fisheries as well as the high rate of self-reported non-compliance by fishermen have contributed to diluting the sense of secure ownership of resources in the reserve. Exploring the possibilities for the use of property rights-based systems for the management of Gala´pagos’ fisheries is another possible road of future research. Finally, the empirical results of this study provide evidence indicating a positive contribution of the Participatory Management System implemented in the Gala´pagos Marine Reserve on compliance with management regulations. Furthermore, the results also underline the potential of the system to increase compliance with the regulations between the boat-owners of the Gala´pagos’ artisanal fleet. This is a desirable objective considering that the conservation and sustainable management of this important ecosystem depend upon them. Exploring this and other issues might contribute to a better design of conservation programs for the protection of the archipelago. Acknowledgements Our thanks go out to the Gala´pagos National Park for the information provided, especially to the Unit of Marine Resources and the Law Department. Also, we thank the Gala´pagos Fishermen Cooperatives, the Araucaria Project of the Spanish Cooperation Agency, and the Participatory Management Board for their support during fieldwork. Additionally, we are grateful to the Marine Conservation and Research Department of the Charles Darwin Research Station for the information provided. The fieldwork was possible thanks to the financial of the Catherine and John MacArthur Foundation, in the framework of the program for graduate degree theses in protection of biodiversity and conservation of ecosystems (Master Program on Natural Resource and Environmental Economics, University of Concepcio´n). The authors gratefully acknowledge detailed and insightful comments and suggestions on previous versions of this article from the Editor and an anonymous referee. Our colleagues Jorge Dresdner and Hugo Salgado also provided helpful comments and suggestions on early stages of this research. References [1] Toral MV, Murillo JC, Piu M, Nicolaides F, Moreno J, Reyes H, et al. La pesquerı´ a de pepino de mar (Isostichopus fuscus) en la Reserva Marina de Gala´pagos en el an˜o 2005. Fundacio´n Charles Darwin/Parque Nacional Gala´pagos, Santa Cruz, Gala´pagos, Ecuador, 2005. 41pp. [2] Hearn A. Evaluacio´n poblacional de la langosta roja en Gala´pagos, 2005. Fundacio´n Charles Darwin, Santa Cruz, Gala´pagos, Ecuador, 2005. 17pp. [3] Viteri C, Cha´vez C. Fiscalizacio´n y Cumplimiento de las Regulaciones Pesqueras en la Reserva Marina Gala´pagos (RMG): 1998–2001. Cuestiones Econo´micas 2003;19(33):135–77. [4] Sutinen J, Andersen P. The economics of fisheries law enforcement. Land Economics 1985;61(4):387–97. [5] Anderson L, Lee D. Optimal governing instrument, operation level, and enforcement in natural resource regulation: the case of the fishery. American Journal of Agricultural Economics 1986;68(4):679–90. [6] Milliman S. Optimal fishery management in the presence of illegal activity. Journal of Environmental Economics and Management 1986;13(4):363–81. [7] Sutinen J, Kuperan K. A socioeconomic theory of regulatory compliance in fisheries. International Journal of Social Economics 1999;26(1/2/3):174–93.

ARTICLE IN PRESS 274

C. Viteri, C. Cha´vez / Ocean & Coastal Management 50 (2007) 253–274

[8] Charles A, Mazani L, Cross M. The economics of illegal fishing: a behavioral model. Marine Resource Economics 1999;14:95–110. [9] Sutinen J, Gauvin J. An econometric study of regulatory enforcement and compliance in the commercial inshore lobster fishery of Massachusetts. In: Neher PA, et al., editors. Rights based fishing. Dordrecht: Kluwer Academic Publishers; 1989. p. 415–28. [10] Furlong W. The deterrent effect of regulatory enforcement in the fishery. Land Economics 1991;67(1):116–29. [11] Hatcher A, Jaffry S, The´baud O, Bennett E. Normative and social influences affecting compliance with fishery regulations. Land Economics 2000;76(3):448–61. [12] Gordon D, Hatcher A. Further investigations into the factors affecting compliance with UK fishing quotas. Land Economics 2005;81(1):71–86. [13] Nielsen JR, Mathiesen C. Important factors influencing rule compliance in fisheries: lessons from Denmark. Marine Policy 2003;27(5):409–16. [14] Eggert H, Ellegard A. Fishery control and regulation compliance: a case for co-management in Swedish Commercial Fisheries. Marine Policy 2003;27(6):525–33. [15] Randall J. Improving compliance in US Federal Fisheries: an enforcement agency perspective. Ocean Development & International Law 2004;35:287–317. [16] Eggert H, Lokina R. Regulatory compliance in Lake Victoria Fisheries. Unpublished Working Paper, Department of Economics, Go¨teborg University, Go¨teborg, Sweden, 2005. [17] Beckenkamp M, Gu¨mbel W. Emotional appraisal of different situations in commons dilemmas depending on social value orientation. In: Proceedings of the IASCP 2000 Conference, University of Saarland— Psychology. [18] Kuperan K, Sutinen J. Blue water crime deterrence, legitimacy and compliance in fisheries. Law and Society Review 1998;32(2):309–37. [19] Nielsen J. An analytical framework for studying compliance and legitimacy in fisheries management. Marine Policy 2003;27(5):425–32. [20] Heylings P, Bravo M. El Sistema de Manejo Participativo de la Reserva Marina de Gala´pagos-Principales Actividades en el an˜o 2001. In: Informe Gala´pagos 2001–2002, Fundacio´n Natura/WWF Quito, Ecuador, 2002. p. 71–6. [21] Heylings P, Cruz F, Bustamante R, Cruz D, Escarabay M, Granja A, et al. El Grupo Nu´cleo de la Reserva Marina de Gala´pagos. In: Informe Gala´pagos 1997–98, Fundacio´n Natura/WWF, Quito, Ecuador, 1998. p. 14–6. [22] Murillo JC. ‘‘Indicadores Socio-econo´micos de la pesca’’, En: ‘‘Reserva Marina de Gala´pagos, Lı´ nea Base de la Biodiversidad’’. In: Danulat E, Edgar GJ, editors. ECCD-SPNG, Santa Cruz, Gala´pagos, Ecuador, 2002. p. 440–8. [23] Becker G. Crime and punishment: an economic approach. Journal of Political Economy 1968;76(2):169–217. [24] Bardhan P. Analytics of the institutions of informal cooperation in rural development. World Development 1993;21(4):633–9. [25] Ca´rdenas JC, Strandlund J, Willis C. Local environmental control and institutional crowding-out. World Development 2000;28(10):1719–33. [26] Scheaffer R, Mendenhall W, Ott L. Elementos de Muestreo. Tercera Edicio´n. Estados Unidos de Ame´rica: PWS Publishers; 1986.