UC Santa Barbara
UC Santa Barbara Previously Published Works
Title
Cooperation as a solution to shared resources in territorial use rights in fisheries.
Permalink
https://escholarship.org/uc/item/5cb2t1ft
Journal
Ecological applications : a publication of the Ecological Society of America, 30(1)
ISSN
1051-0761
Authors
Aceves-Bueno, Eréndira
Miller, Steve J
Cornejo-Donoso, Jorge
et al.
Publication Date
2020
DOI
10.1002/eap.2022
Peer reviewed
eScholarship.org Powered by the California Digital Library
University of California
Cooperation as a solution to shared resources in territorial use rights
in fisheries
ER
ENDIRA ACEVES-BUENO,
1,5
STEVE J. MILLER,
2
JORGE CORNEJO-DONOSO,
3
AND STEVEN D. GAINES
4
1
Nicholas School of the Environment, Duke University, 135 Duke Marine Lab Road, Beaufort, North Carolina 28516 USA
2
Environmental Studies Program, University of Colorado, Boulder, 4001 Discovery Drive Boulder, Colorado 80303 USA
3
Instituto de Fomento Pesquero (IFOP), Col
on 3656 Talcahuano, Regi
on del Biobio, Chile
4
Bren School of Environmental Science and Management, University of California Santa Barbara, Santa Barbara, California
93106-51312 USA
Citation: Aceves-Bueno, E., S. J. Miller, J. Cornejo-Donoso, and S. D. Gaines. 2020. Cooperation as a
solution to shared resources in territorial use rights in fisheries. Ecological Applications 30(1):e02022.
10.1002/eap.2022
Abstract. Territorial use rights in fisheries (TURFs) are coastal territories assigned to fish-
ermen for the exclusive extraction of marine resources. Recent evidence shows that the incen-
tives that arise from these systems can improve fisheries sustainability. Although research on
TURFs has increased in recent years, important questions regarding the social and ecological
dynamics underlying their success remain largely unanswered. In particular, in order to create
new successful TURFs, it is critical to comprehend how fish movement over different distances
affects the development of sustainable fishing practices within a TURF. In theory, excessive
spillover outside a TURF will generate incentives to overharvest. However, many TURFs have
proven successful even when targeted species move over distances far greater than the TURF’s
size. A common attribute among some of these successful systems is the presence of inter-
TURF cooperation arrangements. This raises the question of how different levels and types of
cooperation affect the motivations for overharvesting driven by the movement of fish outside
the TURF. In this paper, we examine equilibrium yields under different levels of inter-TURF
cooperation (from partial to full) and varying degrees of asymmetry across TURFs of both
biological capacity and benefit-sharing. We find that partial cooperation can improve yields
even with an unequal distribution of shared benefits and asymmetric carrying capacity. How-
ever, cooperation arrangements are unstable if the sharing agreement and biological asymme-
tries are misaligned. Remarkably, we find that asymmetry in the system can lead to the
creation of voluntary no-take zones.
Key words: bio-economic model; catch shares; game theory; small-scale fisheries; social-ecological
systems; spatial management; TURF.
INTRODUCTION
Human population growth is particularly fast in
coastal areas (Neumann et al. 2015). In the absence of
effective management schemes, the increased depen-
dency on fish as a global food source and income for
local communities is causing coastal ecosystems to dete-
riorate (Vitousek et al. 1997, Pauly et al. 1998, 2005,
Pauly and Zeller 2016). Ultimately, this situation not
only affects the coastal environment but also the liveli-
hoods of fishing communities (Mcclanahan et al. 2015).
The development of innovative management tools can
offer a solution to these problems. In recent years, catch
shares have been widely studied as a solution to prob-
lems of overexploitation and inefficiency in fisheries
(Grafton 1996, Grafton et al. 2006, Costello et al. 2008,
Griffith 2008, Birkenbach et al. 2017). A particular form
of catch shares, called territorial use rights in fisheries
(TURFs), has emerged as a very promising tool for the
management of coastal fisheries. TURFs provide spa-
tially explicit access rights to resource users, and empiri-
cal studies have shown they can provide the right set of
conditions to achieve more sustainable and profitable
artisanal coastal fisheries (Smith and Panayotou 1984,
Uchida and Baba 2008, Costello and Kaffine 2010,
Gelcich et al. 2010).
Although TURFs are increasingly used, there is still
considerable uncertainty around the design characteris-
tics that promote the best fishery outcomes. Many
TURF systems around the world are relatively small in
order to restrict the number of users. Although the role
of group size in collective action is dependent on the
local institutional arrangements (Ostrom 2009), small
groups tend to be formed by members with homogenous
social characteristics, which facilitates coordination,
monitoring, and enforcement (Olson 1965, Agrawal and
Goyal 2001, Agrawal 2002, Poteete and Ostrom 2004,
Manuscript received 9 April 2019; revised 3 August 2019;
accepted 16 September 2019. Corresponding Editor:
Eva
Elizabeth Plaganyi.
5
E-mail: ere.aceves@duke.edu
Article e02022; page 1
Ecological Applications, 30(1), 2020, e02022
© 2019 by the Ecological Society of America
Cancino 2007, Ostrom 2010). However, this poses a
challenge since, in order to be successful, TURFs need
to provide clear access rights. Since ocean ecosystems
are highly connected systems where fish can escape the
TURFs (a process known as spillover) and be captured
in neighboring fishing grounds, the exclusivity can be
compromised when TURFs are small. Hypothetically,
fish spillover can have a great impact on fisher behavior,
since expected losses from the TURF can induce a race
to fish and drive overfishing (White and Costello 2011,
Aceves-Bueno and Halpern 2018). How is the uncon-
trolled spillover problem solved?
One potential explanation for the unexpected success
of small TURFs with high spillover levels is the develop-
ment of partial cooperation schemes among TURFs. If
two TURFs fully coordinate their harvests and share
benefits, they can effectively act as a single larger TURF
and reduce the incentives to race (Kaffine and Costello
2011). Often, however, cooperation is less than complete.
Examples of inter-TURF partial cooperation can be
found across a range of TURFs in Japan, Mexico, and
Chile, where partial cooperation has emerged as either
profit sharing (unitization), joint monitoring and
enforcement, or shared marketing of catch. These
arrangements entail varying degrees of cooperation,
with many involving only partial sharing of the benefits
of fishing (Cancino et al. 2007, Uchida and Baba 2008,
Kaffine and Costello 2011).
Many of these arrangements are also characterized by
biological and benefit-sharing asymmetry. The conse-
quences of benefit-sharing asymmetries have been stud-
ied in a variety of cooperative settings (e.g., Osborne and
Pitchik 1983, Darrough and Stoughton 1989, Lundberg
and Pollak 1996). In TURFs, asymmetries can be
expected due to differences in membership, access to
markets, access to information, and government sup-
port. For example, in the Sakura-Ebi fishery in Japan, a
pooling arrangement was set between the Yui Harbor
and Ohigawamachi cooperatives. In this pooling
arrangement, the income is spread equally among all 60
fishing units of both cooperatives, 42 of which belong to
the Yui Harbor cooperative. In spite of this asymmetry
in shared benefits distribution, this pooling system has
been in place since 1967 and has allowed the Sakura-Ebi
fishery to thrive as one of the most profitable in Japan
(Uchida and Baba 2008).
Biological asymmetry is also common in coastal sys-
tems and leads to differential production capacity
among TURFs. Such is the case for the Vigia Chico
Cooperative in Punta Allen, Mexico. Fishermen in this
cooperative have developed a unique fishing system
involving casitas, concrete shades where lobsters aggre-
gate in the search for shelter (Cunningham et al. 2013,
M
endez-Medina et al. 2015). The carrying capacity of
each TURF depends not only on the presence of natural
suitable habitat, but also on the capacity of each TURF
owner to invest in casitas for artificial habitat. Variation
in investment has resulted in a system of TURFs with
different carrying capacities. Despite this asymmetry,
fishermen in Punta Allen not only maintain sustainable
levels of harvest, but have also developed conservation
measures, such as the protection of reproductive females
(Defeo and Castilla 2006). The reason for this behavior
is that all individual TURF owners are part of the Vigia
Chico cooperative. That cooperative system involves all
of its members in decision making, monitoring, and
enforcement. Furthermore, although TURF owners
have full autonomy regarding the number of casitas and
catch sizes inside their fishing areas (Cunningham et al.
2013; C. Mendez-Medina, personal communication), they
partially cooperate by paying a fee to maintain the coop-
erative. In return, the cooperative provides infrastructure
to process the catch, offers access to market and market-
ing tools, and absorbs all the costs of monitoring and
enforcement (Cunningham et al. 2013).
We seek to better understand when partial coopera-
tion such as that in Punta Allen can arise among TURF
owners despite asymmetries in both biology and bar-
gaining power. The interactions between spillover of fish
and the level of inter-TURF partial cooperation have
previously been analytically explored only along the
margins, varying spillover at extreme levels of partial
cooperation (full or absent) or changing levels of partial
cooperation for a fixed level of spillover (Kaffine and
Costello 2011). This approach does not allow the analy-
sis of a wide range of stable partial cooperation scenar-
ios. As a consequence, it limits our ability to fully
explore the interconnected dynamics between movement
and partial cooperation to develop appropriate guideli-
nes for more effective TURF designs with different tar-
get species.
Here we rectify these shortcomings by numerically
examining the effects of different partial cooperation
levels. By examining coalitions with partial harvest pool-
ing in symmetric and asymmetric scenarios, we are able
to discern the magnitude of gains or losses compared to
full cooperation. Our approach allows exploring the full
interaction between the extents of fish movement and
inter-TURF cooperation, to identify design guidelines
that will enhance expected yields. We ask, when is coop-
eration among TURFs the best solution for the spillover
problem? How do the gains from cooperation scale with
the level of partial cooperation and fish distribution
asymmetry? We find that partial cooperation will be
beneficial for both TURF owners across a wide range of
shared benefits distribution arrangements. However,
with strong fish spillover, high levels of partial coopera-
tion are necessary to achieve yields close to maximum
sustainable yield (MSY). Interestingly, we find that
asymmetrical systems can lead to the creation of volun-
tary fishery closures.
M
ETHODS
Based on White and Costello (2011), we use a bio-eco-
nomic model that consists of two TURFs, each owned
Article e02022; page 2 ER
ENDIRA ACEVES-BUENO ET AL.
Ecological Applications
Vol. 30, No. 1
by a group acting as a single agent. In the model, the
adult stock density in patch i at time t (N
t
i
) evolves as
follows:
N
tþ1
i
¼ðN
t
i
þ M
t
i
Þð1 dÞþ
S
t
i
1 þ a
i
S
t
i
:
here M
t
i
is net migration into patch i, d characterizes the
natural annual mortality of adults (calculated as the
inverse life span in years), S
t
i
describes the density of set-
tlers, and a
i
is a parameter describing the strength of
density dependence. In particular,
a
i
¼
CR 1
CRdK
i
;
where CR is the Goodyear compensation ratio. CR
was fixed to 4, a value commonly used to represent
coastal fish species (White and Costello 2011). A sen-
sitivity analysis of the CR is presented in the supple-
mentary information (Appendix S1: Fig. S1, S2). K
i
is
the carrying capacity in patch i. The carrying capacity
is allowed to differ across patches so that we can
study the influence of asymmetry (either inherent or
influenced by investments, e.g., casitas in Punta Allen)
on the viability of cooperation.
The production of settlers is determined by the follow-
ing production function:
S
t
i
¼ P
i
ðN
t
i
þ M
t
i
Þ;
where P
i
is the per-capita larval production by adults in
patch i:
P
i
¼
d
1 da
i
K
i
:
Adult net migration is a function of a migration parame-
ter m and the difference in relative density between the
two patches:
M
t
i
¼M
t
j
¼ mð N
t
i
þ N
t
j
Þ
N
t
j
K
j
N
t
i
K
i
:
Note,
N
tþ1
i
¼ N
t
i
þ M
t
i
¼ð1 mÞN
t
i
þ mðN
t
i
þ N
t
j
Þ
N
t
i
N
t
i
þ N
t
j
þ
N
t
j
K
j
N
t
i
K
i
"#
so that m can be thought of as the fraction of fish from
both patches that swim around, with the fraction of
swimmers that end up in i determined by the term in
square brackets. Harvest occurs after adult movement
and larval settlement. The harvest function in steady
state conditions is represented by
Y
i
¼ M
i
þ
S
i
1 þ aS
i
dðN
i
þ M
i
Þ:
Each patch is managed by a single agent that selects
the escapement level (N
l
) to maximize a yield-based
objective. The agents choose harvest independently and
simultaneously, taking the other owner’s decision as
given; we therefore examine what choices the patch own-
ers make in a Nash equilibrium.
To study inter-TURF partial cooperation in the
model, we allow the objective that each owner maxi-
mizes to depend partly on yields in the other patch. Par-
tial cooperation allows a more general approach that
includes the fully cooperative and noncooperative sce-
narios, but it is able to also incorporate intermediate
cooperation scenarios. For partial cooperation, the ben-
efits of one patch depend on a portion of the harvest of
the second patch (Kopel and Szidarovszky 2006). Patch
owners still choose Y independently and simultaneously
to maximize their own benefits, but their benefits are
now determined by the sharing arrangement. In order to
cooperate, both patches provide a portion of their catch
to a common pool C:
C ¼ hðY
i
þ Y
j
Þ
where h 2 [0, 1] is a parameter determining the strength
of partial cooperation. Setting h to zero represents no
cooperation, and a value of one represents full coopera-
tion. The distribution of shared benefits (dividend) for
each patch from the pooled catch depends on the param-
eter b and is defined as
D
i
¼ bC:
D
j
¼ð1 bÞC:
With this setup, the benefits under partial cooperation
for the patch owners are:
Y
C
i
¼ð1 hÞY
i
þ bC ¼ð1 ð1 bÞhÞY
i
þ bhY
j
;
Y
C
j
¼ð1 hÞY
j
þð1 bÞC ¼ð1 bhÞY
j
þð1 bÞhY
i
:
In other words, each owner maximizes a weighted
average of the yields in the two patches, with weights
determined by b and h.
To consider whether a particular cooperative arrange-
ment (choice of b and h) is stable, we compare each
patch owner’s benefits to those she would receive if
h ¼ 0: the fully noncooperative outcome. If both
patches earn higher benefits when cooperating than not,
we consider partial cooperation to be stable (D’Aspre-
mont et al. 1983). For much of our analysis, we treat b
and h as fixed parameters, as if TURF owners were
handed an agreement they could choose to sign or not.
This helps retain focus on the potential for partial coop-
eration. In Appendix S1, we briefly consider how TURF
owners might arrive at a specific choice of b and h
through negotiation.
After studying this model in a general setting, we use
it to analyze the spiny lobster (Panulirus argus) coopera-
tive in Punta Allen (Fig. 1). The parameters used for this
case study are presented in Table 1. The fishing grounds
in that cooperative vary in size, with an average along-
shore length of 1.4 km. The differences in area and in
the presence of natural and artificial suitable area for
lobster create asymmetry in the distribution of resources
January 2020 TURFS AND COOPERATION Article e02022; page 3
among owners. In addition, the owners engage in partial
cooperation, fishing separately but putting a portion of
proceeds toward comarketing and other joint activities.
Thus, this case study provides a rich opportunity for the
application of our spatially asymmetric model of partial
cooperation.
We calculate the movement of lobsters between
patches (m) based on the spiny lobster
0
s home range rel-
ative to the average TURF size, according to the model
of Kramer and Chapman (1999). For this case study, we
use the average TURF length (1.4 km) and set the home
range to 4 km, which is the largest movement performed
by a lobster before their movement is considered noma-
dic (Bertelsen 2013). For consistency with previous
application of similar models (White and Costello 2011,
Aceves-Bueno et al. 2017), we use the inverse life span in
years (Maxwell et al. 2007) as a measure of the species
natural mortality.
R
ESULTS
Our models show that partial cooperation can lessen
the negative effects of fish spillover across TURFs on
yields over wide ranges of fish mobility and degrees of
cooperation.
Fig. 2 shows the overall yields of both patches result-
ing from all possible combinations of cooperation levels
(h) and different shared benefits distribution arrange-
ments represented by the fraction of benefits assigned to
patch i (b). Intuitively, overall yields rise to maximum
sustainable yield at full cooperation (Fig. 2).
Fig. 3 illustrates the benefits under partial coopera-
tion to patches i and j under a range of scenarios, vary-
ing (1) the degree of fish movement between patches
(lines per plot), (2) distribution of shared benefits (left
to right), and (3) asymmetry in patch productivity (top
vs. bottom). All benefits are shown relative to those
received if the patch owners do not cooperate (h ¼ 0),
shown by the gray horizontal line at 1, allowing identi-
fication of scenarios that lead to better (or worse) out-
comes than full noncooperation. In general, the net
benefits for both patches are higher for more mobile
targeted species. In perfectly symmetric scenarios
(Fig. 3E), both patches are equally benefited by cooper-
ation. As both patches receive more under partial coop-
eration than noncooperation (benefits lie above the
noncooperative line at 1), TURF owners could plausi-
bly agree on partial cooperation across a range of h
choices.
The prospects for partial cooperation in asymmetric
systems (Fig. 3A–F) are more complex. Some patterns
are intuitive: a patch owner benefits more from cooper-
ation when she receives a greater share of pooled yields
(left vs. center or right panels) and when the other
patch is more biologically productive (top vs. bottom
panels). Similarly, net benefits from cooperation remain
positive under many combinations of asymmetry in
biology and benefit sharing (Fig. 3A, B, D). However,
if a TURF owner receives a small share of pooled bene-
fits relative to the carrying capacity of her patch
(Fig. 3C, F), partial cooperation may no longer be ben-
eficial to her. In those cases, increases in the degree of
profit-sharing h also may not lead to greater net bene-
fits from cooperation. As a result, partial cooperation
could not be stable if asymmetries in the biology and
the sharing rules are sufficiently misaligned. In
Appendix S1, we also examine a different role of asym-
metry, studying how aversion to inequity may affect the
behavior of TURF owners.
FIG. 1. Individual territorial use rights in fisheries (TURFs)
of the Vigia Chico Cooperative in Punta Allen, Mexico.
Adapted from a map provided by Comunidad y Biodiversidad
A.C (COBI).
TABLE 1. Parameters used for the analysis of the Vigia Chico
Cooperative territorial use rights in fisheries (TURFs).
Parameter Value Source
Mean along shore
length of the Vigia
Chico Cooperative
TURFs
1.43 km map provided by
Comunidad y
Biodiversidad (COBI)
Spiny lobster (Panulirus
argus) adult home range
4 km Bertelsen (2013)
Spiny lobster
(Panulirus argus)
life span
20 yr Maxwell et al. (2007)
Article e02022; page 4 ER
ENDIRA ACEVES-BUENO ET AL.
Ecological Applications
Vol. 30, No. 1