Content uploaded by Sebastian Villasante
Author content
All content in this area was uploaded by Sebastian Villasante on Dec 30, 2022
Content may be subject to copyright.
Vol.: (0123456789)
1 3
Rev Fish Biol Fisheries
https://doi.org/10.1007/s11160-022-09750-7
ORIGINAL RESEARCH
Adaptive factors andstrategies insmall‑scale fisheries
economies
LudmilaM.A.Damasio ·
MariaGraziaPennino· SebastiánVillasante·
AdrianaRosaCarvalho· PriscilaF.M.Lopes
Received: 14 February 2022 / Accepted: 21 December 2022
© The Author(s), under exclusive licence to Springer Nature Switzerland AG 2022
Abstract Despite its relevance, the economic con-
tribution of small-scale fisheries to poverty allevia-
tion is still poorly understood. This study investigates
why some fishers perform economically better in
fisheries than others under similar conditions and
whether these variations in performance were due
to individual adaptive strategies related to fish-
ing technology and effort. A pairwise comparison
between fishers’ income from the Brazilian equato-
rial region in 1994 and 2014 was performed while
modeling individual changes related to the fishing
activity (Generalized Linear Model, GLM) and the
factors that would explain why fishers became richer
or poorer over time (Proportional odds model). Fish-
er’s geographical region, the use of motorized boats
and the adoption of hookah compressors explained
income in 1994, whereas having larger boats and
fishing with hook and line explained it in 2014. Fish-
ers were slightly more likely to gain income if they
changed their type of boat. Some fishers are trapped
in poverty, and the changes they made were either
not enough to leave this condition or made it worse.
Escaping poverty traps in fisheries may require efforts
beyond those available to the individuals, especially
as stocks become increasingly overfished.
Supplementary Information The online version
contains supplementary material available at https:// doi.
org/ 10. 1007/ s11160- 022- 09750-7.
L.M.A.Damasio(*)
Graduate Program inEcology, Federal University ofRio
Grande doNorte, Campus Universitário Lagoa Nova,
Natal, RN59078-900, Brazil
e-mail: ludmila.damasio@hotmail.com
L.M.A.Damasio· M.G.Pennino· A.R.Carvalho·
P.F.M.Lopes
Fishing Ecology, Management andEconomics Group
(FEME), Department ofEcology, Universidade Federal
doRio Grande doNorte – UFRN, Natal, RN, Brazil
e-mail: grazia.pennino@ieo.csic.es
P. F. M. Lopes
e-mail: priscila.lopes@ufrn.br
L.M.A.Damasio
BW Institute, Rua Sueli Brasil Flores Number 88,
Araruama,RiodeJaneiro78970-000, Brazil
M.G.Pennino
Instituto Español de Oceanografía (IEO, CSIC), Centro
Oceanográfico de Vigo, Subida a Radio Faro, 50-52,
36390Vigo,Pontevedra, Spain
M.G.Pennino
Statistical Modeling Ecology Group (SMEG), Valencia,
Spain
S.Villasante
Department ofApplied Economics, Faculty ofBusiness
Administration andManagement, University ofSantiago
de Compostela, 15782SantiagodeCompostela, ACoruña,
Spain
e-mail: sebastian.villasante@usc.es
S.Villasante
EqualSea Lab-CRETUS, SantiagodeCompostela,
ACoruña, Spain
Rev Fish Biol Fisheries
1 3
Vol:. (1234567890)
Graphical Abstract
Keywords Brazil· Overcapacity· Poverty-traps·
Socioeconomics· Small-scale fisheries
Introduction
Fish are one of the world’s single-most traded food
commodities (Asche and Smith 2010). In 2016,
60 million tons of fish and fisheries products were
exported throughout the world, valued at USD 143
billion. In addition to their global commercial impor-
tance, fisheries employ 40.3 million people world-
wide, while fish are one of the main proteins con-
sumed, especially among the poor.
Small-scale fisheries (SSF), specifically, are
responsible for generating 90% of all jobs in the
global fishing sector, and for harvesting almost half of
global fish harvests, generally using less technologi-
cally intensive methods (Smith and Basurto 2019).
The fact that this type of fishery occurs mainly in
developing countries, where income generating activ-
ities may be lower, further increases its vulnerability
to different drivers (e.g., overfishing, climate change
and habitat loss) (Béné etal. 2005; FAO 2017).
The importance of SSF does not necessarily ensure
their ecological sustainability or their proper fisheries
management (Steneck and Pauly 2019). Many coun-
tries, especially in the global south, lack basic fish-
ing information such as harvesting tracking data (e.g.,
quantities, species, ex-vessel prices, discards, and
fishing grounds used) and vessel data (e.g., fleet size,
capacity of fishing boats and operating gears), needed
to implement effective policies (FAO 2018). The lack
of SSF data spans from very basic information on fish-
ers, such as the number of artisanal fishers in a given
place, to more detailed economic information, such
as fisher incomes and how they vary over time (Sala
etal. 2018). Still, it is well known that SSF is often
associated with processes of poverty, marginalization,
vulnerability and exclusion (Garcia etal. 2018). The
explanations for this correlation are multifold.
Generally, small-scale fishing is on the fringes
between the formal and informal activity. Accessing
certain public policies, especially those related to labor
rights, tend to encounter barriers because legal regula-
tions and normative instructions often trigger bureau-
cratic difficulties in accessing such policies (Garcia
etal. 2018). Another explanation is the poor biological
conditions of the exploited natural resources, caused
by the open-access nature of many fisheries and, par-
ticularly, by the lack of proper management of SSF
(Béné 2003; Sala et al. 2018). In addition, individu-
als who support themselves with livelihoods based on
the extraction of natural resources have to deal with a
highly variable income caused both by the economy,
such as changes in demand or price, and by natural
fluctuations in resources (Sethi etal. 2012).
There is no doubt that understanding the factors
that influence fisher incomes, which have economic
consequences on households, communities, and
countries, is critical to both better managing fisher-
ies and informing effective policy aimed at eradicat-
ing hunger and poverty towards achieving UN SDGs.
Therefore, the objective of this study is to analyze the
adaptive strategies that fishers (can or cannot afford
to) adopt that could help explain why some fish-
ers get caught in poverty traps while others see their
economic condition improve over time. A poverty
trap represents a situation where people are unable
to mobilize the resources needed to overcome shocks
or chronic low-income situations and, consequently,
remain in poverty despite their best efforts (Cinner
etal. 2009; Barrett etal. 2011).
A large Brazilian coastal region, marked by high
levels of poverty and a high dependency on the
oceans, was chosen as the study area, in which fisher-
ies were used to understand changes in income lev-
els over time. Specifically, it was evaluated whether
fishers that changed their target species or switched to
species of different prices, invested in larger or more
powerful boats or changed their fishing effort made
them more or less poor (i.e., changed their fishing
Rev Fish Biol Fisheries
1 3
Vol.: (0123456789)
income, used here as a proxy of economic perfor-
mance). It was assumed that changing was a neces-
sary requirement to improve economic conditions to
deal with failing resources, but that not all changes
would have the same effect. For example, increasing
boat size could eventually worsen poverty as their
expenses (e.g., fuel) are higher (Damasio etal. 2016).
Methodology
Study area
The northeast is one of the poorest regions of Brazil,
with almost half of its population living in poverty
(NIS 2020). The eight coastal municipalities cho-
sen for data collection (Caiçara do Norte, Galinhos,
Guamaré, Macau, Porto do Mangue, Caucaia, São
Gonçalo do Amarante, and Paracuru) are located in
the states of Rio Grande do Norte and Ceará (Fig.1).
These places were specifically chosen because they
represent important landing ports for artisanal fisher-
ies (MMA 2008).
Overall, on the Brazilian northeastern coast, arti-
sanal marine extractive fishing predominates, whose
production represents 48% of the total fish in the
region (IBAMA 2007). Artisanal fishing is based on
the family unit and relies mainly on the use of small
vessels sometimes not owned by the fisher. When the
fisher does not own the boat, part of the production is
used to pay the boat owner’s rent. In these cases, the
working relationship between the boat owner and the
crew is informal. In general, the largest percentage of
the fish revenue is destined for the boat owner, who
bears the fishing expenses (engine oil and ice, for
example) and the maintenance costs.
The fishing sector in this Brazilian region has
endured without any significant or major technologi-
cal advance, perhaps with the exception of the use
of GPSs, which is now commonly used. A large part
of the fishing fleet is still formed by wooden vessels
that provide unsafe working conditions. Fishing infra-
structure, such as fish processing units and ice facto-
ries, is either in poor operating conditions or does not
exist (FAO 2008; MMA 2008).
Fig. 1 Study area highlighting the eight municipalities selected for this study
Rev Fish Biol Fisheries
1 3
Vol:. (1234567890)
Data
The database used in this study came from a partner-
ship between the authors’ research group and Petro-
bras (the Brazilian state-owned oil company). The
former developed a protocol for the latter to collect
data on fisheries in areas potentially under the impact
of oil platforms (i.e., more likely to be affected by a
spill or even by socioeconomic changes driven by the
presence of the platform, as the attraction of outside
workers). The data were obtained using semi-struc-
tured interviews applied between December 2014
and December 2015 and aimed to obtain past (1994,
assumed to be 20years ago) and present (at the time
of data collection) information. These years, specifi-
cally, are not marked by any environmental disaster,
economic crises or strong El Niño event, which could
have biased the results.
A period of 20years was established as a reason-
able recall period because it was assumed to be suf-
ficient to detect major environmental and social
changes without requiring people to assess very old
memories (as in Tesfamichael et al. 2014; Dama-
sio et al. 2015), which are more subjected to bias
and memory losses (Diamond et al. 2020). To aid
the recall process, fishers’ memories for 1994 were
elicited by an important event in the Brazilian cul-
ture: winning the world cup. When asked about their
catches, the question would end with a reference to
the year when Brazil beat Italy in the final game.
Using eliciting landmark events is a technique shown
to work in this type of interview (Matlin 2009).
Following the Brazilian research code of eth-
ics (Federal Resolution No 466/2012), all fishers
were informed about the research aims, benefits,
and possible risks. Permission was granted by ver-
bal consent before the interview began. Data sam-
pling was approved by the Federal University of
Rio Grande do Norte Ethics Committee (CAAE
68,531,917.6.0000.5537). A total of 394 fishers were
interviewed. They were on average 49.3 years old,
had very low levels of schooling (2.8years on aver-
age, i.e., incomplete elementary school) and had lived
in the same village for most of their lives (average of
43years).
The interviews collected: (1) socioeconomic data,
such as age, education, fishing experience, and num-
ber of economic activities carried out, and (2) fishing
data, such as species caught, quantity of each species
caught, average fishing time per trip and name of fish-
ing grounds, for both 1994 and 2014/2015, in addi-
tion to more general technological information on
fishing, such as types and sizes of fishing boats, type
of fishing gear and the amount of fuel used on each
fishing trip (for the detailed questions, see Supple-
mentary Information).
Additionally, the Sea Around Us Database (seaar-
oundus.org) was used to extract the ex-vessel price
per species. The latest prices available (2010) in the
database were used for 2014, and it was assumed
that no significant changes (other than inflation)
occurred during the period 2010–2014. The average
fisher income for each year (1994 and 2014) was cal-
culated by using the ex-vessel price per species and
the average amount of fish per target species that fish-
ers claimed to have caught on an average fishing trip,
multiplied by the number of fishing trips per month,
on average, in each period. To compare fishing
income with the country’s minimum wage in those
same years, the minimum wage for both years (1994
and 2014) was converted to USD, using the annual
average conversion rate of each year.
Fishing incomes were also compared with the
value of a Brazilian basket of goods. Given that this
value is adjusted according to inflation, it can more
accurately show the variation in fishing incomes dur-
ing the period analyzed. Minimum wage, on the other
hand, can be either adjusted below or above inflation.
The species were classified into three catego-
ries (first, second and third) according to the system
established by fishers and middlemen, which divides
the categories by price (established based on market
demand and desirability of a fish). Fish in the 1st cat-
egory are the most expensive, followed by fish in the
2nd category, and less valuable fish in the 3rd cate-
gory. Given that the fish price database obtained from
the Sea Around Us is national and not local, the fish
species cited by fishers were sorted from the high-
est to the lowest prices and then divided into three
quartiles in order to deal with the lack of local price
details. Fish in the first quartile of the national data-
base were assumed to also be in the first quartile in
the studied region. The first quartile included the 25%
most expensive species and was, therefore, classified
as 1st category. The third quartile included the 25%
cheapest species and was classified as 3rd category
species. The second quartile and the third quartile
Rev Fish Biol Fisheries
1 3
Vol.: (0123456789)
were pulled together to accommodate the remaining
species, which represented the 2ndcategory fish.
Statistical analyses
The analytical strategy adopted involved four steps:
(i) verify if the poorest fishers in 1994 remain the
poorest in 2014, i.e., whether they were in a poverty
trap (Spearman’s rank correlation); (ii) investigate
which variables (i.e., state, residence time, education,
fishing experience, number of economic activities,
type and size of boat, gear, fishing hours and category
of fish caught) influence income in each year (Gener-
alized Linear Model); (iii) test whether there is a cor-
relation between income in 1994 and the number of
changes a fisher will eventually perform until 2014,
as eventual negative correlation could suggest that
the poorest fishers simply could not afford to change
(Pearson’s correlation); and (iv) analyze whether the
number of changes a fisher performs in his fishing
grounds, effort and gear or target species influences
the variation of income he has between 1994 and
2014 (Proportional Odds Model).
Analyzing whetherfishers are inapoverty trap
To investigate whether fishers with the lowest income
in 1994 continued to have the lowest income in 2014,
all fishers were ranked by income, from the lowest
to the highest income, for both years. The two ranks
were then compared using the Spearman’s rank cor-
relation, considering the order that fishers appeared to
measure the strength of association between the two
ordinal variables.
The Spearman’s rank-order correlation is the
nonparametric version of the Pearson product-
moment correlation (see below). Spearman’s cor-
relation coefficient (ρ) measures the strength and
direction of the association between two ranked
variables (here, income in 1994 × income in 2014).
This measure was deemed appropriate given that
the idea was simply to compare whether fishers
maintained their income position over time, in rela-
tion to other fishers. For example, a strong correla-
tion would suggest low economic mobility (up or
down) among fishers.
Analyzing thevariables thataffect fisher incomes
each year
A Generalized Linear Model (GLM) was performed
for each year (two models) to investigate the influ-
ence of 11 independent variables on fisher incomes,
namely: State (categorical—Rio Grande do Norte
or Ceará), time of residence in the village (number
of years), formal education (number of years), age
(years), fishing experience (number of years), eco-
nomic activities (number of different activities),
vessel type (categorical—sail boat, motor canoe and
motorboat) and vessel size (meters), gear (categori-
cal—compressor, gillnet, line and trap), average
time spent on each fishing trip (hours) and category
of fish caught (categorical—1, 2 or 3, where 1 rep-
resents the most expensive fish and 3 the cheapest
ones). These variables represent the socioeconomic
characteristics of fishers as well as attributes of the
fishing material used by them. To achieve the nor-
mality assumption for the residuals, the response
variables (fisher incomes in each year) were loga-
rithmic transformed and a Gaussian distribution
with an identity link was implemented in the GLMs.
To determine which explanatory variables would
fit into the model, all variables were first checked
for correlation (see more details in Figures SI1 and
SI2 in the Supplementary Information) and then
evaluated for collinearity using the Variance Infla-
tion Factor (VIF). Only variables with a VIF < 3
were included in the model (Supplementary Infor-
mation Figures SI3 and SI4). To establish the best
model, two different measures were computed: the
Akaike Information Criterion (AIC) and the devi-
ance explained (D2%). The best model was the one
with the lowest AIC (Akaike 1974) and the high-
est D2%. Only the final model is presented in the
results. The software R (Team 2018) was used here,
together with the ‘car’ (Fox etal. 2019), ‘broom’
(Robinson and Hayes 2019), ‘Hmisc’ (Harrell Jr
2019), ‘jtools’ (Long 2019), ggplot2 (Wickham
et al. 2019), ‘ggstance’ (Henry et al. 2019), and
‘corrplot’ (Wei and Simko 2017) packages.
Analyzing whetheradaptive strategies are associated
withincome in1994
To investigate whether the fishers who had the low-
est income in 1994 remained with the lowest income
Rev Fish Biol Fisheries
1 3
Vol:. (1234567890)
in 2014 because, for example, their original financial
condition prevented them from making changes, a
Pearson’s correlation was performed. The intention of
this correlation was to verify whether income in 1994
was associated to the number of fisheries-related
changes made by fishers in the following years.
Investigating whetherfishing‑related changes
contributed toincrease fisher incomes
To analyze whether the number of changes made by
the fishers influenced income variation, the percent-
age of income variation between 1994 and 2014
was first calculated for each fisher. As some fishers
showed a negative variation in income (the income
in 2014 was lower than in 1994), a regular GLM was
not fit, because the interpretation of the estimated
coefficients of each explanatory variable would be
unclear. Thus, here it was opted for the categorization
of changes in income into four classes (see below),
which then allowed the use of a proportional odds
model, run with the package ‘mass’ in R Program,
using the function ‘polr’: where P gives the probabil-
ity of a level (j; here four levels) of income (Y) in one
of the two years;
𝛼
is the intercept and
𝛽
is the slope
coefficient for each predictor x (EquationS1).
In this model, the dependent variable ‘variation
in income’ was treated as a response scale (levels): 1
(n = 134)—income in 1994 was at least 50.1% higher
than in 2014 (highest losses); 2 (n = 111)—income
in 1994 was up to 50% higher than income in 2014
(intermediate losses); 3 (n = 56)—income in 2014
was up to 50% higher than income in 1994 (inter-
mediate gains); and 4 (n = 88)—income in 2014 was
at least 50.1% higher than income in 1994 (highest
gains). The classification of fishers into four groups,
instead of simply into losers and winners, intended to
account for the fact that those that lose the most may
be at a higher risk of being trapped in poverty.
The independent variables initially used in the
full model belonged to two different types. The first
one controlled for socioeconomic aspects: State (cat-
egorical—Rio Grande do Norte or Ceará), time of
residence in the village (number of years), formal
education (number of years), age (years), and fish-
ing experience (number of years). The second group
of variables accounted for any changes that fishers
made to how they fish. In this case, the number of
changes was summed (one point per change). These
changes included: type (50.2% of the fishers changed
their boat type) and size of boat (50.8%), category of
fish caught (53.3%), species caught (29.7%, here the
most caught species cited by the fishers in both years
was considered), time spent fishing (32.2%), and
gear used (14.47%). These were all binary variables
(0 = no change, 1 = changed).
The direction of the change (increase or decrease)
was not considered, as there was not enough varia-
tion in the data (most changes were in a single direc-
tion, such as increasing the size of the boat). Finally,
two other variables that were used interchangeably
in the model attempted to capture the total number
of changes in two different ways: number of changes
(sum of all the changes made up to a maximum of
5) and categorical changes, in which people that did
not perform any change were classified as “did not
change”, those performed between 1 and 3 changes
were classified as “few changes” and those that per-
formed more than 3 changes were classified as “many
changes”. Variables were dropped out of the initial
models whenever their significance was < 0.1. To
test if the final results were not obtained by chance,
a training data set was used with 80% of the original
data. All results were compared between the final
model and the training model. Given that the results
were very similar, the final model was considered
reliable.
In addition to identifying the most significant
model, a confusion matrix was also calculated to test
whether the final model was capable to accurately
predict the category of ‘variation in income’ of a
fisher.
Results
For the vast majority of interviewed fishers, their
overall income continued to come only from fish-
ing: 355 fishers out of 394 listed fishing as their only
source of income in 1994, whereas 328 listed fishing
as their only source of income in 2014.
In general, fishing incomes were 30% lower in
2014 than in 1994 and fishers lost income regardless
of the comparison made, whether between minimum
wages or between Brazilian basket of goods. Whereas
in 1994 fisher incomes were 4.3 times the value of
minimum wage and 5.24 times the value of a basket
of goods, in 2014 fisher incomes were only 1.2 times
Rev Fish Biol Fisheries
1 3
Vol.: (0123456789)
the value of minimum wage and 3.11 times the value
of a basket of goods. Thus, in the most conservative
estimate (basket of goods), fishers lost 41% of their
income over the course of 20years.
Yet, not all fishers lost income in the same propor-
tion and some even increased their fishing income.
Of all the fishers interviewed, 38% (149) reported an
increase in their fishing income between 1994 and
2014, compared to 62% (245) who experienced a
decrease.
Fishers caught in a poverty trap
The fishers with the lowest incomes in 1994 in gen-
eral tended to also have the lowest incomes in 2014
(p = 0.00, rho = 0.6).
Fisher income in each period
Fisher incomes in 1994 were mainly explained by
the State where fishing took place, their individual
experience, type of boat, the gear used, time spent
fishing, category of species caught, and, secondar-
ily, size of boat (Supplementary Information, Table
SI1) (Fig.2a). Those who used the hookah compres-
sor and larger motorized boats tended to have better
incomes (as these are on the right side of the graph
and further away from zero), whereas living in the
State of Rio Grande do Norte and catching 2nd and
3rd category fish had the opposite effect (on the left
side of the graph). Although less relevant (effect size
close to zero), having more experience in fishing and
spending longer hours fishing also contributed to hav-
ing higher incomes in 1994 (Fig.2a).
In 2014, higher incomes were mainly explained
by ownership of a larger boat, using hook and line
to fish, and more time spent fishing (Supplementary
Information, Table SI2). Although these three vari-
ables are significant, boat size has an effect size close
to zero (Fig.2b).
Number of adaptive strategies adopted did not
correlate with past income
Only 60 fishers (15.2%) made no changes to the fish-
ing technology used (size and type of boat and fish-
ing gear used), fishing hours, species and species cat-
egory caught or fishing grounds used. With respect to
fishers who made or experienced at least one change,
these changes occurred mainly in the category of fish
caught (53.7%), followed by boat type (50.1%), boat
size (49.1%), and time spent fishing (32.6%). Fish-
ers tended to continue catching higher value species
while also increasing the catch of lower value species.
They also tended to adopt larger, motorized boats,
and increase the average time spent fishing on each
trip. These changes were observed in both the fisher’s
group that had increased income and the fishers who
lost income. The analysis showed that income in 1994
did not associate with the number of changes made
by each fisher (R = 0.12), suggesting that the adop-
tion or not of changes between 1994 and 2014 was
not related to how much money fishers were making
from fishing in 1994.
Fig. 2 Effect size of significant variables in the GLM model
from fisher incomes in 1994 (a) and 2014 (b). When the vari-
able is on the negative side, the effect of the variable on the
response variable is negative, whereas if it is on the positive
side it has the opposite effect. The magnitude of the coeffi-
cient represents how important the variable is, the higher the
coefficient, the greater the effect. As none of the variables
crosses the zero, they are all significant. Acronyms are: State
RN = fishers in the state of Rio Grande do Norte (in compari-
son with fishers from Ceará State); Experience: fishing experi-
ence; Boat Type – 3: motor boat; Fish category: classification
of fish according to value, with 2 and 3 being the lowest levels
Rev Fish Biol Fisheries
1 3
Vol:. (1234567890)
Adaptive strategies to increase fisher incomes
The proportional odds logistic regression showed that
what most influenced whether a fisher gained little or
much or lost little or much of his income over time
was his initial income in 1994, his experience and
whether or not he changed the type of boat he owned
(Table1).
Specifically, fishers who changed their type of
boat, who were less experienced and who had lower
revenues in 1994, in that order, were slightly more
likely to be found in larger categories of earnings
(lost less or actually gained something) in 2014.
Although significant, the model should be interpreted
with caution, given the low value of the estimates,
especially for revenues in 1994 and experience. In
addition, the confusion matrix showed a poor overall
performance of the model (61.44% of misclassifica-
tion): it accurately predicted fishers who suffered gen-
eral losses of more than 50% (9% of incorrect clas-
sification), showed relatively good prediction power
for those who lost up to 50% of their income (31%
of misclassification), and could not predict well who
gained income (70 and 72% of misclassification for
gains > 50% and up to 50%, respectively).
Discussion
In general, fishers from the equatorial region of Brazil
have become poorer over time, and the poorest fishers
in 1994 continued to be the poorest in 2014. Fishers’
income has declined by 41% while fishing remains
the only source of income to the interviewed fish-
ers. Perhaps in an effort to deal with harsher condi-
tions, most fishers made fishing-related changes that
increase fishing effort, such as fishing in larger, yet
small-scale, or more powerful vessels. But some of
the changes may also have been an adaptation to deal
with stock declines, including switching their focus to
catching less expensive and desirable species (second
and third category species).
Some factors that positively influenced income in
1994, such as living in the State of Ceará and using a
hookah compressor for gear, may be related to lobster
fishing. Lobster was one of the main export products
in the 1990’s and the State of Ceará has the largest
abundance in Brazil of two lobster species (Panulirus
argus e P. laevicauda) due to favorable oceanographic
conditions in the region (Fonteles Filho 1994). Hook-
ahs, however, were forbidden in 1995, although there
is poor enforcement of their use (Dias Neto 2010).
The capture of 2nd and 3rd category fish in 1994
is negatively related to income precisely because
they are the fish with the lowest economic values. In
2014, fishers using lines had higher yields, probably
because species caught through line fishing are usu-
ally large bodied species or with high commercial
value (MMA 2006). Interestingly, different gears had
a positive relationship with income in the years stud-
ied (hookah in 1994 and line in 2014), yet the vast
majority of fishers did not change the gear they use
for fishing.
The pattern of vessel improvement seen in this
study, which is related to income in the two years ana-
lyzed, has also been documented globally, whereby
motor boats and overall fleet sizes have increased by
more than six-fold, yet global catches have declined
(Rousseau etal. 2019). Previous studies carried out in
a close-by region suggest that in small-scale fisheries,
larger boats (but they remain artisanal or small-scale)
do not necessarily result in higher catches and more
profit, although they can increase fishing expenses
(Damasio etal. 2016, 2020). Clearly, income losses
are expected if costs rise without being followed by
equivalent increases in catch or in catch value. When
associated with declining fish stocks, and therefore
catches, improvements to boats may only be mask-
ing economic losses (Pauly etal. 2002; Damasio etal.
2016), delaying a real solution to ecological and eco-
nomic matters, and further worsening the financial
situation of fishers.
An important aspect identified among the factors
that put some fishers in the poorest economic strata
was not to change the type of boat they use. This
effect, although small, may suggest that changing to a
different type of boat, for example, from sailboat to a
Table 1 Significant predictors, derived from proportional
odds model, for the changes in income fishers over a 20-year
period (1: > 50% losses; 2: < 50% losses; 3: < 50% gains;
4: > 50% gains)
Variable Estimate SE t value p-value
Changed his boat
type
0.56131 0.18734 2.99 0.00273
Revenue in 1994 − 0.00064 0.000132 − 4.88 0.00000
Experience − 0.01599 0.00857 − 1.86 0.06234
Rev Fish Biol Fisheries
1 3
Vol.: (0123456789)
motorboat, possibly allowed fishers to explore differ-
ent species or parts of stocks not previously accessed.
This result is a phenomenon known as hyperstability,
as the characteristics of the fishers and exploited fish
help to maintain catches as stocks decline (Erisman
et al. 2011). Hyperstability seems to be especially
likely to occur in the tropics because many commer-
cially important reef species form predictable repro-
ductive aggregations that are well known or easily
identified by fishers, especially as the latter adapt
their strategies to explore different grounds (Ham-
ilton et al. 2016). An indication that hyperstability
has probably happened in this region is the recorded
overfishing of species exploited in the region, such
as snappers and groupers, despite some evidence of
increased catches (Silvano etal. 2017). Another indi-
cation that fishers may be trying to adapt to declining
stocks is the fact that the spatial distribution of local
fisheries has shifted from shallow to deeper waters
(Damasio et al. 2020; Freire et al. 2021), which
allows them to exploit previously unassessed parts of
the stock.
With the data collected for this study it was not
possible to clearly identify factors that, on their own,
promote an increase or a decrease in incomes over
time. This was especially true for the factors that
could have contributed to making people less poor (as
measured through fishing income), as the model used
here did not perform well to predict strategies that
would have resulted in economic gains, possibly due
to the small sample size of fishers who have become
richer over time. However, the model predicted,
with good accuracy, factors that kept people poor or
impoverished them in the long run.
Although the results presented here are not a defin-
itive answer to the factors that promote an increase or
a decrease in incomes in fisheries, the fact that fish-
ers with lower incomes in 1994 remained as such in
2014 characterizes a poverty trap (Azariadis and Sta-
churski 2005; Cinner etal. 2009; Barrett etal. 2011).
A poverty trap emerges whenever the opportunity to
increase income or wealth is limited for those who
have very little to invest, but it does not impact those
who can invest a little more (Banerjee and Duflo
2012; Villasante etal. 2022).
One aspect that could explain increased fishing-
related poverty is the depletion of fish stocks, given
that despite their accumulated experience (over
20years), which should have facilitated their access
to stocks (Chen and Chiu 2009; Heldt etal. 2021),
fishers are now gaining less and catching more of
the less desirable species. Yet, we do not claim that
overfishing alone explains some aspects of poverty in
fishing communities (e.g., low physical and financial
capital) (Béné 2003; Hirway 2010).
Poverty is a multidimensional variable that mainly
encompasses breakdowns in socio-institutional mech-
anisms, leading to social exclusion, failure to access
rights and political powerlessness (Sen 1981). A lim-
ited understanding of the mechanisms driving aspects
of poverty may also contribute to accentuating it. For
example, harmful subsidies that neglect fishing costs
and reinforce fisher’s beliefs that improving boats
will lead to increased earnings, associated with a gen-
eral lack of fishing data, prevent fishers and manag-
ers from grasping the intertwined dynamics of (over)
fishing and social conditions (Sumaila et al. 2021).
This leads to a vicious cycle whereby poorer fish-
ers are less likely to emerge from declining fishing,
which further degrades the resource and generates
less income (Cinner 2005; Cinner etal. 2009; Villas-
ante etal. 2022). If fishers are trapped, it is unreason-
able to expect they will escape fishing-related poverty
on their own. In this case, specific socio-institutional
mechanisms are necessary (Béné 2003).
A solution widely explored in literature to improve
financial capital in communities that depend on the
exploitation of natural resources focuses on income
diversification. But this measure does not address
the cause of fishing poverty. Even though it is known
that fishing activity is subject to several fluctuations
in addition to market variations (natural fluctuations
of species, bad weather, loss or destruction of equip-
ment, etc.) fishers are often found in an economic,
political and institutional marginalization. Such a
state leads to most fishers being denied access to eco-
nomic institutions or even not getting help in times of
crisis, a small crisis like a boat breaking down can be
doom for a marginalized fisher. Without such access,
some fishers remain unable to reach the minimum
level of investment that would allow them to generate
greater financial benefits and to lift themselves out of
the level of low productivity and income poverty in
which they are trapped (Pauly 2005; Béné and Friend
2011).
Like elsewhere in the world, small-scale fishers
in Brazil face major threats and challenges that can
push them to the edge of poverty and misery, such as
Rev Fish Biol Fisheries
1 3
Vol:. (1234567890)
conflicts over access to resources and fishing grounds,
depletion of fish stocks, depletion of grounds due to
coastal development, concentration of market power
and climate change (Jentoft etal. 2011; FAO 2016).
Poverty alleviation strategies directed at small scale
fisheries have mostly included attempts to acceler-
ate economic growth, market-led economic pol-
icy reforms and technological and infrastructural
improvements (World Fish Center 2005). The latter
are the most tangible actions that fishers can take
to maximize their earnings, as observed here. How-
ever, these actions overlook the consequences on fish
stocks and temporarily mask stock depletion by com-
pensating declining catches with increasing access to
otherwise inaccessible resources (Pauly etal. 2002).
As complex problems require complex solutions,
reducing poverty in fisheries requires interventions
aimed at improving access to public services (trans-
port, education, water, electricity, health) and eco-
nomic institutions, while actions are taken to improve
fisheries productivity. It should also involve the col-
lective effort of a set of actors and institutions: state
governments, civil society organizations, university,
fishing and fish workers (Jentoft etal. 2018).
Conclusions
Understanding the factors that influence small-scale
fisher income is important from both the point of
view of fisheries management and eradication of hun-
ger and poverty. This is especially relevant in places
like the Brazilian equatorial region where few liveli-
hood alternatives and social security programs are
available, and where poverty, measured as fishing-
related income, seems to have worsened over time
among fishers. Moreover, this is a region that has
been severely impacted by recent environmental dis-
asters (e.g., an oil spill in 2019) and had to deal with
the effects of a global pandemic (Silva etal. 2022).
It was certainly not ready to bear both burdens con-
sidering that the pre-disaster situation had already
exposed the existence of poverty traps.
In Brazil specifically, the collection of basic socio-
economic data, such as number of fishers and income,
in addition to basic fishing data, could contribute to
designing tailor-made strategies to lift people out of
poverty. Turning a blind eye to the efforts of fishers to
cope with their difficulties at a time when fish stocks
continue to deteriorate and recent environmental
and economic/health disasters have hit them almost
simultaneously, will only further endanger the live-
lihoods of fishers and accentuate the pressure on the
environment.
Acknowledgements LMAD thanks the Coordenação de
Aperfeiçoamento de Pessoal de Nível Superior (CAPES,
Brazil; Finance Code 001) for a PhD scholarship. ARCA
(313334/2018-8) and PFML thanks CNPq for a productivity
Grant (301515/2019-0).
Funding This research did not receive any specific grant
from funding agencies in the public, commercial, or not-for-
profit sectors.
Data availability All data generated or analysed during this
study are included in this published article [and its supplemen-
tary information files].
Declarations
Conflict of interest The authors declare that they have no
known competing financial interests or personal relationships
that could have appeared to influence the work reported in this
paper.
References
Akaike H (1974) A new look at the statistical model identifica-
tion. IEEE Trans Autom Control 19:716–723. https:// doi.
org/ 10. 1109/ TAC. 1974. 11007 05
Asche F, Smith M D (2010) Trade and fisheries: key issues for
the world trade organization. WTO
Azariadis C, Stachurski J (2005) Chapter5 Poverty Traps. In:
Handbook of economic growth. Elsevier, pp 295–384
Banerjee A, Duflo E (2012) Poor economics. PublicAffairs,
New York
Barrett CB, Travis AJ, Dasgupta P (2011) On biodiversity
conservation and poverty traps. Proc Natl Acad Sci
108:13907–13912. https:// doi. org/ 10. 1073/ pnas. 10115
21108
Béné C (2003) When fishery rhymes with poverty: a first step
beyond the old paradigm on poverty in small-scale fish-
eries. World Dev 31:949–975. https:// doi. org/ 10. 1016/
S0305- 750X(03) 00045-7
Béné C, Friend RM (2011) Poverty in small-scale fisheries: old
issue, new analysis. Prog Dev Stud 11:119–144. https://
doi. org/ 10. 1177/ 14649 93410 01100 203
Béné C, MacFadyen G, Allison EH (2005) Increasing the con-
tribution of small-scale fisheries to poverty alleviation and
food security. FAO, Rome
Chen C-S, Chiu T-S (2009) Standardising the CPUE for the
Illex argentinus fishery in the Southwest Atlantic. Fish Sci
75:265–272. https:// doi. org/ 10. 1007/ s12562- 008- 0037-1
Rev Fish Biol Fisheries
1 3
Vol.: (0123456789)
Cinner J (2005) Socioeconomic factors influencing customary
marine tenure in the Indo-Pacific. Ecol Soc 10:36
Cinner JE, Daw T, McCLANAHAN TR (2009) Socioeco-
nomic factors that affect artisanal fishers’ readiness to exit
a declining fishery. Conserv Biol 23:124–130. https:// doi.
org/ 10. 1111/j. 1523- 1739. 2008. 01041.x
Damasio LMA, Peninno MG, Lopes PFM (2020) Small
changes, big impacts: geographic expansion in small-scale
fisheries. Fish Res 226:105533. https:// doi. org/ 10. 1016/j.
fishr es. 2020. 105533
de Damasio LMA, Lopes PFM, Guariento RD, Carvalho AR
(2015) Matching fishers’ knowledge and landing data
to overcome data missing in small-scale fisheries. PLoS
ONE 10:e0133122. https:// doi. org/ 10. 1371/ journ al. pone.
01331 22
de Damasio LMA, Lopes PFM, Pennino MG etal (2016) Size
matters: fishing less and yielding more in smaller-scale
fisheries. ICES J Mar Sci J Cons 73:1494–1502. https://
doi. org/ 10. 1093/ icesj ms/ fsw016
Diamond NB, Armson MJ, Levine B (2020) The truth is out
there: accuracy in recall of verifiable real-world events.
Psychol Sci 31:1544–1556. https:// doi. org/ 10. 1177/ 09567
97620 954812
Erisman BE, Allen LG, Claisse JT etal (2011) The illusion of
plenty: hyperstability masks collapses in two recreational
fisheries that target fish spawning aggregations. Can J
Fish Aquat Sci 68:1705–1716. https:// doi. org/ 10. 1139/
f2011- 090
FAO (2008) Small-scale capture fisheries—a global overview
with emphasis on developing countries. A preliminary
report of the Big Numbers Project. FAO
FAO F and AO of the U (2016) Integrated ocean manage-
ment—fisheries, oil, gas and seabed mining, by Melanie
Torrie. Rome, Italy
FAO F and AO of the U (2017) Improving our knowledge
on small-scale fisheries: data needs and methodologies:
workshop proceedings, 27–29 June 2017, FAO, Rome,
Italy
FAO (ed) (2018) Meeting the sustainable development goals.
Rome
Fonteles Filho AA (1994) A pesca predatória de lagostas no
estado do Ceará: causas e consequências. Bol Tec Cient
CEPENE 2:107–131
Fox J, Weisberg S, Price B (2019) car package—companion to
applied regression
Freire KMF, da de Almeida ZS, Amador JRET etal (2021)
Reconstruction of marine commercial landings for the
Brazilian industrial and artisanal fisheries from 1950 to
2015. Front Mar Sci 8:659110. https:// doi. org/ 10. 3389/
fmars. 2021. 659110
Garcia F, Abdallah P, Sachsida A (2018) Avaliação de efeitos
do programa PROFROTA Pesqueira sobre indicadores do
mercado de trabalho. IPEA, Brasilia
Hamilton RJ, Almany GR, Stevens D etal (2016) Hypersta-
bility masks declines in bumphead parrotfish (Bolbome‑
topon muricatum) populations. Coral Reefs 35:751–763.
https:// doi. org/ 10. 1007/ s00338- 016- 1441-0
Harrell Jr F (2019) Hmisc package—Harrell miscellaneous
Heldt K, Stobart B, Mayfield S (2021) Diving into fisher
experience: Do new entrants and fleet turnover depress
catch rates in abalone (Haliotis laevigata and H. rubra)
fisheries. Fish Res 238:105906. https:// doi. org/ 10.
1016/j. fishr es. 2021. 105906
Henry L, Wickham H, Chang W (2019) ggstance package—
horizontal “ggplot2” Components
Hirway I (2010) Understanding poverty: insights emerging
from time use of the poor. In: Antonopoulos R, Hirway I
(eds) Unpaid Work and the Economy. Palgrave Macmil-
lan, London, pp 22–57
IBAMA (2007) Estatistica da Pesca 2007, Brasil—grandes
regioes e unidades da Federacao. Estat Pesca 1–151
Jentoft S, Bavinck M, Alonso-Población E etal (2018) Work-
ing together in small-scale fisheries: harnessing collec-
tive action for poverty eradication. Marit Stud 17:1–12.
https:// doi. org/ 10. 1007/ s40152- 018- 0094-8
Jentoft S, Eide AH, Gunnarsdottir M-V (eds) (2011) Poverty
mosaics: realities and prospects in small-scale fisheries.
Springer, New York
Long J (2019) jtools package—analysis and presentation of
social scientific data
Matlin MW (2009) Cognition, 7th edn. Wiley, Hoboken
MMA (2006) Programa REVIZEE: avaliação do poten-
cial sustentável de recursos vivos Na Zona Econômica
Exclusiva do Brasil—Relatório Executivo. Ministério
do Meio Ambiente, Brasília
MMA (2008) Monitoramento da atividade pesqueira no lito-
ral nordestino—Projeto ESTATPESCA. Ministério do
Meio Ambiente, Tamandare
Neto JD (2010) Pesca no Brasil e seus aspectos institucion-
ais—um registro para o futuro. Rev CEPSUL - Biodi-
versidade E Conserv Mar 1:66–80. https:// doi. org/ 10.
37002/ revis tacep sul. vol1. 30066- 80
NIS N de IS (2020) Índice de pobreza multidimensional.
Belo Horizonte
Pauly D (2005) Rebuilding fisheries will add to Asia’s prob-
lems. Nature 433:457–457. https:// doi. org/ 10. 1038/
43345 7a
Pauly D, Christensen V, Guénette S et al (2002) Towards
sustainability in world fisheries. Nature 418:689–695.
https:// doi. org/ 10. 1038/ natur e01017
Robinson D, Hayes A (2019) broom package —convert sta-
tistical analysis objects into tidy tibbles
Rousseau Y, Watson RA, Blanchard JL, Fulton EA (2019)
Evolution of global marine fishing fleets and the
response of fished resources. Proc Natl Acad Sci
116:12238–12243. https:// doi. org/ 10. 1073/ pnas. 18203
44116
Sala E, Mayorga J, Costello C etal (2018) The economics of
fishing the high seas. Sci Adv 4:eaat2504. https:// doi. org/
10. 1126/ sciadv. aat25 04
Sen A (1981) Poverty and famines: an essay on entitlement and
deprivation. Clarendon Press; Oxford University Press,
Oxford
Sethi SA, Dalton M, Hilborn R (2012) Quantitative risk meas-
ures applied to Alaskan commercial fisheries. Can J Fish
Aquat Sci 69:487–498. https:// doi. org/ 10. 1139/ f2011- 170
Silva MRO, Silva AB, Barbosa JC et al (2022) Empower-
ing fisherwomen leaders helped reduce the effects of the
COVID-19 pandemic on fishing communities: insights
from Brazil. Mar Policy 135:104842. https:// doi. org/ 10.
1016/j. marpol. 2021. 104842
Rev Fish Biol Fisheries
1 3
Vol:. (1234567890)
Silvano RAM, Nora V, Andreoli TB etal (2017) The ‘ghost
of past fishing’: small-scale fisheries and conservation
of threatened groupers in subtropical islands. Mar Policy
75:125–132. https:// doi. org/ 10. 1016/j. marpol. 2016. 10. 002
Smith H, Basurto X (2019) Defining small-scale fisheries and
examining the role of science in shaping perceptions of
who and what counts: a systematic review. Front Mar Sci.
https:// doi. org/ 10. 3389/ fmars. 2019. 00236
Steneck RS, Pauly D (2019) Fishing through the Anthropo-
cene. Curr Biol 29:R987–R992. https:// doi. org/ 10. 1016/j.
cub. 2019. 07. 081
Sumaila UR, Skerritt DJ, Schuhbauer A et al (2021) WTO
must ban harmful fisheries subsidies. Science 374:544–
544. https:// doi. org/ 10. 1126/ scien ce. abm16 80
Team RDC (2018) R: a language and environment for statisti-
cal computing
Tesfamichael D, Pitcher TJ, Pauly D (2014) Assessing changes
in fisheries using fishers’ knowledge to generate long time
series of catch rates: a case study from the Red Sea. Ecol
Soc. https:// doi. org/ 10. 5751/ ES- 06151- 190118
Villasante S, Gianelli I, Castrejón M etal (2022) Social-eco-
logical shifts, traps and collapses in small-scale fisher-
ies: envisioning a way forward to transformative changes.
Mar Policy 136:104933. https:// doi. org/ 10. 1016/j. marpol.
2021. 104933
Wei T, Simko V (2017) corrplot package—visualization of a
correlation matrix
Wickham H, Chang W, Henry L, et al (2019) ggplot2 pack-
age—create elegant data visualisations using the grammar
of graphics
World Fish Center (2005) Towards an interdisciplinary
approach to the assessment of small-scale fisheries and its
role in food security and poverty alleviation and sustain-
able resource use. Penand and Rome
Publisher’s Note Springer Nature remains neutral with regard
to jurisdictional claims in published maps and institutional
affiliations.
Springer Nature or its licensor (e.g. a society or other partner)
holds exclusive rights to this article under a publishing
agreement with the author(s) or other rightsholder(s); author
self-archiving of the accepted manuscript version of this article
is solely governed by the terms of such publishing agreement
and applicable law.
- A preview of this full-text is provided by Springer Nature.
- Learn more
Preview content only
Content available from Reviews in Fish Biology and Fisheries
This content is subject to copyright. Terms and conditions apply.
