Aquacultural Engineering 78 (2017) 146–154

Contents lists available at ScienceDirect

Aquacultural Engineering
journal homepage: www.elsevier.com/locate/aque

The financial feasibility of producing fish and vegetables through MARK

aquaponics
⁎
Roel H. Bosmaa, , Lysette Lacambrab,1, Ynze Landstrac,2, Chiara Perinib,3, Joline Poulied,4,
Marie J. Schwanerd,5, Yi Yinb,6
a
Aquaculture & Fisheries group, Animal Science Group, Wageningen University & Research (WUR), Post Office Box 338, 6700AH Wageningen, The Netherlands
b
Plant Sciences, Wageningen University & Research, The Netherlands
c
Environmental Economics & Natural Resources, Wageningen University & Research, The Netherlands
d
Animal Sciences, Wageningen University & Research, The Netherlands

A R T I C L E I N F O A B S T R A C T

Keywords: Aquaponics, producing fish and vegetables in a closed-loop water system, reduces fertilizer use and water dis-
Aquaponics charge, and is therefore promoted as a sustainable venture. A recent global study found that the majority of 257
Cost-Benefit-Analysis surveyed aquaponics farms made losses, but the reasons have been poorly analyzed. Based upon grey literature
Catfish and a post-hoc cost-benefit analysis for an investment in an aquaponics farm in the Philippines, this paper aims
Jade perch
to assess the factors contributing to an appropriate level of returns. In the Philippines, vegetables are relatively
Vegetables
well-priced, but if catfish were produced, a venture producing 1250 kg fish, 6000 kg lettuce, and 300 kg tomato
per year would have a Net-Benefit-Cost Ratio of below 1.3 after 20 years. This ratio should be higher when one
considers the risks against which a farmer cannot and should not insure its production, as argued in the paper.
Depending on the species of both fish and vegetables, the quantity of nutrients coming from the former com-
ponent imposes a fish volume: vegetable area ratio ranging from 1:30 to 1:100, thus the quantity of marketable
fresh vegetables determines the size of an aquaponics enterprise. As a consequence, the investments in the fish
component are relatively high and weigh heavily on the financial balance. For producers to successfully adopt
aquaponics, they need to consider starting first with catfish, and then, as they get to master the system’s
management, they can shift to a high-value fish species for niche markets.

1. Introduction
Aquaponics combines aquaculture and hydroponic plant production
in a closed-loop water system that simultaneously grows both plants
and fish. In a symbiotic relationship, the waste that is produced by fish
in water tanks, which could become toxic for the fish if not cleaned, is
used directly or converted by bacteria into useful nutrients for plants.
As plants assimilate these nutrients, the water becomes clean and is
recycled back to the fish tank (Liang and Chien, 2013; Chaves et al.,
1999). Aquaponics allows the sustainable growth of two crops, while
reducing the need of chemical fertilizer for plant and the discharged
water. Aquaponics finds its roots in ancient mixed systems, e.g., the
rice-fish and the poultry-fish systems of Asian countries (FAO, 2001).
Currently, up-scaling and out-scaling of aquaponics are promoted in
both developed and developing countries, with support from science
(INAPRO, 2014; Karimanzira et al., 2016; Wongkiewa et al., 2017) and
guidelines from the FAO (2014). Recently, Love et al. (2015), after
assessing 257 aquaponics enterprises worldwide, concluded that the
majority of the small aquaponics enterprises incurred losses, and Engle
(2015), based upon an analysis of articles, reports and presentations,
listed possible challenges and economic issues.
For its double purpose, aquaponics might be a sustainable solution
to the low provision of fish and vegetables in some countries. Though
the Philippines has a long coast and a vast marine zone (Fig. 1), the

⁎
Corresponding author.
E-mail addresses: roel.bosma@wur.nl, roel.bosma@xs4all.nl (R.H. Bosma), lysette_lacambra@yahoo.com (L. Lacambra), ynzelandstra@hotmail.com (Y. Landstra),
chiaram.perini@gmail.com (C. Perini), jolinepoulie@hotmail.com (J. Poulie), janneke.schwaner@gmail.com (M.J. Schwaner), yin.yi.cg@outlook.com (Y. Yin).
1
East-West Seed Knowledge Transfer, Chiang Mai, 50290 Thailand.
2
Friesland-Campina, Bronland 20, 6708 WH Wageningen, The Netherlands; Vinkenburgstraat 160, 3512 AB Utrecht, The Netherlands.
3
Agroservice s.r.l., Cormano, Italy.
4
Van der Waalsstraat 16 2, 6706 JN Wageningen, The Netherlands.
5
Department of Biological Sciences, University of Idaho, Moscow, ID, USA.
6
Rixin Road 1533, Kunming, Yunnan 650000, China.

http://dx.doi.org/10.1016/j.aquaeng.2017.07.002
Received 28 April 2017; Received in revised form 4 July 2017; Accepted 5 July 2017
Available online 08 July 2017
0144-8609/ © 2017 Published by Elsevier B.V.
R.H. Bosma et al.
country imports fish; and over the past 15 years, the importation of
vegetables and fruits is increasing (Cororaton and Corong, 2009). Im-
porting both fish and vegetables has contributed to the country’s trade
deficits, which can be attributed to the increasing demand relative to
the domestic supply (Cororaton and Corong, 2009). Next to the increase
in sustainable fish and vegetable production, aquaponics can also re-
duce unemployment and trade deficit. Most of the aquaponics ventures
in the Philippines are small scale producing only for household, family
or community (Aquaponics Philippines, 2013). A handful of the ven-
tures produce for the market; one of them is near Cebu, and the others
in the vicinity of metro Manila (Aquaponics Philippines, 2014).
In this paper, we used a cost-benefit analysis (CBA) to assess the
financial feasibility of an investment in a small-scale aquaponics system
in the province of Negros Oriental, Philippines. Using this case study, in
addition to refereed papers and grey literature, we present generic re-
commendations for up- and out-scaling of aquaponics.
2. Methods
The system design was based on a nutrient balance of a two-system
147
Aquacultural Engineering 78 (2017) 146–154
Fig. 1. Map of the Philippines showing the mentioned cities.
Legend: The longitudinals are 120° and 125° E (Source: http://www.
mapsof.net/)
component. We collected data through survey from experts and from
literature review. To assess the financial feasibility using a CBA, we
calculated the Net Present Value (NPV) and the discounted benefit-cost
rate (DBCR). Both the nutrient balance and the CBA with the related
sensitivity analysis were calculated in Excel®.
2.1. The data collection
Detailed information about the aquaponics system already present
in a similar environment was acquired through an e-mail survey of four
commercially oriented farms, of whom all responded, in the
Philippines. Questions reflected in the survey included information on
farm size, fish and vegetable combination, sales per year, type of fish
tanks and greenhouse, electricity and water supply, storage and
packaging, transportation, staff, training, building materials, cost of
investment and operation, and record-keeping.
The market information on Dumaguete, the capital of Negros
Oriental, and Cebu, an international tourist destination, was collected
from two local experts by e-mail and phone. The potential markets at
Dumaguete city were located about 9 km away from the planned farm
R.H. Bosma et al. Aquacultural Engineering 78 (2017) 146–154

Table 1 Table 3
Amount of feed and FCR of catfish per fish size. Stocking density per size class up to 500 gr.

Fish age (days) Fish size (gram) Feed (gram/day) FCR* Size class Kg per Days Weeks No. of No. Kg fish m3 per m3 per
class tanks fish per fish tank
6 23.0 7.26 0.64 tank
13 36.9 6.71 0.67
20 54.1 6.31 0.70 10–100 80 35 5 5 56 5.6 0.070 0.4
27 74.4 5.84 0.73 100–250 150 32 5 5 53 13.3 0.088 0.5
34 97.9 5.37 0.75 250–500 200 37 5 5 50 25.0 0.125 0.6
41 124.5 4.92 0.78
48 154.3 4.46 0.80
55 187.1 4.02 0.81
Table 4
62 223.0 3.58 0.83
Equations used to calculate the N-outflow of the fish component.
69 261.9 3.15 0.84
76 303.9 2.72 0.86
Step Purpose of calculation Unit Equation
83 348.8 2.31 0.87
90 396.8 1.91 0.89 1 Fecal loss (FL) g kg -1
FLx = Fx.feed * (1-Dx)
97 447.8 1.52 0.90 feed
104 501.8 1.14 0.91 2 Fish gain (G); g kg-1 Gx,gain = ((Wf * Cx,f/100) – (Wi * Cx,i
feed /100))/(Wf -W,i)*(1000g feed/FCR)
3 Non-fecal loss (NFL) g kg-1 NFLx= Gx.feed *(1 - Dx) - Gx,gain - FLx
site. The docking port to the ferry for Cebu in Sibulan was a 30-min-car feed
drive from the farm site. The boat trip was less than one hour, but the 4 DM body composition %; W DM = 17 * W0.088
in g
arrival port was a two-hour-car drive from Cebu city. Complementary
5 Nitrogen content of fish %; W Protein = 13 * W0.059
data were collected from literature and local experts. However, we in g
made assumptions for some data (see below). 6 O2 consumption / g/kg O2 = 1531.02 * R-0.656
feeding level (R)
7 CO2 production / g/kg CO2 = 1937.18 * R-0.657
2.2. The nutrient balance of the components feeding level (R)

The nutrient balance focused on nitrogen because this is the most
needed nutrient for both plants and fish. The balance of other nutrients
will be discussed. First we present the methods for the fish and the plant
component.
2.2.1. The fish component
This assessment followed the first three steps in calculating the
output of a recirculating aquaculture system (RAS), as described by
Hogendoorn (1983) and Nguyen et al. (2016). These calculations were
based on the feed use efficiency, survival, and growth of the cultured
species.
The number of fish surviving was calculated by using mortality rates
for four sizes (Table 2). The produced biomass of catfish was managed
according to three sizes; the three stocking densities, 56, 53, and 50
pieces were assumed to be housed in three different water volumes
(Table 3). The tanks were assumed to have the same standard size. For
the catfish, each period was assumed to last for 5 weeks. Alternative
freshwater species that are more predatory and of higher value, such as
Jade perch (Scortum barcoo), need to be graded more frequently, but
the final density in the final weeks was assumed to be the same as that
for catfish: 25 kg/m3.
In the Philippines, feeds are available for most cultured fish species.
However, not all these feeds are based on the specific requirements of
the species; e.g., diets of Asian river catfish are based on those devel-
oped in Europe for African catfish cultured in RAS. The catfish feed for
sale in the Philippines contains 36% crude protein for the starter and
29% for the finisher pellets. In general, fish effectively use about 30%
only of ingredients present in the diet; they excrete or waste the
Table 2
Mortality rate per size class of the total amount of catfish during the year.
Size class (gr) Mortality rate (%) Number of fish in system
Start
10–100 5 3038
100–250 6 2894
250–500 5 2730
500–900 4 2600
Legend:
Cx,f = Concentration of component x in fish at Wf; Cx,i = Concentration of x in fish at Wi;
Dx = Digestibility of component x; DM = Dry Matter; Fx.feed = Content of x in the feed;
Gx,feed = Weight of component x in the feed; Gx,gain = Fish weight gain; R = feeding
level;
W = weight of the fish; Wf = Final weight fish; Wi = Initial weight fish.
Sources: 1–5: Hogendoorn (1983).; 6–7: Eding and Weerd (1999).
remaining 70%. However in RAS, the waste is generally low because
the biofilms recycle some nutrients. Consequently, the Feed Conversion
Ratio (FCR; i.e., the quantity of feed used to produce 1 kg of wet weight
fish) in RAS is generally below 1; see Table 1 for the FCR used in this
study.
The nutrient outflow of the fish component was calculated by
considering a Total Available Nitrogen (TAN) removal efficiency of
45%, leaving 55% for the plants. The equations used to calculate the N-
output of the fish component are given in Table 4.
2.2.2. The plant component
The choice of the vegetables was based on three parameters: the
market demand (see 4.1), the convenience for growing fish and vege-
tables in an aquaponics system, and the match between nutrient input
and requirements. The plants successfully being grown in aquaponics
systems include: lettuce, cucumbers, bell peppers, tomatoes and egg-
plant; with some extra care, also root crops such as carrot can be grown
(Graber and Junge, 2009; Kamal, 2006; Sajjadinia et al., 2010; Roosta,
2014). Therefore this study considers a combination of lettuce having
low and tomato having high nutrient requirements. To have a constant
nutrient demand from the vegetables in this system, we considered a
staggered crop production for both lettuce and tomato. This ensures a
continuous production of these vegetables having no seasonal climate
limitation in the Philippines.
The nitrogen demand from the plants was calculated as the sum of
the nitrogen demand of lettuce and tomato. Tomato nitrogen demand
End was calculated according to the vegetative and fruit growth (Mori et al.,
2008) considering a specific % of nitrogen concentration in the plant’s
2894 organs (Minkenberg and Ottenheim, 1990; Yoshida et al., 1997). De-
2730
mand was reported on a daily basis which considered a cycle of about
2600
2500 55 days (which also corresponded to the cycle length of the chosen
tomato variety). Nitrogen demand was assumed as constant throughout

148
R.H. Bosma et al.
the entire growth period. Similarly, nitrogen demand for lettuce was
calculated by considering the plant growth and a fixed nitrogen con-
centration in the leaf tissue (De Pinheiro Henriques and Marcelis,
2000). Also in this case, we made the assumption that the demand is
constant throughout the cycle. This assumption could be made because
a staggered production was set for both tomatoes and lettuce; this
means that at any given time, different stages of plant coexist in the
same system and the nutrient demand remains constant throughout the
year.
2.3. The cost-benefit analysis
The profitability of the project was estimated by calculating the
NPV, the discounted pay-back time and the DBCR for the project
duration. For the present case, we applied a project length of 20 years.
The discounted pay-back time is the number of years until the cumu-
lated discounted cash flow is positive. According to cost-benefit analysis
theory, the project may be undertaken when the NPV is larger than zero
after the project life span (Perkins, 1994).
The NPV is defined as the present value of the net benefit stream.
The net benefits of a period are the benefits minus the costs in that
particular period, and subsequently discounted by using an appropriate
discount rate, and then summed for the project life span (Perkins,
1994). Inflation and other price changes were not taken into account
with the assumption that costs and benefits would rise proportionally
(Boardman et al., 2011). The NPV, or the discounted net benefit, is
calculated by using the following formula:
n and water was pumped around for both, and 20% because fish would be
NPV = ∑ [(Bt − Ct )/(1 + r )t ]
t=0
in which: Bt = benefits in year t; Ct = cost in year t; n = length of
project in years; r = discount rate. The undiscounted net benefits
(UNB) are calculated without applying the quotient in above formula.
The DBCR is an indicator of the profitability of an investment −
when DBCR is lower than 1 it is not advised to start a project; DBCR is
below 1 when NPV is negative. The DBCR is defined as the ratio be-
tween the sum of discounted annual revenues and the sum of the cor-
responding costs, which is generated during a specific period
(Daneshvar and Kaleibar, 2010, cited by Testa et al., 2015); thus, cal-
culated as follows:
n n
⎡ ⎤ ⎡ ⎤ circulating aquaculture system (RAS), this species is relatively easy to
DBCR = ⎢∑ Bt /(1 + r )t ⎥/ ⎢∑ Ct /(1 + r )t ⎥;
⎣t=0 ⎦ ⎣t=0 ⎦
in which the first term are the discounted revenues (TDR); and the
second term, the sum of the discounted investments (TDIC) and op-
erational cost (TDOC), as mentioned in Table 9.
The choice of the appropriate discount rate is an issue of debate. The
higher the discount rate, the higher is the preference for money now
rather than later; and the higher the relative value of cost and benefits
today compared with that of tomorrow (Hanley and Spash, 1993). In
this project, a discount rate of 8% was used because the discount rate in
similar projects lies between 6% and 10% (Rupasinghe and Kennedy,
2006). This discount rate is in line with present interest rates applied by
the banks in the Philippines for commercial investments. Applying a
discount rate makes sure that costs and benefits in the future are given a
relatively lower value than costs and benefits today, and considers a
sense of time and the opportunity cost of capital. The opportunity cost
of capital allows a comparison of the present value of capital with other
investments, including the deposit on a bank account against an interest
that increases the value of the deposit.
The costs for investment and operation were calculated by using the
prices collected after the size of the farm was determined. The invest-
ments were depreciated by using the replacement cost of a specific
material or equipment after the expected lifespan (Table 7). In the
operational cost, the insurances were not included because the
149
Aquacultural Engineering 78 (2017) 146–154
information on amounts to pay varied too much. For taxes also the
information varied; therefore, we implemented two scenarios: without
and with taxes. The considered tax is the value added tax (VAT) which
farmers do not pay when they sell directly to consumers in the Phi-
lippines, and when their total gross sale is below PhP 1.919.500 (BIR,
2015).
To calculate the revenues, we assumed that wholesale prices, i.e.,
farm gate price, are half of the retail price. Some hotels and restaurants
might be willing to pay more. We used the farm gate price of PhP
18 kg−1 and PhP 55 kg−1 for tomatoes and the lettuce, respectively
(CountrySTAT, 2015); and farm-gate price of PhP 79 kg−1 and PhP
300 kg−1 for catfish and Jade perch, respectively.
A sensitivity analysis was done to estimate the effect of the discount
rate, the fish species, the payment of VAT on the DBCR, and the sales at
consumer prices. The discount rate of 8% was compared with the range
of 4–16%. For the shift to Jade perch, we used the price of fingerling at
PhP 12 per piece and that of feed at PhP 34 kg−1; double and 1.5 times
higher compared to that of catfish, respectively. For the sales directly to
consumers, a price increase of 50% was assumed considering also losses
(Engle, 2015), and adding 20–50% to the cost of the 1.5 full time staff
was accounted for in the baseline.
To complete the post-hoc analysis, we calculated the contribution of
the fish component to revenue, investments, and operation.
Specifically, we accounted to the fish component the following opera-
tional costs: 100% for cost of fingerlings; 50% for feed and electricity;
and 20% for transportation, repairs and labor. We used 50% because
about half of the feed (55% of TAN) was assumed to be used by the fish
sold once a week only and vegetables, 5–7 days.
3. Results
Our market survey concluded that the local consumers preferred
fish weighing about 500 g only. The elasticity for the demand for low-
priced fish such as catfish (PhP 79 kg−1) is high. The demand for the
better-priced species such as Jade perch (PhP 300 kg−1) is limited but
not satisfied, and thus the elasticity is lower. The survey found that
about 500 plants of lettuce could be marketed per week (i.e., about 70
plants per day). The demand for lettuce is the limiting factor for an
integrated production system.
This assessment includes the culture of catfish because in the re-
manage. Moreover, catfish is less sensitive to low water quality and can
turn to direct air breathing in case of low dissolved oxygen content due
to electricity failure. Once the system is operating well the enterprise
could be converted to the more sensitive Jade perch.
3.1. The system components
3.1.1. The plant component
The production of 70 plants of lettuce would demand 12.8 gN d−1
(Table 5). With the remaining N, about 4 kg of tomato can be grown per
week. Together this requires the supply of 31.6 gN d−1 (Table 6). The
production of lettuce required an area of 80m2, and that of the tomato,
an area of 96 m2. This fits the size of a small commercial greenhouse
available in the Philippines.
3.1.2. The fish component
About 30 gN d−1 output can be provided by a RAS producing 25 kg
of catfish weekly. To keep the pressure on the bio-filter equal and to
regularly provide fish to the market, we calculated a culture system
with 15 fish tanks and a total water volume of 1.4 m3. The first batch of
fingerlings will be placed in the tank at week 1 and from this moment
onward, every three weeks, a new batch of about 170 fingerlings will be
added to the system. After 15 weeks the first batch of about 50 fish
weighing 500gr will be harvested.
R.H. Bosma et al. Aquacultural Engineering 78 (2017) 146–154

Table 5
The characteristics of the plant components for the scaling of the system.

Density N leaves N fruits DM fruits DM plant Batches Cycles year-1

−1
plant-1 m2 % % % % n year 1st yr Later yr

Tomato 4 0.045 0.02 4 0 3 8.25 15.5

Lettuce 25 6.00 – – 5 – 36 52

3.2. The system profitability Table 7

Investment cost (rounded in 1000 PHP) for the different materials needed for the aqua-
The investments, i.e., the fixed costs in infrastructures for the cal- ponics farm and the life span (yr) considered for depreciation.

culated farm size include 35% labor cost for the building and installa- Material Amount Life span
tion (Table 7). For each component, a specific life span was used to
calculate the depreciation cost considered in the CBA. The cost of Greenhouse 542.5 5; 10; 20a
plumbing was the highest among the costs in the survey. We assumed Media beds 411.9 1; 3; 5; 20a
Land 115.1 n.a.
that a well needs to be constructed in order to get clean water from an Roof for fish tanks 40.7 10
aquifer, after which water can be used against cost for pumping and Bio-filter 30.8 10
maintenance. If we allocate 50% of the cost of land and general infra- Fish tanks 27.0 10
structure to the fish component, the investment on this is about Plumbing 27.0 10
Crates for vegetables 26.0 3
PhP194,000.
Well for fresh water 15.5 20
The estimated annual operational or variable cost, i.e., expenses Air pump 5.9 10
returning every year for a catfish-oriented aquaponics farm, was about Fishing gear 5.0 5
PhP 300,000 yr−1 in year 0, and thereafter more than PhP 316,000 Water pump 4.9 5
yr−1 (Table 8). Labor cost represented more than half of the operating Test toolkit 2.2 10
Clarifier 1.4 5
cost, and was equivalent to the pay of 1.5 full time staff. The operation Trays for seedling 1.2 0.33; 1a
cost attributable to catfish was estimated at 28%. Fish transport container 0.7 5
In the baseline scenario, taxes and insurance costs were not taken Total amount 1,259.8
into account because not all interviewed farmers paid insurances or
a
taxes. Moreover, the amounts paid for taxes or insurances varied Depending the component and its material.

widely. For the repair costs, the highest indicated % in the survey was
implemented (Table 8). For a 100% Jade perch aquaponics farm, the
higher assumed cost of fingerling and feeds increased the total oper-
ating cost to about PhP 345,000. The operation cost that was attributed
to Jade perch was estimated at 34%.
The revenue of the farm originated from the sales of fish, tomatoes,
and lettuce. The CBA assumed that all harvested products are sold. The
calculated annual harvest weights were 1250 kg of fish, 8320 kg of
lettuce, and 1985 kg of tomato once the system is operational; for year
one, these weights were 890, 5760 and 1060 kg, respectively.
Against the average wholesale price for catfish, in year 0 the total
revenue would be slightly more than PhP 400,000; and in later years,
the non-discounted revenues would be close to PhP 600,000, which
would equal to a discounted revenue (TDR) of PhP 555,000 in year 2
(Tables 8 and 9). If the farm were to shift to Jade perch in year 5, the
TDR would be about PhP 870,000; the latter would be double the
amount (437,000) of the catfish-oriented enterprise (Tables 8 and 9). In
case of catfish only, the revenues from fish would be close to 17% of the
total; after the shift to Jade perch, this contribution would increase to
more than 40%.
For the life span of 20 years, a catfish aquaponics farm has an es-
timated NPV of PhP 1.13 million; and the DBCR is 1.23 (Tables 9 and
10). For a lifespan of ten years the DBCR is 1.10. The payback time of
the aforementioned is 7 years. For a Jade perch-oriented aquaponics
business, the DBCR for 10 and 20 years is estimated at 1.5 and 1.7,
respectively (Tables 9 and 10), and pay-back time is 3 years. For a farm

Table 6
The computations of the plant components for the scaling of the system.

Day Above-ground plant growth N-content Nitrogen demand

Total mass Fruits Leaves N in leaves N in fruits Average Total N

gDW plant-1 gDW plant-1 gDW plant-1 g plant-1 g plant-1 g plant-1 g plant-1 d-1 g m-2 d-1 Total g d-1
Tomato 0 20 0 20 0.9 0 0.9
21 40 10 30 1.35 0.2 1.55
41 90 30 60 2.7 0.6 3.3 0.046 0.19 17.8
56 100 40 60 2.7 0.8 3.5

Day Above-ground plant growth N-content Nitrogen demand

Total mass Fruits Leaves N in leaves N in fruits Average Total N

-1 -1 -1 -1
gDW plant gDW plant gDW plant g plant g d-1 plant-1 g m-2 d-1 g d-1 Total g d-1

Lettuce 0 4 80 0.24 0.017 0.43 8.6 12.8

7 6 120 0.36 0.017 0.43 8.6
14 8 160 0.48 0.034 0.86 17.1
21 12 240 0.72 0.034 0.86 17.1
28 16 320 0.96

For lettuce the calculation considers 20 m2 per batch.

150
R.H. Bosma et al. Aquacultural Engineering 78 (2017) 146–154

Table 8 require dilution of the fish effluent (Vergote, 2009). Formulating a

The operational cost (rounded in 1000 PHP) of the aquaponics farm. specific fish feed low in Na might be difficult; therefore, Diver (2006)
suggested to grow low-nutrient requiring plants in aquaponics systems.
Description Amount
The latter however may have trade-offs on revenues and on cost if, e.g.,
Seeds 27.6 more space is required; Karimanzira et al. (2016) designed a system
Fertilizer 7.0 level decision support model. However, Wongkiewa et al. (2017) con-
Fingerlings catfish 17.5
cluded that quantitative parameters (e.g., ratios of different forms of
Fingerlings seabass 35
Feed catfish 22.6 nitrogen) differ from system to system due to the complexity of the
Feed Jade perch 34.0 biological system and are affected by several factors. Therefore such
Treatment pests & diseases 5.5 model parameters should be systematically controlled in a well-en-
Electricity 35.5 gineered system.
Transportation 3.9
Reducing the quantity of lettuce would allow the farmers to produce
Repairs 12.0
Labor 184.8 more tomato, but the latter requires more space, and thus a higher
Total amount 316.4 investment in greenhouse area. The cost of investing in greenhouses is
relatively high because of the high risk of typhoons in the region.
Because of the high frequency of strong storms in Negros Oriental, we
that shifts after 5 years from catfish to Jade perch, these figures are chose not to produce more tomato in order to limit the vegetable area to
estimated at 1.3 and 1.5 respectively, and the estimated cash flow 96m2. The latter is the maximum area of small greenhouses for which
would be positive after 6 years. the risk of destruction and capital loss are lower. Culturing vegetables
outdoors in the region is even more risky, also because of reduced
3.3. Sensitivity analysis options for biological pest control.
The quantity of fish produced is relatively small, as well as the ratio
When accounting a 10% VAT on farm sales of the system with of fish volume to vegetable area (1.4:180), close to the range
catfish, the NPV (for r = 8%) would become negative and the DBCR (1:30–1:100) for other documented farms (Langen, 2009). The fish
would fall below 1 for both the 10- and 20-year-project period. When component requires specific skills for its management and relatively
accounting this 10% VAT on the system shifting to Jade perch, the NPV high operational cost inducing losses as long as the scale remains lim-
would remain positive and the DBCR above 1, even for the higher ited (Engle, 2015; Langen, 2009). For a system with catfish, the capital
discount rates. When the VAT is included, the part of the operational investments are balanced with the revenues (about 17% each), but the
cost attributable to fish would increase about 7% points (34 and 41%, variable cost, estimated at about 28% of the operation for the farm with
without and with VAT, respectively). catfish, is higher than that of the fish component’s contribution to the
Assuming a 50% higher price for all products, and gross income revenue (17%). While the operational cost increases after the shift to
below the VAT boundary, the DBCR for the catfish scenario would Jade perch (34%), the contribution of the fish component to the income
reach values above 1.2 for all discount rates even if labor cost were is higher (41%). This strengthens the recommendation to shift to Jade
increased by 50% (Table 10). However, assuming that the farmer would perch, an example of a species with a higher wholesale price. Though
pay VAT, our sensitivity analysis shows that the positive effect on this after the shift to Jade perch the estimated operational cost of the fish
scenario would be limited if labor cost increases with more than 30% component is higher, the increase in revenue outweighs the cost, as
for discount rates above 10%. demonstrated by the higher DBCR. A trade-off may come when the
benefits get too high, as this will result in taxes increasing the part of
4. Discussion the operational cost attributable to fish with about 5% points.

Our study confirms the conclusions of Buysens and Gobin (2011)
that the economic feasibility of an aquaponics system would depend on
the choice of the fish species, the size of the fish component (PRIVA,
2009), and then specifically its turn-over. Tokunaga et al. (2015) con-
cluded that the commercial aquaponics farms in Hawai were less ben-
eficial than what the experiments promised; therefore, carefully
choosing fish species and geographical location is crucial. We will
discuss these factors below, as well as the assumptions, the use of
wholesale prices, and the system balance.
4.1. The system balance
The waste generated by fish fed on commercially available fish diets
does not perfectly match plant nutrient requirements; in particular Na,
Mg and Cl might cause imbalances (Seawright et al., 1998) and might
4.2. The risks of farming and insurances, taxes and other non-considered
cost
To compensate for the risks inherent to farming, a farmer will aim at
DBCR higher than 1.2. This is required to be able to accumulate a buffer
for bad years, e.g., partial harvest loss once every five years due to
diseases, or increased cost due to natural hazards. If risks are higher
they should aim for a higher DBCR, e.g., 1.5 for hazards occurring every
two years. On a case-to-case basis, to reduce impact of new epidemics
or natural disasters on farmers, the government may need to support
them, in particular if the farm gate prices are low due to mechanisms
keeping consumer prices low.
Several studies support the non-inclusion of harvest insurances.
First of all, while being set as examples, most existing crop insurance
programs function with subvention (Gardner and Kramer, 1986).

Table 9
The total discounted investment costs (TDIC), total discounted operating costs (TDOC) and total discounted revenues (TDR) at a discount rate of 8%, the undiscounted (UNB) and the
discounted net benefits (DNB = NPV) for a period of 20 years in the aquaponics farm with catfish only (rounded in 10,000 PHP).

Year 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Total

TDIC 168 1.8 1.6 4.2 1.4 3.9 3.4 1.1 1.0 2.7 8.3 0.8 2.1 0.7 0.7 2.9 0.8 0.5 1.1 0.6 208
TDOC 30 29 27 25 23 22 20 18 17 16 15 14 13 12 11 10 9 9 8 7 334
TDR 41 55 51 47 44 41 38 35 32 30 28 26 24 22 20 19 17 16 15 14 612
UNB −115 26 26 23 26 22 23 26 26 23 10 26 24 26 26 19 25 26 23 25 334
NPV −115 24 22 18 19 15 14 15 14 12 5 11 9 9 9 6 7 7 6 6 113

151
R.H. Bosma et al. Aquacultural Engineering 78 (2017) 146–154

Table 10
The DBCR for two project periods (10 and 20 years), various discount rates and with or without VAT, for three production systems: catfish only, seabass only, or switch from catfish to
Jade perch after 5 years, and a scenario of direct consumer sales with prices and labor cost increased 50% (PL50).

r Catfish Jade perch Catfish / Jade perch PL50 with Catfish

No VAT No VAT No VAT 10% VAT No VAT

10 yr 20 yr 10 yr 20 yr 10 yr 20 yr 10 yr 20 yr 10 yr 20 yr

4 1.16 1.31 1.61 1.81 1.38 1.66 1.21 1.42 1.47 1.63
8 1.10 1.23 1.53 1.70 1.29 1.52 1.14 1.32 1.40 1.54
10 1.07 1.19 1.49 1.64 1.25 1.45 1.11 1.27 1.37 1.49
12 1.05 1.15 1.46 1.69 1.21 1.39 1.08 1.22 1.34 1.45
16 0.99 1.07 1.39 1.49 1.13 1.27 1.02 1.13 1.28 1.37

Moreover, such an insurance may reduce the incentive of farmers to
aim for efficiency and increase their tendency for rent-seeking
(Roumasset, 2007); thus, increase the systemic risks of productivity
(Chambers and Quiggin, 2002). Last but not the least, an insurance
system might lead to a free rider’s behavior of the farmer who might not
aim for optimal investment in a robust and sustainable farm. Current
policies offering insurance to farmers seem to lean more on the interest
of shareholders of financial institutions rather than that of farmers. The
shareholders expect more than 12% of benefit, and they set the price of
the insurance; while most farmers earn less than 10% only on their total
investments.
When accounting a tax rate of 10% on the farm revenue, the NPV of
the recommended system shifting to Jade perch is profitable for all
considered discount rates and the DBCR for r = 8% remains above 1.3.
In years with good results, the farm owner might be subjected to tax on
income for percentages of up to 32% in the Philippines. To reduce these
taxes, the farm could pay premiums to the workers in good years or pay
for their training; and include in the operational cost the time spent for
arranging permits and paying taxes.
4.3. The level of product prices
All harvested products were assumed to be sold at wholesale prices.
Products not sold but consumed by farm owner, staff, and family would
decrease farm revenues; but would result in net gain for these con-
sumers. Their benefits, when accounted in the salaries, would reduce
the farm’s labor cost except in periods of high retail prices.
In the Philippines, the price of vegetables is relatively high due to
their scarcity (Cororaton and Corong, 2009), and a first condition for
feasible aquaponics is therefore fulfilled (Engle, 2015; Love et al., 2015)
To stimulate local production and to reduce trade deficits, government
imposes heavy tax on the imports of products from agriculture and
fisheries. Consequently, consumer price for fruits and vegetables is
higher than that of the global market price; and more than 60% of the
total household budget of poor people in the Philippines is spent on
food (Engle, 2015; Love et al., 2015). Thus, a shift to vegetables, like
sweet pepper which is better-priced than lettuce, might also increase
efficiency; but a market research should be done before making shifts in
the designed system.
The price used in the calculations, which is half of the retail price,
could be underestimated. Selling the products directly to consumers in
the market, or to specific customers for a price closer to retail price,
would give higher revenues. For the simulated farm with catfish only,
assuming 50% increase in product prices and in labor cost, our sensi-
tivity analysis confirmed a 30%-point increase in the long-term profit-
ability. Better wholesale prices for the fish and vegetables might be
negotiated or obtained in particular during periods when retail prices
escalate, such as during Easter, Christmas, and New year periods; and
thus increase farm revenues. Focusing the vegetable production on such
periods will have a limited effect due to the continuous staggered
production set-up of the farm, but might be worthwhile for the fish
component, if farmers would grow larger fish by temporarily adding
some tanks to the system.
Though the targeted market weight of fish (500 gr) seems low, but
for most customers this provides a convenient one-plate-dish. From
farmer’s perspective, the production of low live weight fish is profit-
able: the higher the weight, the more feed the fish uses for its main-
tenance requirements and the higher will be the cumulated risk of
mortality; and thus the production cost.
4.4. Upscaling and other options
Increasing scale can likewise result in economies of scale and might
improve the enterprise’s profitability, which in particular, is important
for the fish component. For fish, the lower cost per unit of output would
come from labor and transportation, while for some inputs such as
seeds and fertilizers, a lower price might be negotiated. However, in-
creasing the size of the fish component would require a larger market
and marketing effort for vegetables. If the DBCR is above 1.2, farmers
can afford the extra cost for marketing and transport. However, con-
sidering the VAT boundary and the risks inherent to a huge increase of
the vegetable component, more than tripling the size of the modelled
farm seems not recommendable; then 75 kg of fish can be produced
weekly. Near cities with a huge market paying high prices for vege-
tables and (specific) fish, an enterprise can make the necessary econo-
mies of scale on labor and transportation. Elsewhere, the small-scale
(family) farm, like modeled, selling directly to consumers, can fulfill its
multiple socio-economic functions.
Thus the fish component of an ideal aquaponics farm will remain
small and have relatively high cost which can only be recovered if high-
value fish for a niche market could be cultured (Engle, 2015; Buysens
and Gobin, 2011). The latter authors, after technical trial and testing
consumer preferences of Sander lucioperca, Lota lota and Scrotum barcoo
in Belgium, concluded that the last two were the preferred options as
they serve niche markets. The present study assumed a start up with
catfish because of the fish’s tolerance to low oxygen level and water
quality. A shift to Jade perch (Scrotum barcoo), or the endemic Lobed
river mullet (Cestraeus plicatilis), however, may likely increase the
profitability of the system.
Using the Life Cycle Assessment methodology, Forchino et al.
(2017) concluded that one of the main pathways to improve sustain-
ability is the optimization of management practices. Culturing the fish
in the same greenhouse with the plants can contribute to a higher CO2
content, and thus promote better plant growth (PRIVA, 2009). Vergote
(2009) and PRIVA (2009) concluded that large-scale aquaponics is
technically feasible and might be an efficient option to reduce en-
vironmental effluents and taxes in countries like Belgium and the
Netherlands, which enforce an environmental tax for the waste water
disposal. Gravel et al. (2015) demonstrated that the fish effluents sup-
press fungal diseases and stimulate root growth in tomato; using tomato
in aquaponics systems might thus have a positive environmental trade-
off through reduced use of pesticides and improved production. The

152
R.H. Bosma et al.
reduced use of pesticides might provide a temptation to sell the pro-
ducts as organic (Engle 2015); producers should consider that this
implies cost for certification to ensure quality and safety to the con-
sumers.
5. Conclusion
An aquaponics project with the low-priced catfish is not feasible,
more so, under non-tax and non-insurance assumptions. However, a
farm shifting to a higher value species like Jade perch within 5 years
can support taxes as well as cope with risks. In general, the main turn-
over will come from vegetables; and having a nearby market with high
demand and good prices is the first condition for success. Aquaponics
farms, producing low-priced fish or vegetables, seem to have grim fu-
ture. Given the required high market price to make the system finan-
cially feasible, the best prospects for the relatively small fish component
in the aquaponics system is a niche market that is too small for com-
mercial mono-culture fish farmers.
Funding
The initial study was financially supported by the potential investor,
Mr Frank Schinkel, according to the rules of the Wageningen
University’s MSc course: Academic Consultancy Training.
Complementary analysis and writing of this article did not receive any
support.
Acknowledgements
The design of the system and the CBA were performed by a group of
international MSc students in the framework of the course Academic
Consultancy Training (ACT) at Wageningen University. The authors
acknowledge Mr. Zhihan Li, team-member who changed his e-mail
address, for his contribution, the three anonymous referees for their
suggestions, and Mrs. Lorna Calumpang (lornamiracalumpang745@
gmail.com) for the grammatical editing of the paper.
References
Aquaponics Philippines, 2013. Aquaponics in the Philippines. Retrieved from http://
aquaponicsphilippines.com (Accessed 27, June 2017).
Aquaponics Philippines, 2014. Commercial Aquaponics Setups in the Philippines.
Retrieved from http://aquaponicsphilippines.com (Accessed 27, June 2017).
BIR, 2015. Value Added Tax. Bureau of Internal Revenues, Republic of the Philippines
http://www.bir.gov.ph/index.php/tax-information/value-added-tax.html (Accessed
25, September 2015).
Boardman, A.E., Greenberg, D.H., Vining, A.R., Weimer, D.L., 2011. Cost-Benefit
Analysis: Concepts and Practice, Fourth Edition. Pearson Education, Inc., Upper
Saddle River USA.
Buysens, S., Gobin, B., 2011. Consumentenonderzoek vis. Report from: Provinciaal
Proefcentrum voor de Groenteteelt Oost-Vlaanderen vzw.
Chambers, R.G., Quiggin, J., 2002. Optimal producer behavior in the presence of area-
yield crop insurance. Amer. J. Agr. Econ. 84 (2), 320–334.
Chaves, P.A., Sutherland, R.M., Laird, L.M., 1999. An economic and technical evaluation
of integrating hydroponics in a recirculation fish production system. Aquac. Econ.
Manag. 3 (1), 83–91.
Cororaton, C.B., Corong, E.L., 2009. Philippine Agricultural and Food Policies. IFPRI
Discussion Paper 161.
CountrySTAT, 2015. Food and agriculture data network. Retrieved from http://
countrystat.bas.gov.ph/selection.asp (Accessed 05, February 2015).
Daneshvar, S., Kaleibar, M.M., 2010. The minimal cost-benefit ratio and maximal benefit-
cost ratio, Second International Conference on Engineering System Management and
Applications. March 30 - April 1 2010, (Accessed on: http://ieeexplore.ieee.org/xpl/
mostRecentIssue.jsp?punumber=5523174)
De Pinheiro Henriques, A.R., Marcelis, L.F.M., 2000. Regulation of growth at steady-state
nitrogen nutrition in lettuce (Lactuca sativa L.): interactive effects of nitrogen and
irradiance. Annals of Botany 86 (6), 1073–1080.
153
Aquacultural Engineering 78 (2017) 146–154
Diver, S., 2006. Aquaponics-Integration of hydroponics with aquaculture. ATTRA 1–28.
Eding, E.H., van Weerd, J.H., 1999. Grundlagen, Aufbau und Management von
Kreislaufanlagen. In: Bohl, M. (Ed.), Zucht und Produktion von Süsswasserfischen,
2nd edn. DLG −Verlag, Frankfurt, München, pp. 436–491.
Engle, C.R., 2015. Economics of Aquaponics. Southern Regional Aquaculture Center
(SRAC) Publication No. 5006.
FAO, 2001. Integrated Agriculture-aquaculture: A Primer. Issue 407. ISBN 9251045992.
FAO, 2014. Small-scale Aquaponic Food Production − Integrated Fish and Plant Farming.
FAO Fisheries Aquac. Techn. (Paper No. 589).
Forchino, A.A., Lourguioui, H., Brigolina, D., Pastresa, R., 2017. Aquaponics and sus-
tainability: the comparison of two different aquaponic techniques using the Life Cycle
Assessment (LCA). Aquac. Eng. 77, 80–88.
Gardner, B.L., Kramer, R., 1986. Experience with crop insurance in the United States.
Crop Insurance for Agricultural Development: Issues and Experience. Published for
IFPRI. The Johns Hopkins University Press, Baltimore and London, pp. 1986.
Graber, A., Junge, R., 2009. Aquaponic Systems: nutrient recycling from fish wastewater
by vegetable production. Desalination 246 (1), 147–156.
Gravel, V., Dorais, M., Dey, D., Vandenberg, G., 2015. Fish effluents promote root growth
and suppress fungal diseases in tomato transplants. Can. J. Plant Sci. 95 (2), 427–436.
Hanley, N., Spash, C.L., 1993. Cost-Benefit Analysis and the Environment. Edward Elgar
publishing, Cheltenham UK, Northhampton, U.S.
Hogendoorn, H., 1983. Growth and production of the African catfish, Clarias Lazera
(C. & V.) III. Bioenergetic relations of body weight and feeding level. Aquaculture 35,
1–17.
INAPRO, 2014. Innovative model and demonstration-based water management for re-
source efficiency in integrated multi-trophic aquaculture and horticulture systems.
Retrieved from http://www.inapro-project.eu/ (Accessed 10, January 2015).
Kamal, S.M., 2006. Aquaponic production of Nile tilapia (Oreochromis Niloticus) and Bell
pepper (Capsicum Annuum) in recirculating water system. Egypt. J . Aquat. Biol. Fish
10 (3), 85–97.
Karimanzira, D., Keesman, K.J., Kloas, W., Baganze, D., Rauschenbach, T., 2016. Dynamic
modeling of the INAPRO aquaponic system. Aquac. Eng. 75, 29–45.
Langen, E., 2009. Tilapia-pioniers leggen toe op hun vis. Groenten + fruit: weekblad voor
de voedingstuinbouw 49, 11.
Liang, J.Y., Chien, Y.H., 2013. Effects of feeding frequency and photoperiod on water
quality and crop production in a tilapia–water spinach raft aquaponics system. Intern.
Biodeteriorat. Biodegrad. 85, 693–700.
Love, D.C., Fry, J.P., Li Ximin Hill, E.S., Genello, L., Semmens, K., Thompson, K.E., 2015.
Commercial aquaponics production and profitability: findings from an international
survey. Aquaculture 435, 67–74.
Minkenberg, O.P., Ottenheim, J.J., 1990. Effect of leaf nitrogen content of tomato plants
on preference and performance of a leafmining fly. Oecologia 83 (3), 291–298.
Mori, M., Amato, M., Di Mola, I., Caputo, R., Chiaranda, F.Q., Di Tommaso, T., 2008.
Productive behaviour of cherry-type tomato irrigated with saline water in relation to
nitrogen fertilisation. Eur. J. Agron. 29 (2–3), 135–143.
Nguyen, N., Quan, N.H., Hao N.V., Bosma, R.H., Verreth, J.A.V., Verdegem, M.C.J.,
Eding, E.H., 2016. Nutrient mass balances, water quality and water use of striped
catfish (Pangasianodon hypophthalmus, Sauvage, 1878) in flow-through and re-
circulation systems. Chapter 3 in WUR-PhD thesis available through: http://edepot.
wur.nl/394644; forthcoming in Aquaculture.
Perkins, F., 1994. Discounted project assessment criteria. Practical cost benefit analysis:
Basic concepts and applications. MacMillan Education Australia PTY, LTD, Sydney.
PRIVA, 2009. Visteelt in de Glastuinbouw. Eindrapport Project EcoFutura. www.priva.nl.
Roosta, H.R., 2014. Effects of foliar spray of K on mint, radish, parsley and coriander
plants in aquaponic system. J. Plant Nutr. 37 (14), 2236–2254.
Roumasset, J., 2007. The economics of agricultural development: what have we learned.
In: Balicsan, A.M., Fuwa, N. (Eds.), Reassessing the Rural Agenda. SEARCA & Institute
for SEAsian Studies, Singapore, pp. 33–61.
Rupasinghe, J.W., Kennedy, J.O.S., 2006. Optimal batch lengths for barramundi farming
under seasonal variations: a dynamic programming approach. In: Proceedings IIFET.
Porthsmouth.
Sajjadinia, A.R., Ershadi, A., Roosta, H.R., 2010. Comparison of eco-physiological char-
acteristics of pepper in hydroponic and aquaponic systems. J. Sci. Technol. Greenh.
Cult. 1 (2), 1–8.
Seawright, D.E., Stickney, R.R., Walker, R.B., 1998. Nutrient dynamics in integrated
aquaculture–hydroponics systems. Aquaculture 160, 215–237.
Testa, R., Foderà, M., Trapani, A.M.D., Tudisca, S., Sgroi, F., 2015. Choice between al-
ternative investments in agriculture: the role of organic farming to avoid the aban-
donment of rural areas. Ecol. Eng. 83, 227–232.
Tokunaga, K., Tamru, C., Ako, H., Leung, P.S., 2015. Economics of small scale Aquaponics
in Hawai. World Aquac. Soc 46 (1), 20–32.
Vergote, N., 2009. Is tomaat met tilapia het nieuw recept? Provinciaal Proefcentrum voor
de Groenteteelt Oost-Vlaanderen vzw.
Wongkiewa, S., Hub, Z., Chandranc, K., Leed, J.W., Khanala, S.K., 2017. Nitrogen
transformations in aquaponic systems?: A review. Aquac. Eng. 76, 9–19.
Yoshida, T., Ueda, H., Manishi, K., Fukumoto, Y., 1997. Effects of irrigation and nitrogen
application on the accumulation of calcium and nitrogen in tomato (Lycopersicon
esculentum) fruits and their relation to blossom-end rot. Japan. J. Soil Sci. Plant Nutr.
68 (2), 178–180.
R.H. Bosma et al. Aquacultural Engineering 78 (2017) 146–154

Roel H. Bosma Joline Poulie

Lysette Lacambra Marie J. Schwaner

Ynze Landstra Yi Yin

Chiara Perini

154