JOURNAL OF THE Vol. 48, No.

2
WORLD AQUACULTURE SOCIETY April, 2017
doi: 10.1111/jwas.12423

Economics of Sustainable Intensification of Aquaculture: Evidence

from Shrimp Farms in Vietnam and Thailand
Carole R. Engle 1

Engle-Stone Aquatic$ LLC, 320 Faith Lane, Strasburg, Virginia, 22657, USA

Aaron McNevin and Phoebe Racine

World Wildlife Fund, Washington, District of Columbia, 20037, USA

Claude E. Boyd
Fisheries and Allied Aquacultures, Auburn University, 203B Swingle Hall, Auburn, Alabama, 36849, USA

Duangchai Paungkaew and Rawee Viriyatum

World Wildlife Fund, Bangkok, Thailand

Huynh Quoc Tinh and Hang Ngo Minh

World Wildlife Fund, Can Tho, Vietnam

Abstract
There is growing interest in sustainable intensification of aquaculture production. Yet little economic
analysis has been done on farm-level effects of the economic sustainability of production intensification.
Data from 83 shrimp farms (43 in Vietnam and 40 in Thailand) were used to identify (through principal
component and cluster analyses) 13 clusters of management practices that reflected various scales
of production intensity that ranged from 0–1999 kg/ha/crop to 10,000 kg/ha/crop and above, for
both Penaeus monodon and Litopenaeus vannamei in Vietnam and Thailand. The clusters identified
reflected sets of management practices that resulted in differing yields despite similarities in stocking
densities among some clusters. The enterprise budget analysis developed showed that the more
intensively managed clusters outperformed the less intensively managed clusters in economic terms.
More intensively managed farm clusters had lower costs per metric ton of shrimp produced and were
more profitable. The greater yields of shrimp produced per hectare of land and water resources in
more intensively managed shrimp farms spread annual fixed costs across a greater volume of shrimp
produced and reduced the cost per metric ton of shrimp. Costs per metric ton of shrimp produced
decreased from the lowest to the highest intensity level (from US$10,245 at lowest intensity to US$3484 at
highest for P. monodon and from US$24,301 to US$5387 for L. vannamei in Vietnam and from US$8184
at the lowest intensity level to US$3817 at the highest intensity level per metric ton for L. vannamei in
Thailand). Costs of pond amendments used in shrimp production were particularly high in Vietnam
and largely unwarranted, whereas fixed costs associated with the value of land, production facilities,
equipment, and labor were sufficiently high in Thailand so that net returns were negative in the long
run. Nevertheless, economic losses in Thailand were less at greater levels of intensification. The study
demonstrated a clear value proposition for shrimp farmers to use natural resources (such as land) and
other inputs in an efficient manner and supports findings from corresponding research on farm-level
natural resource use efficiency. Additional research that incorporates economic analysis into on-farm
studies of sustainable intensification of aquaculture is needed to provide ongoing guidance related to
sustainable management practices for aquaculture.

KEYWORDS

shrimp economics, sustainability, sustainable intensification, Thailand, Vietnam

1 Correspondence to: [email protected]

© 2017 The Authors. Journal of the World Aquaculture Society published by Wiley Periodicals, Inc. on behalf of the World Aquaculture Society.
This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any
medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.

227
 17497345, 2017, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jwas.12423 by Yemen Hinari NPL, Wiley Online Library on [28/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
228 ENGLE ET AL.

Growth of aquaculture is necessary to provide more intensive production and cite lower feed
food for increasing global population levels, and conversion ratios (FCRs), fewer disease prob-
the need for such growth to be developed in lems, and greater rates of return. For US catfish
a sustainable and responsible manner is widely production, adoption of more intensive produc-
recognized as a necessary goal. However, much tion systems has been shown to reduce (Kumar
of the literature on sustainability lacks a sys- et al. 2016) or increase (Goode et al. 2002)
tematic and data-driven approach from which to per-unit costs of production depending on the
identify more sustainable production systems or capital requirements, production performance,
species (Engle and D’Abramo 2016). To add to and relative prices and costs.
the complexity of the search for more sustainable There is agreement in the aquaculture sustain-
aquaculture production is the reality that busi- ability literature on the importance of access
nesses must be profitable to be sustained over and availability of capital (Bush et al. 2010;
time. Yet there are few studies that have assessed Belton and Little 2011). While greater capital
the economic sustainability of alternatives sug- is needed for more intensively managed farms,
gested to increase environmental sustainability. even extensive production of aquaculture crops
Studies of aquaculture sustainability have often requires greater capital than does artisanal
tended to intertwine farm size and intensity of fishing or rice farming. Intensification increases
production. “Small-scale” production is often requirements for both capital and management
assumed to refer to low-input, extensive, or skill. Even if strong economic incentives exist to
subsistence production, while “large-scale” is improve productivity, farmers without adequate
frequently used to refer to intensive production
capital or management skill will be unable to
for export markets (Nakamura 1985; Bush
benefit from productivity gains (Waite et al.
et al. 2010). Vandergeest et al. (1999), however,
2014).
argued that small shrimp farms can be managed
Recent calls for sustainable intensifica-
as intensively as large farms. This is largely
tion of food production (Little et al. 2012;
supported by the technical efficiency literature
Waite et al. 2014; FAO 2016) may indicate
on aquaculture in which farm size has not been
an increased understanding of the potential
shown to consistently explain economic efficien-
environmental–economic–social benefits of
cies of aquaculture production generally (Iliyasu
et al. 2014), across counties/provinces within a intensification of aquaculture production. The
given country (Tan et al. 2011), or across coun- Food and Agriculture Organization of the United
tries and levels of production intensity within Nations (FAO) highlighted examples of 12 sus-
those countries (Dey et al. 2005). tainable intensification aquaculture systems,
Production intensification, on the other hand, one of which was closed/semi-closed intensive
spreads annual fixed costs over greater produc- production of shrimp in Thailand. However, no
tion volumes (economies of scale) and thereby economic analysis was performed of the shrimp
reduces costs per metric ton of production, for systems highlighted.
equivalent production systems managed accord- There is a surprising lack of farm-level eco-
ing to profit-maximizing conditions (Engle nomic analysis related to effects of intensifi-
2010). The trend toward increasing intensi- cation levels in aquaculture. Moreover, while
fication of shrimp culture is likely driven by there has been a tendency to define intensifi-
economies of scale that are common and widely cation based on stocking density (World Bank
recognized by aquaculture business managers. and MOFI 2006; Joffre and Bosma 2009), John-
However, diseconomies of scale can also occur son et al. (2014) showed that the combined set
if businesses grow too large to be managed of management practices, rather than a single
efficiently or through overcapitalization. Some production parameter such as stocking density,
authors (Thongrak 1995; New 1996; Patmasiri- must be considered when assessing farm-level
wat 1997) have argued that low-input extensive economic performance. Detailed farm-level data
systems are more economically efficient than and analyses are needed to determine the effects
 17497345, 2017, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jwas.12423 by Yemen Hinari NPL, Wiley Online Library on [28/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
ECONOMICS OF SUSTAINABLE INTENSIFICATION IN SHRIMP 229

of intensification on farm economic performance be found in Johnson et al. (2014) and Kumar
in major shrimp-producing countries. and Engle (2017). Such variation is not unique
Using farm-level survey data from Thailand to aquaculture, and multivariate techniques
and Vietnam, Boyd et al. (2017) showed that have been developed to group observations in a
intensive shrimp production is more efficient, way that identifies sets of on-farm management
uses fewer resources, and results in less environ- strategies that are not only distinct from each
mental impact per metric ton of shrimp produced other but also minimize within-group variation
than more extensive shrimp production systems. (MacQueen 1967). Cluster analysis is a useful
The present study used the same farm-level sur- multivariate tool that does not require prior
vey data to assess whether more intensive shrimp information on the population (Pielou 1984) but
production was economically preferable to less yet identifies groups of homogeneous entities
intensive shrimp production in Thailand and that are distinct from other groups (Prein et al.
Vietnam. Specific objectives were to (1) com- 1993; Hair 1995; Johnson and Wichern 2007).
pare cost structures and total costs across pro- The approach used in this study to identify
duction intensity levels, (2) compare net returns groupings of farms with similar management
above variable costs (short-term profits), and (3) strategies followed that of Johnson et al. (2014)
compare net returns above total costs (long-term and Kumar and Engle (2017). A principal com-
profitability) across production intensity levels ponent analysis was conducted to identify sets of
in Thailand and Vietnam. Results of this anal- variables that contributed similarly to the overall
ysis contribute to the ongoing search for ways to variability within the data set.
improve sustainability of the resources used in The second stage of the analysis was to group
shrimp production in a way that is also econom- the observations into clusters with similar char-
ically sustainable. acteristics. A cluster analysis (agglomerative
hierarchical algorithm) was conducted to iden-
tify groups of farm observations that were more
Methods
similar to each other than to observations in the
A survey was conducted in 83 farms (43 in other clusters in terms of key variables such as
Vietnam and 40 in Thailand) from April to country, species raised, pond size, stocking den-
November 2015. The survey covered a purposive sity, feeding rate, aeration use, and number of
sample of farms in each country and included days in production of each crop.
farms that raised shrimp in a variety of levels of Complete enterprise budgets were developed
intensity in three provinces in Vietnam and seven for each cluster identified based on standard bud-
provinces in Thailand. The survey instrument geting techniques (Engle 2010). Copies of com-
elicited information on the farm size, equip- plete budgets are available on request from the
ment used, production practices (stocking den- corresponding author. ANOVA was performed
sity, feeding rates, aeration rates, length of pro- on key parameters that included stocking rate,
duction cycle, number of crops used, and aver- survival, yield, feeding rate, FCR, days per crop,
age size of shrimp harvested), and production number of crops per year, pond size, and farm
input quantities and costs. Farmers were asked size. Mean values that were not significantly dif-
to report data for the previous production year. ferent from those of other clusters were averaged
Response rates for the survey were 100% in Viet- across all clusters for use in enterprise budgets.
nam and 98% in Thailand. When significant differences were found, mean
Surveys of commercial aquaculture farms values for each cluster were used in the corre-
demonstrate substantial variation in manage- sponding budget. Survival rates used were the
ment decisions related not just to stocking averages as reported by farms.
density but also to feeding rates, aeration rates, Tables of investment and depreciation costs
and harvesting strategies. Recent examples of of land, ponds, reservoirs, buildings, other
such variability and the interrelated effects on infrastructure items, and equipment were first
farm yields and per-unit costs of production can developed for the relevant farm size and then
 17497345, 2017, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jwas.12423 by Yemen Hinari NPL, Wiley Online Library on [28/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
230 ENGLE ET AL.

converted to a per-hectare basis. Tables of Table 1. Principal components and eigenvalues.

annual costs and returns were based on the mean
Principal Percent Cumulative
pond size for each cluster. Values for all line component Eigenvalue variance (%) variance (%)
items were converted to a per-hectare basis and
1 3.084 34 34
multiplied by the mean pond size (in hectares) 2 2.171 24 58
for each budget for each cluster. 3 1.312 15 73
4 0.656 7 80
5 0.553 6 86
Results
6 0.509 6 92
Principal Component and Cluster Analyses 7 0.391 4 96
8 0.207 2 98
Eight principal components were found to
account for 98% of the variability in the data
set (Table 1). Specific variables included in
these eight principal components were coun- Table 2. Clusters identified for economic analysis,
survey of shrimp farms in Vietnam and Thailand, 2015.
try, species raised, amount of aeration, pond
size, and the number of days in production of Intensity Yield range
each crop. Thirteen clusters were identified that Country Species category (kg/ha/crop)
minimized the production and management vari- Vietnam Penaeus monodon Low 0–1999
ability within each, including eight clusters in Vietnam P. monodon Medium 2000–4999
Vietnam and five in Thailand (Table 2). Clusters Vietnam P. monodon High 5000–9999
identified included production of Litopenaeus Vietnam P. monodon Very high 10,000 and
above
vannamei at all intensity levels in both countries,
Vietnam Litopenaeus Low 0–1999
production of Penaeus monodon at all intensity vannamei
levels in Vietnam, and P. monodon production Vietnam L. vannamei Medium 2000–4999
at a very extensive level in Thailand. The most Vietnam L. vannamei High 5000–9999
extensive level of production in Thailand was Vietnam L. vannamei Very high 10,000 and
above
a system that relied on water exchange from
Thailand P. monodon Low 0–1999
tidal flows that introduced blood cockles and Thailand L. vannamei Low 0–1999
mangrove crabs, which were then harvested and Thailand L. vannamei Medium 2000–4999
sold in addition to shrimp from the ponds, and Thailand L. vannamei High 5000–9999
was also distinct from the least intensive pro- Thailand L. vannamei Very high 10,000 and above
duction system of P. monodon in Vietnam. The
clusters identified resulted in four categories of
production intensity as identified by differences size of 6737 m2 , whereas the low-yield cluster
in yields of shrimp as follows: very high (yields with L. vannamei had a mean pond size of 4000
of 10,000 kg/ha/crop and above), high (yields of m2 . With P. monodon in extensive production in
5000–9999 kg/ha/crop), medium (2000–4999 Thailand, the mean pond size was 9.9 ha.
kg/ha/crop), and low (0–1999 kg/ha/crop).
Tables 3 and 4 present the mean values of
key production variables for each cluster that Production Performance across Levels
included stocking density, feeding rate, aeration of Intensification
rate, and yield, among others. Yield and feeding rate increased as the inten-
Mean farm size was significantly different sity of shrimp production increased, regardless
(P < 0.05) between countries (mean farm size of of country or species raised (Tables 3 and 4).
2.2 ha in Vietnam and 7.7 ha in Thailand). Mean There were some differences between countries
pond size (3620 m2 ) did not differ significantly for the very high intensity level, with mean
across clusters within Vietnam but did differ by yields in Thailand (13,560 kg/ha/crop) greater
cluster in Thailand. In Thailand, the very high, than those at the same intensity level in Vietnam
high, and medium clusters had a mean pond (11,324 kg/ha/crop for P. monodon and 11,702
 17497345, 2017, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jwas.12423 by Yemen Hinari NPL, Wiley Online Library on [28/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
ECONOMICS OF SUSTAINABLE INTENSIFICATION IN SHRIMP 231

Table 3. Mean values for key production parameters by categories of intensity/yield levels, Vietnam clusters.

Penaeus monodon Litopenaeus vannamei

Item Low Medium High Very high Low Medium High Very high
Stocking density (PL/m2 ) 12 23 58 59 26 31 66 73
Feeding rate (kg/ha/crop) 1119 5822 11,318 14,116 1246 4556 9114 15,485
Days in crop 143 124 120 90 102 90 87 110
Aeration rate (hp/ha) 8 10 21 21 4 12 28 50
Yield (kg/ha/crop) 895 3790 7962 11,324 265 3469 6974 11,702
Feed conversion ratio 1.25 1.54 1.42 1.25 4.7 1.31 1.31 1.32
Harvest weight (shrimp/kg) 40 37 41 66 58 85 72 59
Survival (%) 49 74 83 70 23 75 81 81
Crops per year 1 1 2 3 2 2 2 2
PL = postlarvae.

Table 4. Mean values for key production parameters, by categories of intensity/yield levels, Thailand clusters.

Penaeus monodon Litopenaeus vannamei

Item Low Low Medium High Very high
Stocking density (PL/m2 ) 2 62.5 62.5 82 99
Feeding rate (kg/ha/crop) 0 1681 4797 10,956 18,666
Days in crop Year-round 93 92 92 95
Aeration rate (hp/ha) 0 9 33 35 49
Yield (kg/ha/crop) 196 1301 3483 6982 13,560
Feed conversion ratio n.a. 1.29 1.38 1.57 1.38
Harvest weight (shrimp/kg) Not reported 70 66 64 57
Survival (%) 20 75 67 73 80
Crops per year 1 2 2 2 2
n.a. = not applicable (no feed was used by farms in this cluster); PL = postlarvae.

kg/ha/crop for L. vannamei). Yields of the other per year showed similar results in that higher
intensity levels in Thailand were more similar to intensity of P. monodon production in Vietnam
those in Vietnam. In Vietnam, the mean yield was associated with a greater number of crops
values for the two species raised within each raised in a year, whereas most of the clusters
intensity level were similar, while in Thailand, associated with production of L. vannamei in
the only observations of P. monodon produc- both Vietnam and Thailand produced two crops
tion were at a very extensive level, with very a year. Farms in the most extensive management
low yields. Feeding rate differences showed sim- clusters reported significantly lower survival
ilar patterns, with the exception that farms in the rates (20–40%) as compared with farms in
low-intensity production cluster of P. monodon the medium to very high intensity clusters
in Thailand used no feed. (68–83%).
Stocking density generally increased with
increasing levels of production intensity
(Tables 3 and 4). Higher intensity levels of P. Economic Performance across Levels
monodon production in Vietnam were associated of Intensification
with a shorter production cycle, from 90 d/crop Gross receipts per hectare per year increased
at the higher intensity levels to 143 d/crop for with greater production intensity, across all clus-
the lower intensity levels. However, there was no ters, as would be expected (Tables 5 and 6). The
clear trend in the days required to produce a crop greater yields produced at greater levels of inten-
for L. vannamei production in either Vietnam sity resulted in greater volumes of shrimp sold
or Thailand. The number of crops produced that resulted in greater gross receipts.
 17497345, 2017, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jwas.12423 by Yemen Hinari NPL, Wiley Online Library on [28/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
232 ENGLE ET AL.

Table 5. Key economic results of enterprise budget analyses, Vietnam clusters.a

Item Penaeus monodon Litopenaeus vannamei

Low Medium High Very high Low Medium High Very high
Gross receipts $5601 $23,718 $99,654 $191,340 $3435 $44,969 $90,405 $151,695
Feed cost $1572 $8181 $31,809 $59,510 $3502 $12,805 $25,615 43,521
PL cost $467 $894 $4511 $6883 $2162 $2577 $5487 $6069
Amendment cost $992 $2261 $5248 $5248 $986 $4106 $21,556 $48,033
Sediment removal cost $334 $419 $1110 $2628 $283 $583 $665 $718
Energy (total electricity and fuel) $339 $2667 $3422 $3422 $526 $924 $2182 $2336
Labor 0 0 $2964 $24,080 0 0 $5186 $7588
TVC $4550 $16,412 $55,239 $113,371 $8372 23,905 67,676 $120,102
Income above variable costs $1051 $7306 $44,415 $77,970 −$4937 $21,065 $22,729 $31,594
TFC $4619 $4645 $4967 $4981 $4508 $4690 $5100 $5972
Total cost $9169 $21,057 $60,206 $118,352 $12,879 $28,594 $72,777 $126,073
Net returns −$3568 $2662 $39,446 $72,988 −$9444 $16,375 $17,629 $25,622
BEP above VC $5.08 $4.33 $3.47 $3.34 $15.80 $3.45 $4.85 $5.13
BEP above TC $10.24 $5.56 $3.78 $3.48 $24.30 $4.12 $5.22 $5.39
BEY above VC 727 2623 4413 6710 646 1844 5221 9265
BEY above TC 1465 3365 4810 7004 994 2206 5614 9725
BEP above VC = breakeven price above variable costs; BEP above TC = breakeven price above total costs; BEY above
VC = breakeven yield above variable costs; BEY above TC = breakeven yield above total costs; PL = postlarvae; TFC = Total
fixed costs; TVC = total variable costs.
a Values calculated in US$/ha/yr.

Variable costs, by definition, are costs that cost of equity capital invested in the farm that
increase with greater levels of production. would have earned some amount of interest if
Tables 5 and 6 show that individual variable invested differently.
costs per hectare per year, that is, the cost of Costs per metric ton of shrimp produced
postlarvae, feed, energy use (electricity and decreased from the lowest to the highest inten-
fuel), amendments, labor, and total variable sity levels for each species in each country. For
costs per pond, generally increased with the P. monodon production in Vietnam, the cost
intensity of production in both countries and for per metric ton at the lowest intensity level was
both species produced. US$10,245 and decreased to US$3484 at the
Total fixed costs per hectare per year also highest intensity level. For L. vannamei produc-
increased generally with the level of produc- tion in Vietnam, costs per metric ton decreased
tion intensity for both countries and species. from US$24,301 at the lowest intensity level
The greater total fixed costs on farms that pro- to US$5387 at the highest intensity level. In
duce at greater levels result primarily from addi- Thailand, costs per metric ton for L. vannamei
tional investment in equipment such as aera- decreased from US$8184 at the lowest intensity
tion systems and various types of vehicles. This level to $3817 at the highest intensity level.
additional investment enters into annual costs Net returns per hectare per year for the Viet-
and returns in the form of annual depreciation namese clusters showed increasing overall prof-
and interest on the investment. Annual depreci- its with increased levels of production intensity,
ation accounts for the need to provide an annual with the lowest intensity level of L. vannamei
accounting for the capital that will be needed showing an overall loss (Fig. 1). In Thailand,
to replace equipment when worn out. Charging all farms showed negative net returns above all
interest on the investment in the annual costs and costs (Fig. 2). However, as the level of intensity
returns table standardizes either the interest paid of production increased, the magnitude of losses
if capital were borrowed to finance the equip- decreased for L. vannamei production. Income
ment purchase or to account for the opportunity above variable costs was positive, indicating
 17497345, 2017, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jwas.12423 by Yemen Hinari NPL, Wiley Online Library on [28/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
ECONOMICS OF SUSTAINABLE INTENSIFICATION IN SHRIMP 233

Table 6. Key economic results of enterprise budget analyses, Thailand clusters.a

Item Penaeus monodon Litopenaeus vannamei

Lowb Lowc Medium High Very high
Gross receipts $653 $8673 $23,220 $46,547 $90,400
Feed cost 0 $3736 $10,660 $24,347 $41,480
PL cost $28 $4167 $4167 $5467 $6600
Amendment cost 0 $1389 $992 $1924 $5301
Sediment removal cost 0 0 $345 $476 $659
Energy (total electricity and fuel) $30 $2315 $5343 $5850 $6384
Labor 0 0 $313 $1639 $7763
TVC $2258 $13,668 $25,734 $45,839 $83,310
Income above variable costs −$1605 −$4994 −$2514 $707 $7090
TFC $1104 $11,895 $17,117 $17,856 $20,198
Total cost $3362 $25,563 $42,850 $63,694 $103,508
Net returns −$2709 −$16,889 −$19,630 −$17,149 −$13,108
BEP above VC $11.52 $5.25 $3.69 $3.28 $3.08
BEP above TC $17.15 $9.83 $6.14 $4.56 $3.81
BEY above VC 677 2050 5201 6876 12,496
BEY above TC 1009 3834 8660 9554 15,526
BEP above VC = breakeven price above variable costs; BEP above TC = breakeven price above total costs; BEY above
VC = breakeven yield above variable costs; BEY above TC = breakeven yield above total costs; PL = postlarvae; TFC = Total
fixed costs; TVC = total variable costs.
a Values are US$/ha/yr.
b The very extensive P. monodon cluster had an average pond size of 9.9 ha, much larger than all other clusters.
c The low intensity of L. vannamei cluster had significantly smaller ponds, 0.4 ha as compared with 0.67 ha ponds in the

other L. vannamei clusters.

$80,000 $0

$60,000 ($5,000)

$40,000 ($10,000)

$20,000 ($15,000)

$0 ($20,000)

($20,000) ($25,000)
P. monodon L. vannamei P. monodon L. vannamei
Low Medium High Very high
Low Medium High Very high
Figure 1. Net returns (US$/ha/yr), Vietnam.
Figure 2. Net returns (US$/ha/yr), Thailand.

short-run profitability, for the very high and high

intensity clusters only. importance of annual fixed costs demonstrates a
Within each country and species type, the more efficient use of the fixed resources of land
relative importance of the fixed costs associ- and investment capital with a greater intensity
ated with the investment in land, infrastructure, of production. The greater importance of annual
and equipment decreased as the intensity level fixed costs at the very high yield level in Thai-
increased. In Vietnam, the contribution of annual land may explain the lack of long-term prof-
fixed costs to total costs decreased from 50 to itability demonstrated in the budget analysis in
4% as intensity increased for P. monodon and Thailand as compared with Vietnam.
from 35 to 5% for L. vannamei. In Thailand, Table 7 lists the various types of amend-
the contribution of annual fixed costs to total ments reported on the survey questionnaires
costs decreased from 46 to 20% as production and included more than 90 different types of
intensity increased. This decrease in the relative lime, minerals, antibiotics, probiotics, vitamins,
 17497345, 2017, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jwas.12423 by Yemen Hinari NPL, Wiley Online Library on [28/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
234 ENGLE ET AL.

Table 7. Amendments used on shrimp farms in Vietnam Table 7. continued

and Thailand, as reported by respondents.
Name
Category Name reported Vietnam Thailand Category reported Vietnam Thailand
Lime Calcium carbonate X Fertilizer Brown sugar X
Calcium chloride X DAP X
Calcium dioxide X DAB X
Calcium X Fertilizer X X
hydroxide Molasses X
Calcium oxide X X Phosphorus X
Calcium sulfate X Red sugar X
Daimetin X Yuca X
Dolomite X X Insecticides/ Derris grass X
Lime X X pesticides Saponin X X
Water Biopond X Trichlorfon X
amendments Biowaste X Disease treatment Copper sulfate X X
Hydrogen X Formaldehyde X
peroxide Gluteraldyde X
Magnesium X Metabisulfite X
chloride Potassium X X
Magnesium X permanganate
sulfate Antibiotics Antibiotics X
Potassium X Amoxicillin X
chloride Ciprofloxacin X
K2 NO4 X Enrofloxacin X
pH Buser X Oxytetracycline X
Sodium X Probiotics Aquamax X
bicarbonate BST X
Sodium sulfate X BZT X
Tapondpro X EM X
Zeolite X X Navet Biozyme X
Minerals ADP X Probiotics X X
Azomite X Super Biotic X
CAPHOT X Super Fixer X
Himineral X Super Murras X
HP9 X Super PH Brand X
Mineral A X Super PS X
Mineral CAMID X Vitamins Biosortol X
Minomix X Vitamin C X
Other minerals X X Vitamins X
Pondmin X Other Chitosan X
Remix X Metabisulfite X
Disinfectants Aquadine X Other X
Benzalkonium X X Unknown ABS X
chloride (BKC) BBM X
Biodin X Hitech X
Biodine X X Nasbaq X
Chlorine X X NTS/NT5 X
Chlorite X Organic Gold X
Clear 80 X Osanet Shell X
Dimetyle X Sun Terex X
Dine 9000 X Super Info X
Hidine X Truong Hai X
Tea seed X Hepatic
Vikon X TCK X
 17497345, 2017, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jwas.12423 by Yemen Hinari NPL, Wiley Online Library on [28/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
ECONOMICS OF SUSTAINABLE INTENSIFICATION IN SHRIMP 235

fertilizers, disinfectants, and compounds used variables (such as feeding, aeration, and har-
for sterilization. Given the substantial effect of vesting strategies) constant. However, farms are
the costs of these amendments, particularly in managed based on day-to-day decisions that can
Vietnam, shrimp production costs were esti- be affected by cash flow, incidence of disease,
mated with and without the cost of amendments. influence from peers, and other reasons that
Depending on the specific cluster, the cost per result in categories, in this case of stocking den-
metric ton of shrimp produced was reduced by sities, that do not represent orderly increments,
$170/m.t. to $2262/m.t. in Vietnam and from but yet resulted in statistically distinct levels of
$0/m.t. to $435/m.t. in Thailand by eliminat- shrimp yields.
ing the use of amendments. While some amend- This study found that categorizing levels of
ments, such as lime or minerals, are necessary intensification by initial stocking density may
for good survival and growth of shrimp, daily use obscure the effects of intensification on cost effi-
of antibiotics in shrimp feed, some of which are ciencies and profitability. The extensive litera-
banned for use in livestock feeds in the USA, EU, ture on the economics of aquaculture shows that
Japan, and other countries, or other amendments yields often have the greatest effect on per-unit
for which there is little clear evidence of effects costs of production. Moreover, yields are not
on survival and growth, represent opportunities solely a function of stocking density, but result
to reduce overall costs of producing shrimp. from the combined and interrelated effects of
stocking density, aeration levels, and feeding
rates, among others. Johnson et al. (2014) sim-
Discussion
ilarly found that cost per unit of production, and
Farm-level data are inherently variable hence profitability, can be quite different even
because no two farmers have the same operating with similar stocking densities. In this study, in
protocol, access to capital, work force, and Vietnam, the mean stocking density of P. mon-
management skill and experience. Attempts by odon in the high-yield cluster (58/m2 ) was simi-
researchers, policy makers, or environmentalists lar to that in the very-high-yield cluster (59/m2 ).
to categorize a particular race or culture of However, the yield in the very-high-yield clus-
aquaculture producers are flawed because of this ter was 42% greater than in the high-yield clus-
variability. Economic outcomes of an individual ter, due primarily to a 94% greater feeding rate
farm reflect a set of choices (related to stock- across three crops per year as opposed to two
ing density, aeration rates, and feeding rates) crops per year in the high-yield cluster. In Thai-
made by managers. Identifying management land, the mean stocking densities were the same
clusters reduced this variability and allowed for (62.5/m2 ) between the low- and medium-yield
identification of patterns and trends related to clusters for L. vannamei production. However,
the economics across management clusters of the yield in the medium cluster was 2.7 times
different intensity levels. The cluster analysis greater than that in the low density due primarily
clearly showed that the overall level of produc- to a feeding rate that was 2.9 times greater and an
tion intensity, as defined by the differing yields aeration rate that was 3.7 times greater than that
that resulted from the total set of management of the low-yield cluster. Thus, this study provides
practices, was a key distinguishing factor in evidence from shrimp farms that supports those
terms of which farms were more similar to each of Johnson et al. (2014) for catfish farms on the
other and, thus, to the determination of clusters. importance of use of multivariate tools such as
Results of this analysis also point to the need cluster analysis to identify similar sets of man-
for data obtained from commercial farms rather agement practices as the basis for comparative
than strictly from experimental research. An economic analyses.
experimental study to assess the effects of inten- Farm size was not significant in the principal
sity of production would likely have established component analysis, supporting the assertion by
stocking densities that varied based on some Vandergeest et al. (1999) that farm size does
arithmetic increment, while holding all other not necessarily dictate the level of production
 17497345, 2017, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jwas.12423 by Yemen Hinari NPL, Wiley Online Library on [28/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
236 ENGLE ET AL.

intensity. For example, farms managed at the depreciation” by not replacing equipment and
very high intensity level were of sizes that ranged production facilities in a timely manner and not
from 1 to 18 ha in Vietnam and 2 to 28 ha in accounting for the noncash costs of depreciation.
Thailand. The more extensively managed farms Additional, in-depth research is needed to con-
in Thailand averaged 9.9 ha, larger than the 7-ha clusively determine the root causes of the lack of
average size of more intensively managed farms. long-run profitability of shrimp farming in Thai-
Thus, it may be more useful to discuss shrimp land and develop guidance for shrimp farmers.
production intensity in Thailand and Vietnam as Although more intensive production of shrimp
a factor that is separate from farm size. was shown in this study to be more profitable,
This study showed that increasing intensifica- the ability of a farmer to shift to more inten-
tion of shrimp production resulted in increased sive production practices depends on the farm’s
profits in the short run in both countries. Such access to sufficient capital to provide electrical
a finding is generally consistent with economic service to the farm and to purchase the aeration
analyses of aquaculture in that greater yields equipment, postlarvae, and feed needed. It also
spread fixed costs over greater amounts of pro- depends on the farm manager having sufficient
duction (Engle 2010). In an earlier study of the experience, knowledge, and management skill
profitability of extensive shrimp farming in Viet- to successfully operate a more intensively man-
nam, profitability increased as yields increased aged farm. Access to capital (whether through
with use of higher-quality inputs (Brennan et al. formal channels such as banks, or informal net-
2000). In a recent study of intensification of cat- works of friends and family), the level of expe-
fish production through investment in split-pond rience and knowledge, and location differences
systems, Kumar et al. (2016) showed improved that affect the availability of water control struc-
profitability as per-unit costs of production tures in some locales and not others have been
were reduced. Revenues from the greater yields shown to restrict a farm’s ability to take advan-
exceeded the additional annual fixed costs of the tage of more profitable shrimp production sys-
increased investment. tems (Lebel et al. 2010; Marks 2010).
The lack of long-term profitability of shrimp The excessive use of amendments to ponds
farming in Thailand is, of course, affected by identified in this study, particularly in Vietnam,
early mortality syndrome (EMS), but also may has been reported previously by other authors
reflect increasing opportunity costs of land and (Gräslund and Bengtsson 2001; Boyd 2002).
capital for uses other than shrimp farming. Land While some types of amendments, such as lime
values and construction costs for production or minerals, are no doubt necessary given the
facilities have increased in recent years and water quality in particular farms, in other cases,
increased fixed costs. Greater fixed costs, com- substances such as antibiotics banned for use in
bined with losses due to EMS, may partially many countries and other substances that could
explain the lack of long-run profitability and, not be identified specifically were purchased and
hence, economic sustainability of shrimp farm- used by Vietnamese farmers. There was insuffi-
ing in Thailand. Additional work is needed to cient detail in the data set to separate the eco-
determine conclusively whether the lack of prof- nomic effects of various types of amendments
itability can be attributed to macroeconomic used. Moreover, the use of banned antibiotics
factors that have resulted in increased land in aquaculture production and their indiscrimi-
and capital costs. An alternative explanation nate use for other livestock has the potential to
may be that shrimp farming in Thailand has increase bacterial resistance to those antibiotics
become overcapitalized. For shrimp farming to that are important in human medicine (Rahman
be sustained in Thailand over the long term, et al. 2016). This study showed strong economic
the farms must generate sufficient revenue to incentives to reduce shrimp production costs in
be able to replace equipment and production Vietnam by restricting pond amendments only
facilities as these wear out. It is not uncom- to those that have been proven to have expected
mon for well-established industries to “live off results.
 17497345, 2017, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jwas.12423 by Yemen Hinari NPL, Wiley Online Library on [28/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
ECONOMICS OF SUSTAINABLE INTENSIFICATION IN SHRIMP 237

A
12,000

Breakeven Price
10,000

(US$/MT)
8,000
6,000
4,000
2,000
-

Yield (kg/ha/crop)
B
25,000
Breakeven Price

20,000
(US$/MT)

15,000
10,000
5,000
-

Yield (kg/ha/crop)
C
10,000
Breakeven Price

8,000
(uS$/MT)

6,000
4,000
2,000
-

Yield (kg/ha/crop)

Figure 3. (A) Effect of increased production per hectare on production cost per metric ton, Penaeus monodon, Vietnam. (B)
Effect of increased production per hectare on production cost per metric ton, Litopenaeus vannamei, Vietnam. (C) Effect
of increased production per hectare on production cost per metric ton, L. vannamei, Thailand.

Boyd et al. (2017) found that fewer resources greater yields (metric tons per hectare). The
were used per metric ton of shrimp produced greater yields spread annual fixed costs across
on more intensive farms. The economic analy- greater production volumes and resulted in a
sis in the present study supported this finding decreased cost per metric ton of shrimp pro-
in that production cost per metric ton of shrimp duced with increased intensification. Thus,
produced decreased as production intensity (and within the range of yields and production sys-
yield) increased (Fig. 3A, B, C). Chatvijitkul tems included in this study, the more intensive
et al. (2017) showed that achieving lower FCRs production systems were more economically
reduced costs as well as environmental impacts sustainable. Costs per metric ton of shrimp
associated with fish and shrimp production. produced decreased from the lowest to the
Thus, there is a clear value proposition for farm-
highest intensity levels for each species in each
ers to become more efficient in their use of natu-
country (US$10,245/m.t. at lowest intensity
ral resources and other inputs to produce shrimp.
to US$3484/m.t. at highest intensity for P.
monodon and from US$24,301/m.t. [lowest
Conclusions intensity level] to US$5387/m.t. [highest inten-
Economic outcomes improved with increased sity level] for L. vannamei in Vietnam and
intensification of production that resulted in from US$8184/m.t. [lowest intensity level]
 17497345, 2017, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jwas.12423 by Yemen Hinari NPL, Wiley Online Library on [28/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
238 ENGLE ET AL.

to $3817/m.t. [highest intensity level] for L. and its outcomes. Development Policy Review
vannamei in Thailand). 29(4):459–484.
Boyd, C. 2002. Chemical and biological amendments used
In Vietnam, the least intensive production sys- in shrimp farming. Report prepared under the World
tem was not profitable for either P. monodon Bank, NACA, WWF, and FAO Consortium Program on
or L. vannamei production. In Thailand, none Shrimp Farming and the Environment (work in progress
of the clusters analyzed showed long-term prof- for public discussion). The Consortium, FAO, Rome,
Italy.
its. Annual fixed costs and labor costs were
Boyd, C., A. McNevin, P. Racine, D. Paungkaew, R. Viriy-
relatively higher in Thailand than in Vietnam. atum, H. Q. Tinh, and C. R. Engle. 2017. Resource
Careful attention needs to be paid to seek the use of shrimp Litopenaeus vannamei and Penaeus mon-
most efficient ways to manage farms to improve odon production in Thailand and Vietnam. Journal of
long-term profitability in Thailand. Amendment the World Aquaculture Society 48(2):201–226.
Brennan, D., H. Clayton, and T. T. Be. 2000. Eco-
costs were particularly high in Vietnam and nomic characteristics of extensive shrimp farms in the
included substances banned for use in aqua- Mekong Delta. Aquaculture Economics and Manage-
culture products. Moreover, in both Thailand ment 4(3/4):127–139.
and Vietnam, numerous other amendments not Bush, S. R., P. A. M. van Zwieten, L. Visser, H. Van Dijk,
shown to be effective were used and increased R. Bosma, W. F. De Boer, and M. Verdegem. 2010.
Scenarios for resilient shrimp aquaculture in tropical
production costs. coastal areas. Ecology and Society 15(2):15. http://
This analysis shows the importance of identi- www.ecologyandsociety.org/vol15/iss2/art15/.
fying groups of farms based on similar sets of Chatvijitkul, S., C. E. Boyd, D. A. Davis, and A. A.
management practices identified through mul- McNevin. 2017. Embodied resources in shrimp and
fish feeds. Journal of the World Aquaculture Society
tivariate analyses rather than on a single pro-
48(1):7–19.
duction parameter such as stocking density. Not Dey, M. M., F. J. Paraguas, N. Srichantuk, Y. Xinhua, R.
accounting for important differences in farm Bhatta, and C. D. L. Thi. 2005. Technical efficiency
management strategies other than stocking den- of freshwater pond polyculture production in selected
sity can obscure important differences in eco- Asian countries: estimation and implication. Aquacul-
ture Economics and Management 9(1):39–63.
nomic outcomes.
Engle, C. R. 2010. Aquaculture economics and financing:
This analysis has shown that more intensive management and analysis. Wiley-Blackwell, Ames,
shrimp production, using fewer resources per Iowa, USA.
metric ton of shrimp produced, was also more Engle, C. R. and L. D’Abramo. 2016. JWAS: showcasing
economically sustainable than more extensive research focusing on sustainability of aquaculture enter-
prises and global food security. Journal of the World
practices. On-farm research that incorporates Aquaculture Society 47(3):311–313.
economic effects into studies related to the sus- FAO (Food and Agriculture Organization of the United
tainability and environmental management of Nations. 2016. Sustainable intensification of aquacul-
aquaculture is needed to continue to provide use- ture in the Asia-Pacific region. Documentation of suc-
cessful practices. W. Miao and K. K. Lal, editors. FAO,
ful guidance to farmers and policy makers.
Bangkok, Thailand.
Goode, T., M. Hammig, and D. Brune. 2002. Profitability
Acknowledgments comparison of the partitioned aquaculture system with
traditional catfish farms. Aquaculture Economics and
Funding for this study was generously pro- Management 6(1/2):19–38.
vided by The Gordon and Betty Moore Founda- Gräslund, S. and B.-E. Bengtsson. 2001. Chemicals and
tion (GBMF 4455). We thank the farmers who biological products used in south-east Asian shrimp
farming, and their potential impact on the environ-
participated in the survey and provided data used
ment: a review. The Science of the Total Environment
in the analysis. The views expressed are those of 280:93–131.
the authors and not necessarily those of the fund- Hair, J. F. 1995. Multivariate data analysis with readings, 4th
ing agency. edition. Prentice-Hall, New York, New York, USA.
Iliyasu, A., Z. A. Mohamed, M. M. Ismail, A. M. Abdul-
lah, S. M. Kamarudin, and H. Mazuki. 2014. A
Literature Cited review of production frontier research in aquaculture
Belton, B. and D. C. Little. 2011. Imminent and inter- (2001-2011). Aquaculture Economics and Management
ventionist inland Asian aquaculture development 18:221–247.
 17497345, 2017, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jwas.12423 by Yemen Hinari NPL, Wiley Online Library on [28/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
ECONOMICS OF SUSTAINABLE INTENSIFICATION IN SHRIMP 239

Joffre, O. M. and R. H. Bosma. 2009. Typology of New, M. B. 1996. Responsible use of aquaculture feeds.
shrimp farming in Bac Lieu Province, Mekong Delta, Aquaculture Asia 1:3–12.
using multivariate statistics. Agriculture, Ecosystems Patmasiriwat, D. 1997. Environmentally sensitive sector: a
and Environment 132:153–159. case study of shrimp aquaculture in Thailand based on
Johnson, R. A. and D. W. Wichern. 2007. Applied mul- farm survey. Thailand Development Research Institute
tivariate statistical analysis, 6th edition. Prentice Hall, Draft Paper, Bangkok, Thailand.
New York, New York, USA. Pielou, E. C. 1984. The interpretation of ecological data.
Johnson, K., C. Engle, and B. Wagner. 2014. Compar- John Wiley & Sons, New York, New York, USA.
ative economics of U.S. catfish production strategies: Prein, M., G. Hulata, and D. Pauly. 1993. On the use of
evidence from a cross-sectional survey. Journal of the multivariate statistical methods in aquaculture research.
World Aquaculture Society 45(3):279–289. Pages 1–12 in M. Prein, G. Hulata, and D. Pauly,
Kumar, G. and C. Engle. 2017. Economics of intensively editors. Multivariate methods in aquaculture research:
aerated catfish ponds. Journal of the World Aquaculture case studies of tilapias in experimental and commercial
Society 48(2):320–332. systems. ICLARM, Manila, Philippines.
Kumar, G., C. Engle, and C. S. Tucker. 2016. Costs and Rahman, K. M. M., A. A. M. Hatha, A. D. G. Selvam, and
risk of catfish split-pond systems. Journal of the World A. P. Thomas. 2016. Relative prevalence of antibiotic
Aquaculture Society 47(3):327–340. resistance among heterotrophic bacteria from natural
Lebel, L., R. Mungkung, S. H. Gheewala, and P. Level. and culture environments of freshwater prawn, Macro-
2010. Innovation cycles, niches and sustainability in the brachium rosenbergii. Journal of the World Aquaculture
shrimp aquaculture industry in Thailand. Environmen- Society 47(4):470–480.
tal Science & Policy 13:291–302. Tan, R. I., Y. T. Garcia, M. L. Dator, I. M. A. Tan, and
Little, D. C., B. K. Barman, B. Belton, M. C. Bev- D. E. Permsl. 2011. Technical efficiency of genetically
eridge, S. J. Bush, L. Dabbadie, H. Demaine, P. improved farmed tilapia (GIFT) cage culture operations
Edwards, M. M. Haque, G. Kibria, F. Morales, F. in the lakes of Laguna and Batangas, Philippines.
J. Murray, W. A. Leschen, M. C. Nandeesha, and Journal of the International Society for Southeast Asian
F. Sukadi. 2012. Alleviating poverty through aquacul- Agricultural Sciences 17(1):194–207.
ture: progress, opportunities and improvements, Pages Thongrak, S. 1995. Determinants of technical efficiency in
719–783 in R. P. Subasinghe, J. R. Arthur, D. M. intensive shrimp farming. Songklanakarin Journal of
Bartley, S. S. De Silva, M. Halwart, N. Hishamunda, Science and Technology 17:81–87.
C. V. Mohan, and P. Sorgeloos, editors. Farming the Vandergeest, P., M. Flaherty, and P. Miller. 1999. A
waters for people and food. Proceedings of the Global political ecology of shrimp aquaculture in Thailand.
Conference on Aquaculture 2010, September 22–25, Rural Sociology 64(4):573–596.
2010, Phuket, Thailand. FAO, Rome, Italy, and NACA, Waite, R., M. Beveridge, R. Brummett, S. Casstine, N.
Bangkok, Thailand. Chaiyawannakarn, S. Kaushik, R. Mungkung, S.
MacQueen, J. 1967. Some methods for classification and Nawapakpilai, and M. Phillips. 2014. Improving pro-
analysis of multivariate observations. Proceedings of ductivity and environmental performance of aquacul-
the Fifth Berkeley Symposium on Mathematical Statis- ture. Working Paper. Installment 5 of Creating a Sus-
tics and Probability, volume 1: Statistics, 281–297, tainable Food Future. World Resources Institute, Wash-
University of California Press, Berkeley, California, ington, District of Columbia, USA. Accessed at http://
USA. www.worldresourcesreport.org.
Marks, B. 2010. Small fry in a big ocean: change, resilience World Bank and MOFI. 2006. Guidelines for environmen-
and crisis in the shrimp industry of the Mekong Delta tal management of aquaculture investment in Vietnam.
of Viet Nam. Electronic dissertation. University of Technical Note. World Bank, Washington, District of
Arizona, Tucson, Arizona, USA. Accessed Jun 13, 2016 Columbia, USA.
at http://hdl.handle.net/10150/193955.
Nakamura, R. 1985. Aquaculture development in India: a
model. BioScience 35(2):96–100.