Financial viability of Penaeus setiferus versus Penaeus vannamei with continuous live harvesting and one final harvest strategies in South Carolina

[From the U.S. Government Printing Office, www.gpo.gov]

78* 76* 36* 74c

IN
:@,_ eX @1_

. @n_ 6,

-4
-4k

North Carolina
CAPE LOOKOUT.
FINANCIAL VIABH, OF
36o 74'
PENAEUS
,STING
41,MVE
S W
UITTH
"SO "CA'A"OLIN-AA,
WILMING @k

!7
"Ae

@61
7
TU ipzi 61;Aw -
South Carolina
0@
3z

z
VL3t
0
34
tni,C. law
a shwar -,Cjr'r ii@ikjf-- e-, 4
@
_d W`SfW6i'
-Bj

CHARLESTO

Marine Resources Divis!bh-4,-,,1--@.'--.
Technical Report,, -849,
SAV@NNAH
PM
36

Georg
2.
3
I "A"

RUNSWICAU

JACK ONVILLE
-South Carolina
q.
Department of Natural Resources
"'ST. A'U TINE. Q)
Florida
0
SH222
.S6S6 DAY A BE CH
no .84
82o 80* 780

FINANCIAL VIABILITY OF PENAEUS SETIFERUS VERSUS
PENAEUS VANNAMEI WITH CONTINUOUS LIVE
HARVESTING AND ONE FINAL HARVEST
STRATEGIES IN SOUTH CAROLINA

S. Sureshwaran, C. Greene, R.J. Rhodes
Craig L. Browdy and Al Stokes

Marine Resources Division
Technical Report Number 84
December, 1994

CAR

01P NAVJ
13

South Carolina
/71 Cn
Department of Natural Resources
LIBRARy
NOAA/cCEH
1990 HOBSON AVE.
'6 CHAS. SC 2.9408-2623

FINANCIAL VIABILITY OF PENAEUS SETIFERUS VERSUS PENAEUS VAAW"V
WITH CONTINUOUS LIVE HARVESTING AND ONE FINAL HARVEST
STRATEGIES IN SOUTH CAROLINA

S. Sureshwaran'
Carol Greene'

and

Raymond J. Rhodes'
Craig L. Browdy'
Al Stok&

Technical Report Number 84
December, 1994

'Department of Agribusiness and Economics
South Carolina State University
Campus Post Office Box 7282
Orangeburg, SC 29117

and

'Economic Analysis and Seafood Marketing Program
Office of Fisheries Management
Division of Marine Resources
S. C. Department of Natural Resources
Post Office Box 12559
Charleston, SC 29422-2559

and

3Waddell Mariculture Research and Development Center
Marine Resources Research Institute
Division of Marine Resources
S.C. Department of Natural Resources
Post Office Box 809
Bluffton, SC 29910

This work was partially funded by the U.S. Department of Agriculture as administered by the Oceanic Institute/Gulf
Coast Research Laboratory Consortium. The views expressed in this report do not necessarily reflect those of the
U.S. Department of Agriculture, Oceanic Institute, S.C. Department of Natural Resources, or S.C. State University.
Any commercial product or trade name mentioned herein is not to be construed as an endorsement.

Table of Contents

Page

List of Tables .................................................... iv

List of Figures ................................................... iv

Summary ....................................................... I

Introduction ..................................................... I

Literature Review ................................................. 2

Methods and Data ......................... ........................ 3

Facility Design and Equipment ................................... 3
Production Assumptions ........................................ 5
Major Financial and Operating Assumptions .......................... 5
Production Scenarios .......................................... 5

Results ......................................................... 6

Base Scenarios .............................................. 6
Alternative Scenarios ......................................... 10

Discussion ..................................................... 10

Literature Cited .................................................. 12

Appendix 1: Shrimp Harvest Weight Model ............................. 13

List of Tables

Table Page

1 Summary of Facility and Equipment Costs for a Hypothetical Shrimp
Farm in South Carolina, 1993 ................................... 7
6

2 Projected Annual Income Statement for Operating Years Three
Through Ten, and Discounted Cash Flow Analysis in the Base Scenario
for a Hypothetical Penaeus vannamei Shrimp Farm in South Carolina ........ 8

3 Projected Annual Income Statement for Operating Years Three
Through Ten and Discounted Cash Flow Analysis in the Base Scenario
for a Hypothetical Penaeus setiferus Shrimp Farm in South Carolina ......... 9

4 Discounted Cash Flow Analysis for the Alternative Scenarios for
Hypothetical Penaeus setiferus and Penaeus vannamei Shrimp Farm in South
Carolina, 1993 .............................................. 11

5. Regression Coefficients for the Harvest Weight Model .................. 15

List of Figures

Figure

I Hypothetical Shrimp Farm Design ................................. 4

SUNMARY The ten-year (after tax) internal rate of
return (IRR) for P. vannamei is 15.1 %. The
It is anticipated that prospective aquacult- projected IRR is greater than the assumed
urists, investors, and perhaps lending insti- base-scenario after-tax discount rate of 10%
tutions and policy makers will desire informa- for P.vannamei. However, the projected IRR
tion on the financial feasibility of producing for P. setiferus, -4.9 %. This is less than the
indigenous Penaeus setiferus versus specific base-scenario discount rate. Net present value
pathogen free Penaeus vannamei. This report (NPV) in ten years for P. vannamei is
was undertaken to provide an accurate and $232,000 and for P. setiferus is a negative
objective picture of the profit or loss from $568,000. The negative NPV indicates that at
farming these species. In addition, an evalua- current prices, technology, and the assumed
tion is made of the effects of a continuous discount rate, P. setiferus is n-ot profitable.
live harvesting strategy versus a single final
harvest strategy on the profitability of the two The IRR and NPV are highly sensitive to
species. the assumed stocking densities, survival rates
and expected prices at harvest. Given a dis-
The hypothetical shrimp farm described in count rate of 10%, small changes in stocking
this report includes 24 ponds, each I hectare densities, survival rates and prices can result in
in size, located on 31 hectares of land that is large losses even for a P. vannamei shrimp
leased near a saltwater source. The base sce- farm. However, continuous live harvesting
nario assumes one final harvest strategy, stoc- has positive effects on net cash flows.
king density of 80 postlarvae/meter', an ag- Because of the limited market for live shrimps,
gregate survival rate of 70%, and a price of producers should remain alert for over-supply
$4.95/kg for P. vannamei and $4.73/kg for P. conditions that can have a negative impact on
setiferus. The effects of alternative stocking prices. With continuous live harvesting, IRR
densities, survival rates, prices, and live har- increases to 16.4% for P. vannamei and to
vesting are investigated in 7 other scenarios. 2.2 % for P. setiferus. Even with continuous
All the 8 scenarios assume a feed conversion live harvesting, P. setiferus remains an un-
rate of 2: 1, and the length of the grow-out profitable enterprise.
cycle as 5 months.
Commercial shrimp farms utilizing
Initial investment in facility and equip- specific pathogen free P, vannamei postlarvae
ment is approximately $1 million. Because of (PL) can be profitable in South Carolina.
differences in growth rates and resulting feed Farms can have larger profits by practicing
costs, total initial investment for both constru- continuous live harvesting strategies.
cting and operating the facility depends on However, the indigenous P. setiferus is not
species produced. In the base scenario, total profitable. Therefore, regulatory actions that
initial investment including operating costs, hinder the import of specific pathogen free P.
for P. vannamei is $1.38 million and for P. vannamei PL from out-of-state hatcheries
setiferus is $1.35 million'. After the third could have adverse impacts on the South Caro-
year, when the farm produces at full capacity, lina commercial shrimp farms.
net cash flow after estimated taxes for P.
vannamei is $250,000 and for P. setiferus is MRODUCTION
$103,000.
In 1993, 26 farms in South Carolina,
Hawaii and Texas produced a record crop of
farm-raised shrimp, approximately 2,500
metric tons, 25 % more than the estimated
'This assumes operating at 50% of full capacity in 2,000 tons in 1992 (Rosenberry, 1993).
year 1. However, the United States remained a rela-
tively small producer of farm-raised shrimp,
accounting for less than 2 % of the production

2
in the Western Hemisphere. In 1993, South of aquaculture projects, including discounted
Carolina with 14 fanns and about 100 hectares cash-flow technique, is discussed in Rhodes
(ha) in commercial production, accounted for (1991).
approximately 20% of the nation's cultured
shrimp output (unpublished data, S.C. Depart- LITERATURE REVIEW
ment of Natural Resources, 1994).
Sandifer, et al. (1993) conducted pond
The profitability of South Carolina's (SC) experiments at Waddell Mariculture Center to
shrimp mariculture industry is dependent on compare production characteristics of the
many factors including the availability of native P. setiferus and Pacific P. vannamei
viable postlarvae (PL) at a competitive price. white shrimp in South Carolina. The produc-
The SC farms have been generally dependent tion levels achieved in 1989 are thought to be
on out-of-state suppliers for PL of the species among the highest achieved with P. setiferus in
of choice, P. vannamei, a non-indigenous pond culture. Their results suggest that P.
marine shrimp (Rhodes, et al. 1992). Howev- setiferus may be a viable alternative to P.
er, farmers have become apprehensive about vannamei for intensive cultivation in the con-
future supply and quality of imported PL. In tinental U.S. when P. vannamei are unavail-
1989, many farms had to reduce planned able. They suggest that further evaluation of
stocking densities and/or not stock all their this potential is needed.
ponds due to an apparent shortage of quality
PL's (McGovem-Hopkins et al. 1991). In Griffin, et al. (1984) used a conceptual
addition, environmental concerns have incre- model that included production, engineering,
ased in recent years regarding the perceived marketing, environment and profit as sub-
impacts of farming non-indigenous shrimp in models and a bioeconomic factors simulation
the United States. Industry awareness has also model to evaluate a projected penaeid shrimp
increased relative to possible negative impacts maricultural operation on the Texas coast.
of shrimp diseases carried by PL from out-of- The results showed that the operation would
state hatcheries. Consequently, as concerns prove marginally economically feasible based
have increased, research has been conducted to on assumptions of the study. A 2 % chance of
evaluate the financial feasibility of producing loss and a 4.5 % annual return on investment
commercial quantities of indigenous P. setif- were predicted by using baseline simulations.
erus versus specific pathogen-free P. vannamei
in South Carolina. In addition, the effects of Adams, et al. (1980) developed a bio-
continuous live harvesting versus one final economic engineering model for shrimp mari-
harvest on the profitability of the two species culture systems for a hypothetical grow-out
is evaluated. The information in this report operation in Brazoria County on the northern
should be of use to prospective aquaculturists, Texas coast. A budget simulation was devel-
investors, and perhaps lending institutions and oped to examine economies of size. Budgeting
policy makers. and cash-flow statements were used to examine
penaeid shrimp mariculture systems. Their
Capital budgeting decisions, i.e., all ac- results suggest that the size of the individual
tions relating to the planning and financing of pond which captures most economies of size is
capital outlays for the purpose of purchasing 2.5 acres and the number of ponds which
equipment and facilities, are a key factor in the achieves most of the economies of size for the
long-term profitability of the shrimp farm. In firm is 24. For this operation, IRR to total
this study, discounted cash flow analysis, one investment is 17%.
of the financial tools that is used to aid an
investor in making wise capital budgeting Hollin and Griffin (1985) examined the
decisions, is used to compare the profitability economies of: (1) growing one crop of large
of P. setiferus versus P. vannamei and contin- shrimp per year versus two crops of medium
uous live versus one final harvesting strategies. size shrimp, and (2) small, intensive ponds
The application of financial feasibility analysis versus large, sen-ii-intensive ponds. The after-

3
tax internal rate of return (IRR) in the base of total facilities and pond size generally in-
scenario for the 500 acre system was 14.61 % crease IRR. For example, a 40 ha facility
and for the 40 acre system was 9.1 %. When with 4 ha ponds had a mean after-tax IRR of
the production strategy was changed from one 0.75 %, while a 400 ha facility with 20 ha
crop to two crops, the IRR on the 500 acre pond sizes had a mean IRR of 13.31
system increased to 22.8 %. The most signifi-
cant change in the IRR was brought about by MIETHODS AND DATA
increasing the survival rates. When the sur-
vival rate was increased from 50 to 70%, IRR The hypothetical farm described in this
increased from 14.6 to 25%. paper is based on recommended best manage-
ment practices. Production and cost estimates
Pardy, et al. (1983) estimated density are predicated upon experience at Waddell
dependent growth equations for two species of Mariculture Center (WMC), South Carolina
penaeid shrimps, P. stylirostris and P. van- commercial shrimp farms and vendors of
namei. Based on these equations, a simulation supplies and services to the aquaculture indus-
model was developed to examine the effects of try. Other agencies, such as USDA-Soil Con-
alternative stocking densities and cropping servation Service, U.S. Army Corps of Engi-
schemes on various variables including gross neers, and S.C. Department of Health and
revenue and revenue above total selected costs. Environmental Control, provided data on pond
They found that the one harvest production construction costs, licenses, permits, etc.
strategy resulted in the greatest gross revenue,
no matter what stocking densities were chosen. Facilfty Desigg and Eatdpment
The one crop system generally results in larger
shrimp and thus, greater market price and The hypothetical facility analyzed in this
higher revenue above selected costs. report consists of 24 ponds each of I hectare
in size constructed on a 31 hectare farm that is
Griffin and Lambregts (1993) evaluated leased near a saltwater source. The average
the effects of pond design (pond size, pond pond depth is 1.3 m. A 3-m levee (wide
shape, levee crown size and canal bank slope) enough for service vehicles) separates each
on the after-tax IRR of a 40 ha shrimp farm. pond. It is assumed that the use of the land
Regression analysis was used to examine the was to grow row crops and consequently there
relationship between pond design variables and are few stands of trees.
the IRR. Results suggest that pond shape, fol- Water exchange is made available to im-
lowed by pond size, were the most influential prove water quality (such as dissolved oxygen
variables. An increase in pond size from 2 to and water temperature). A minimum water
10 ha increased IRR from 17.2 to 21.3%. exchange rate of 5 % is used in this study.
The results are specific to the design, size of However, the farm has the capacity to
the farm and soil type and therefore, can not exchange a maximum of 20% of the water in
be extrapolated outside the analyzed designs. any one pond. Therefore, the farm has six
The authors also addressed the possibility of pumps with flow ratings of 9.1 in' of water
increasing operating risks associated with large per minute and 25 horsepower motors. A 12
ponds. in levee separates each pond from the fill and
the discharge canals (Figure 1). The fill canal
Hanson, et al. (1985) analyzed the effects water will be gravity fed into each pond via an
of 12 different facility sizes on the profitability intake riser and released through a discharge
of producing a single species of shrimp, P. riser into the discharge canal (Figure 1).
stylirostris. Stochastic processing of the model These risers regulate water flow with a system
permitted random fluctuations in prices, pro- of boards. Two 10-horsepower paddle-wheel
duction, weather and survival rates within their aerators will be placed in each pond. All
probability density functions. However, risks electrical wiring for pumps and aerators will
associated with larger pond sizes are not exam- be 3 phase.
ined. The results suggest that increased sizes

Discharge P S.,,,...,pettling Basin
Structure-,,
Intake Structure

- ------ ------ Intake Canal

Discharge Canal

Arrows indicate direc-
tion of water flow.

SCALE
0.5 inch = 110 yds.
Legend

12M (39ft.) Levee
Acres (1 HA) 3 M (1 Oft.) Levee
F Each pond is approx. 2.5

Figure 1 - Hypothetical Shrimp Farm Design
(Gravity Feed Water System)

5
Other facilities and equipment include: (1) third year onwards production is at 100 % of
an office trailer, (2) a 1,000 square feet lab/sh- capacity; and (7) the actual operating time is a
op, (3) a 2,000 square feet warehouse, (4) six month season. The straight-line deprecia-
three medium sized silos each holding 50,000 tion method is used. A business manager, a
pounds of feed, (5) one longbed pickup truck technical manager, a clerk/typist/receptionist
and one flatbed truck, (6) one 23 horsepower and two pond management and maintenance
tractor for light work and towing feeder and crew work full time. Part time employees are
one 52 horsepower tractor for canal and pond a security officer, and harvest and other sea-
maintenance, (7) one feed blower, (8) one ice sonal labor. It is also assumed that this hypo-
machine, (9) harvest, office, lab equipment, thetical facility is not funded by debt capital,
and (10) pre-start up project and survey re- i.e., no loans. The annual financial projec-
ports. tions for this facility are generated using a
Lotus 1-2-3 spreadsheet template prepared by
Production AssumRti
:ons Applied Analysis, Inc. (AAI), (Leung and
Rowland, 1989). The model does not account
Specific pathogen free (SPF) P. vannamei for natural disasters such as droughts, hurri-
is used in this study. SPF P. vannamei be- canes, etc.
came available for distribution to the United
States in 1989 (Wyban, et al. 1993). It has Production Scenaxios
been found that use of SPF P. vannamei has
increased survival rate, feed efficiency, pro- In the base scenarios, Scenario S and S2A,
duction and profitability for the shrimp indus- ponds are stocked at 80 PL/ml in late April.
try (Wyban, et al. 1992). An aggregate survival rate of 70% is assumed.
Based on the growth model, harvest weights
The length of grow-out cycle is 5.0 for P. vannamei and P. setiferus were estimat-
months. Risks associated with growing shrimp ed. In addition, it is assumed that all shrimps
include natural disasters nutritional and envi- are produced for normal market sales, i.e., no
ronmental factors, such as low quality feed, live shrimp sales. These shrimps are sold
poor water quality, pollution, and associated head-on to individuals, restaurant distributors,
low dissolved oxygen events. The risks of and wholesalers. Prices are those that pre-
growing shrimp are not examined in this vailed in South Carolina in 1993 for the esti-
study. A feed conversion rate of 2.0:1 is used mated harvest weights.
in this study. The harvest weight is estimated
using a density dependent growth model (see The following 7 scenarios for the two
Appendix 1). This growth model was estimat- species are developed to evaluate the effects of
ed using data collected at Waddell Mariculture alternative stocking densities, survival rates,
Center and adjusted for commercial grow-out and prices' (see Table 4):
experience.
Scenario S: Stocking density of 80
Major Fmmcial and ORtratima PL/m, aggregate survival rate of
Assppjgfions 70%, price of $4.95/kg for P. vanna-
mei and $4.73/kg for P. setiferus
The major financial and operating assum-
ptions are the following: (1) the price of PL Scenario SIA: Stocking density of 60
including transportation is $10.00 per 1,000; PL/m@, aggregate survival rate of
(2) cost of feed (F.O.B., farm site) is $0.55 70%, price of $4.95/kg for P. vanna-
per kilogram (kg); (3) electricity cost is mei and $4.73/kg for P. setiferus
$0.07/kilowatt-hour(kwh); (4) discount rate is
10%; (5) planning horizon is 10 years; (6)
initial investment starts in year 0, first year 11994 prices reported by SC shrimp producers were
output is 50% of maximum capacity, second significantly higher than 1993 prices.
year output is 75% of maximuirn capacity, and

6
Scenario S1B: Stocking density of species produced. For P. vannamei, operating
100 PL/ml, aggregate survival rate of cost in year 1 is $0.38 million. Therefore, P.
70%, price of $4.95/kg for P. vanna- vannamei requires a total initial investment of
mei and $4.73/kg for P. setiferus $1.36 million before any revenue from sales is
received. For P. setiferus, projected operating
Scenario S2A: Stocking density of 80 cost in year 1 is $0.35 million with a total
PL/ml, aggregate survival rate of initial investment of $1.33 million.
70%, price of $4.73/kg.
A simple, pro forma annual income state-
Scenario S2B: Stocking density of 80 ment for operating years three through ten was
PL/m@, aggregate survival rate of generated for the hypothetical P. vannamei
70%, price of $5.28/kg. shrimp farm (Table 2). Projected annual sales
are $ 1. 11 million at full capacity and total cash
Scenario S3A: Stocking density of 80 operating cost is $0.77 million. Feed and PL
PL/ne, aggregate survival rate of accounted for the largest percentage of operat-
65%, price of $4.95/kg. ing cost, 3 3 % and 25 %, respectively. Energy
accounted for 10% of the operating cost.
Scenario S3B: Stocking density of 80 Total annual cash outflow is $0.86 million.
PL/rre, aggregate survival rate of Net cash flow after taxes is $0.25 million.
75 %, price of S4.95/kg.
The ten-year (after estimated income tax-
Another scenario, Scenario S4 (see Table es) internal rate of return (IRR) and net pres-
4), evaluates the effects of live harvesting on ent value are projected. The ten-year IRR is
IRR and NPV. The marketing experience of 15. 1 % (Table 2), which is greater than the
S.C. shrimp growers have shown that these base-scenario after-tax discount rate of 10%.
buyers are seafood distributors or "live haul- Net present value (NPV) in ten years is pro-
ers" selling to Oriental restaurants in the jected to be $232,000.
Northeastern United States (Rhodes, et al.
1994). The projected aggregate demand for The pro forma annual income statement
live marine shrimp by restaurants in the conti- for operating years three through ten for a
nental U.S. is quite small, approximately 30% hypothetical P. setiferus shrimp farm in South
of the 1993 U.S. farmed shrimp production. Carolina was generated (Table 3). Projected
Therefore, in Scenario S4, only 5 % of produc- annual sales are $0.81 million. Total cash
tion is assumed as live marketed. VvIen har- operating cost is $0.71 million. Feed and PL
vesting shrimp for live shrimp buyers, it is accounted for the largest percentage of pro-
assumed that labor costs $200 for a 250 kg jected operating cost, 27%. Energy accounted
shipment. In this analysis, no other changes in for 12% of the operating cost. Total annual
operating costs are associated with live shrimp cash outflow is $0.71 million. Net cash flow
marketing. after taxes is $0. 10 million.

RESLTLTS The ten-year IRR is negative 4.9% and
NPV is negative $568,000. The negative NPV
Base Scenarios for P. setiferus indicates that this operation
will not generate a positive return on equity if
Initial investment in facilities and equip- all shrimp are used for normal market sales.
ment is $998,000 (Table 1). Equipment costs Unless prices rise above $5.48 per kg, an
were approximately 49% of this cost. Land increase of approximately 16% above levels
clearing and construction costs account for for 1993, producing P. setifents will not be
approximately 48 %. profitable based upon this analysis. In con-
trast, the positive NPV for P. vannamei shows
Due to differences in growth rates and the potential profitability of culturing this
resulting feed costs, operating costs depend on species. Prices in 1993 are approximately

Table 1. Summary of Facility and Equipment Costs for a Hypothetical Shrimp Farm in South
Carolina, 1993.

Item Cost Percent' Useful Years

DEVELOPMENT COST
Project Report 10,000 20
Project Manager 20,000
Subtotal $ 30,000 3.0

LAND CLEARING AND FACILITIES
Land Clearing 85,800 20
Pond Construction 180,000 10
Discharge and Intake System 96,000 10
Buildings 44,000 20
Other 71,955
Subtotal $ 477,755 47.9

EQUIPMENT
Harvest Equipment 15,000 10
Feed Storage Bins 36,000 10
Paddlewheels 240,000 5
Trucks/Tractors 50,000 5
Feeding System 40,500 5
Power Equipment 11,000 10
Pumps 82,800 5
Other 15,000
Subtotal $ 490,300 49.1

TOTAL COSTS 998,055 100.0

Percent of total cost

8
Table 2. Projected Annual Income Statement for Operating Years Three Through Ten, and
Discounted Cash Flow Analysis in the Base Scenario for a Hypothetical Penaeus
vannamei Shrimp Farm in South Carolina.

Item Value or Cost Percent'
(In Thousands)

Proiected Annual Sales $ 1,108
(223,910 kilograms at $4.95/kg)

Projected Annual Expenses
Juveniles 192 25
Feed 252 33
Energy 80 10
Lease Rent 60 8
Labor 59 8
Salaried Personnel 83 11
Contingency 8 1
Other 33 4

TOTAL OPERATING COST $ 767 100

Projected Depreciation 129
Total Operating Costs with Depreciation 897

Projected Taxable Income 212
(Sales Minus Total Operating Costs)

Taxes (Federal. and Local) 79

Income After Taxes 132
(Taxable Income Minus Taxes)

Total Annual Cash Outflow 847
(Total Operating Costs plus Taxes)

Net Cash Flow 262
(Sales Minus Annual Cash Outflow)

Discounted Cash Flow Analysis
10 years
Net Present Value (dollars 000's) at 10% 232

Internal Rate of Return (percent) 15.09%

'Percentage of total operating cost.

9
Table 3. Projected Annual Income Statement for Operating Years Three Through Ten, and
Discounted Cash Flow Analysis in the Base Scenario for a Hypothetical Penaeus
setiferus Shrimp Farm in South Carolina.

Item Value or Cost Percent'
(in Thousands)

Proiected Annual Sales $ 810
(171,226 kilograms at $4.73/kg)

Proiected Annual ExRenses

Juveniles 192 27
Feed 192 27
Energy 80 12
Lease Rent 60 8
Labor 59 8
Salaried Personnel 83 12
Contingency 7 1
Other 33 5

TOTAL OPERATING COST $706 100

Projected Depreciation 129
Total Operating Costs with Depreciation 837

Projected Taxable Income -28
(Sales Minus Total Operating Costs)

Taxes (Federal and Local) -6

Income After Taxes -22
(Taxable Income Minus Taxes)

Total Annual Cash Outflow 702
(Total Operating Costs plus Taxes)

Net Cash Flow 107
(Sales Minus Annual Cash Outflow)

Discounted Cash Flow An@ftsis
10 years
Net Present Value (dollars 000's) at 10% -568

Internal rate of return (percent) -4.93%

'Percentage of total operating cost

10
18 % greater than the break even price. That farnis of similar sizes. The estimated IRR's
is, prices of P. vannamei can drop approxi- range from 0. 75 % by Hanson, et al. (1985)
mately 18% and the producers will continue to for a 40 ha facility to 4.5 % by Griffin, et al.
make a positive NPV. (1984). The IRR of 9.06% estimated by
Hollin and Griffin (1985) is with a survival
rate of 50 %, less than that assumed in this
study.
Alternative Scenarios
The estimated IRR for P. setiferus is less
P. setiferus is unprofitable (i.e. a negative than the assumed discount rate for the base
NPV) in all the alternative scenarios consid- scenario. Therefore, this operation fails to
ered in this analysis (Table 4). The profitabili- generate a positive return on equity. The
ty of P. vannamei is sensitive to changes in negative NPV may be improved if better
prices, stocking densities and survival rates. growth can be achieved and/or by developing a
specialty market for P. setiferus. Relatively
In scenario S4, 5 % of the shrimp are live high growth rates have been recorded for this
marketed. Discounted cash flow analysis for species in the wild or at relatively low densi-
P. vannamei and P. setiferus with 5 % of ties in ponds (Sandifer, et al. 1993 and WMC
production for live market sales and the re- unpublished data). Achievement of these
maining 95 % of production for final market growth rates in intensive pond culture will
sales were generated (Table 4). For P. vanna- depend on improved diets and/or rearing
mei, the ten-year IRR is 16.9%, which is procedures for P. setiferus. Live shrimp
greater than the base scenario after-tax dis- market segments for fishing bait are limited to
count rate of 10%. NPV in ten years is native species. Although wholesale bait
$321,000. The higher IRR and NPV in Sce- shrimp prices may be somewhat higher, there
nario S4, as compared to the base scenario is a potential risk of "flooding" this relatively
where 100% of the shrimp are for normal limited market. In addition, the lack of cost
market sales, indicates the potential higher effective storage facilities may hinder entrance
profitability of selling shrimp to "live-haulers". to these markets.
For P. setifew, IRR is 2.2 %, but the NPV is
still a negative -$320,000. Relative to the base The projected IRR for P. vannamei,
scenario, live harvesting does improve the 15. 1 %, is greater than the assumed base sce-
financial viability of a P. setiferus shrimp nario discount rate of 10%. The positive NPV
farm. However, the negative NPV still indi- for P. vannamei indicates that commercial
cates that this operation would not generate a shrimp farm in S.C. could have a positive
positive return on equity even if 5 % of the return to equity assuming specific pathogen
shrimps are used for live market sales. free PL are used. Hollin and Griffin (1985)
estimated an IRR of 22.5 % for a 40 acre farm
DISCUSSION with an 80% survival rate, a survival rate 10%
points higher than used in the base scenario in
Sandifer, et al. (1993) evaluated produc- this analysis. The NPV for both species may
tion data from Waddell Mariculture Center be improved by operating larger facilities,
(WMC) and concluded that P. setiferus may be integrating a nursery "headstart" operation
a viable alternative to P. vannamei for inten- with the farm or if price increases.
sive cultivation in the continental U.S. when
P. vannamei are unavailable. They recom- The profitability of Pacific white shrimp,
mended further evaluation of this potential. P. vannamei and the negative NPV of indige-
The preliminary financial analysis presented in nous P. setifenis have implications on regula-
this report suggests that a P. setifenis commer- tions that might affect the import of PL from
cial shrimp farm operation would only gener- out-of-state. Regulations that exclude P.
ate a negative IRR. This is less than the range vannamei as the target species could severely
of IRR estimated by other researchers for

11
Table 4. Discounted Cash Flow Analysis for the Alternative Scenarios for Hypothetical Penaeus
setiferus and Penaeus vannamei Shrimp Farm in South Carolina, 1993.

Scenarios P. setiferus P. vannamei
NPV IRR (0/,) NPV IRR

Scenario S -$416,000 -0.40% $232,000* 15.09%*

Scenario SIA -$843,000 -14.92% -$275,000 3.40%
Scenario SIB -$166,000 6.11% $516,000 20.84%

Scenario S2A -$568,000* 4.93%* $87,000 11-95%
Scenario S2B -$201,000 5.26% $438,000 19-31%

Scenario S3A -$600,000 -5.96% $55,000 11-25%
Scenario S3B -$236,000 4.40% $402,000 18-60%

Scenario S4 -$320,000 2.22% $321,000 16.94%

*BASE SCENARIO

Scenario S Stocking density of 80 PL/meter', aggregate survival rate of 70%, price of $4.95 / kg.

Scenario SIA Stocking density of 60 PL/meter, aggregate survival rate of 70%, price of $4.95 / kg for
P. vannamei and $4.73 / kg for P. setiferus

Scenario SIB Stocking density of 100 PL/meter, aggregate survival rate of 70%, price of $4.95 / kg
for P. vannamei and $4.73 / kg for P. setiferus

Scenario S2A Stocking density of 80 PL/meter, aggregate survival rate of 70%, price of $4.73 / kg.

Scenario S2B Stocking density of 80 PL/meter2, aggregate survival rate of 70%, price of $5.28 / kg.

Scenario S3A Stocking density of 80 PL/meter2, aggregate survival rate of 65%, price of $4.95 / kg.

Scenario S2B Stocking density of 80 PL/meter, aggregate survival rate of 75%, price of $4.95 / kg.

Scenario S4 Stocking density of 80 PL/meter, aggregate survival rate of 70%, continuous live market
sales of 5% and "normal" market sales of 95%.

NPV = Net present value after taxes.
IRR = Internal rate of return after taxes.
After-tax discount rate is 10%.

12
constrain the profitability of shrimp farming Griffin, W.L., W.E. Grant, R.W. Brick, and
using the techniques outlined in this analysis. J.S. Hanson. 1984. A bio-econornic
model of shrimp maricultaral systems
Continuous live market sales have positive in the U.S.A. Ecological Modelling,
effects on net cash flows for P. setiferus and 25(2):47-68.
P. vannamei. A 1993 survey conducted by
Rhodes, et al. (1994) suggests that opportuni- Griffin, W.L., and J.A.D. Lambregts. 1993.
ties for selling live shrimp probably exist The impact of aquaculture pond
throughout the major metropolitan areas of engineering design on the returns to
continental U.S., especially in the Northeastern shrimp farms. Journal of the World
states, but the lack of cost effective storage Aquaculture Society 24(l):23-30.
facilities may hinder entrance to these markets.
In addition, the live shrimp market for S.C. Hanson, J.S., W.L. Griffin, J.W. Richardson
producers seems relatively small. Unless addi- and C.J. Nixon. 1985. Economic
tional markets are identified, live harvesting of feasibility of shrimp farming in Texas:
large quantities of shrimps can have a negative An investment analysis for semi-inten-
impact on prices. Therefore, because the live sive pond grow-out. Journal of the
shrimp market is relatively small, it is irnpor- World Mariculture Society. 16:129-
tant that producers remain alert for potential 150.
"oversupply" conditions that might signal a
major reduction in prices, particularly for live Hollin, D. and W.L. Griffin. 1985. Prelimi-
market sales. nary economics of shrimp mariculture
in Texas. Presented at Texas Shrimp
This study confirms the findings in Farming Workshop, Corpus Christi,
Rhodes, et al. (1987) that the profitability of Texas.
S.C. marine shrimp operations is strongly
influenced by survival rates, stocking densities, Leung, P.S. and L.W. Rowland. 1989. Finan-
etc. The effects of these variables on the cial analysis of shrimp production: An
financial viability of producing P. setiferus and electronic spreadsheet model. Com-
P. vannamei were examined by developing puters and Electronics in Agriculture
alternative scenarios. The models used in this 3:287-304.
study are based on several simplifying assump-
tions, e.g., the land is leased. Nevertheless, McGovern-Hopkins, K., C.L. Browdy, J.D.
in general, the model did demonstrate potential Holloway and J.S. Hopkins. 1991.
species effects on the direction of changes in Penaeid shrimp hatchery system de-
profits, and to a lesser extent, the magnitude velopment for South Carolina. (Ab-
of such profits. Therefore, it may be fruitful stract) World Aquaculture Society
for prospective shrimp aquaculturists, inves- 22nd Annual Conf. & Expo., San
tors, and perhaps lending institutions and Juan, Puerto Rico, June 16-20, 1991.
policy makers to be aware of these findings.
Pardy, C.R., W. L. Griffin, M.A. Johns, and
A.L. Lawrence. 1983. A preliminary
LITERATURE CITED economic analysis of stocking strate-
gies for penaeid shrimp culture.
Adams, C. M., W.L. Griffin, J.P. Nichols Journal of the World Mariculture
and R.E. Brick. 1980. Application of a Society, 14:49-63.
bio-economic-engineering model for
shrimp mariculture systems. Southern
Journal of Agricultural Economics,
12(l):135-142.

13
Rhodes, R.J., P.A. Sandifer and J.M. Whet- Wyban, James A., J. S. Swingle, J. N.
stone. 1987. A preliminary financial Sweeny, and G. D. Pruder. 1992.
analysis of semi-intensive penaeid Development and commercial perfor-
shrimp farming in South Carolina. mance of high health shrimp using
S.C. Marine Res. Ctr., Tech. Rpt. No. specific pathogen free (SPF) brood-
64. stock Penaeus vannamei. In: J.A.
Wyban (Editor), Proceedings of the
Rhodes, R.J. 1991. Economics of Aqua- Special Session on Shrimp Farming,
culture production: Financial Feasibil- World Aquaculture Society, Baton
ity. In: J.A. Hargreaves and D.A. Rouge, LA.
Alston (Editors), Status and Potential
of Aquaculture in the Caribbean. Wyban, James A., J. S. Swingle, J.N.
World Aquaculture Society, Baton Sweeney, and G. D. Pruder. 1993.
Rouge, LA. Specific pathogen free Penaeus vanna-
mei. World Aquaculture 24(l):39-45.
Rhodes, R.J., K. McGovern-Hopkins, and
C.L. Browdy. 1992. Preliminary
financial feasibility analysis of an inde- APPENDIX 1
pendent marine shrimp hatchery locat-
ed in South Carolina. S.C. Marine SHRIMP HARVEST WEIGHT MODEL
Res. Ctr., Tech. Rpt. No. 80.
The empirical specification of the model is:
Rhodes, R.J., A. Johnson, C.L. Browdy, P.A.
Sandifer. 1994. Analyzing Live LWeight = bo + b, 4Age + b2 LDensity +
Seafood Marketing Opportunities in the b3 Species + b4 Disease + b. Year 87 + b6
United States: The South Carolina Year88 + b7 Year89 + b8 Year9O + bg
Experience with Live Shrimp. (Poster) Year9l blo Year92 + e .
Annual meetings of the World Aqua-
culture Society, New Orleans, LA, Where, LWeight = Log of average harvest
January 1994. weight in grams per week, LAge =Log of the
number of growout days in weeks; Disease
Rosenberry, B. 1993. (Editor) World Shrimp binary variable, equal to I if signs of viral
Farming: Special Report, Shrimp Far- infection were observed during the growout
ming In The United States. Aqua- cycle (0 otherwise); Species = binary variable,
culture Digest, San Diego, CA 92131. equal to one if species is P. vamamei (0 other-
wise); and LDensity = Log of stocking
Sandifer, P.A., J.S. Hopkins, A.D. Stokes density, number stocked/m@ Year87,...,Year92
and C.L. Browdy. 1993. Preliminary = binary variable, equal to one for that year,
comparisons of the native P. setiferus (0 otherwise).
and pacific P. vannamei white shrimp
for pond culture in South Carolina, There were a total of 100 observations
USA. Journal of the World Aqua- (ponds) in the data collected at Waddell Maric-
culture Society, 24(3):295-303. ulture Center during shrimp culture experi-
ments conducted from 1984 to 1992. Each
observation had 18 variables (Pond, year,
stock date, harvest date, crop length, stock
weight, harvest weight, growth rate, pond
size, species, stock source, disease status,
stocking density, number stocked, number
harvested, estimated survival rate, production,
and experimental comparison). The biweekly
sampling data were excluded from the

14
analysis. Among the 100 observations 45 Binary variables, one for each year
ponds satisfied the criteria similar to the best from 1987 to 1992 were included as a proxy
management practices recommended for S.C. for water temperature. To avoid problems
commercial shrimp farms. with multicollinearity, the year 1986 is omitted
and its effects are captured by the intercept.
The selected criteria are: a) survival All the YEAR variables shift the intercept by
rate greater than 60%; b) density between 20 the amount of the estimated coefficient.
and 100 per square meter; c) ponds were
stocked before June 1; d) ponds were harvest-
ed late September or October; e) all aeration
rates were included; f) all exchange rates were
included; g) ponds in which shellfish were cul-
tured with shrimps were included.

The estimated regression coefficients
are presented in Table 5. Significance of the
estimated coefficients were tested at 0.05 and
0. 1 levels. Three criterion were used to ex-
amine the overall "goodness" of fit of the
regression models. The high F-value rejects
the null hypothesis that all the regression
coefficients for the explanatory variables are
simultaneously equal to zero. The W and the
adjusted RI indicate that a large percentage of
the variation in the dependent variable are
explained by the independent variables includ-
ed in the model - The adjusted W takes into
account the number of explanatory variables in
relation to the number of observations. There-
fore, in a multiple regression model, the ad-
justed W is preferred to R.

As expected, age and species have a
positive effect on the final harvest weight of
shrimps. A 1 % increase in age will increase
the harvest weight by 0.52 %. Species, a
binary variable, with a value of 1 for P. vann-
amei and 0 otherwise, increases the intercept
for the regression model for P. vannamei.

Density and disease have a negative
regression coefficient. That is, as expected, at
higher densities the average harvest weight
would be smaller. The results show that a 1 %
increase in density will decrease the harvest
weight by 0. 16 % Disease, a binary variable,
with a value of 1 for post larvae with disease
and 0 otherwise, will reduce the intercept for
the model P. vannamei with disease by the
estimated coefficient.

Table 5. Regression Coefficients for the Harvest Weight Model','

Variable Coefficients
Name

Intercept 0.62*
Age 0.52*
Density -0. 16*
Species 0.13*
Disease -0.07**

Year87 -0.05

Year88 0.07

Year89 0.18

Year9O 0. 15*

Year9l 0. 12*

Year92 0.01

Variables significant at 0.05 level
Variables significant at 0.10 level

'Harvest weight model adjusted for commercial grow-out experience in Penaeus vannamei:
Lweight = 0. 85 + 0.52 LAge - 0. 16 L Density.
bPenaeus setiferus: Lweight 0. 75 + 0.52 LAge - 0. 16 L Density.

@er

low

411

Aaft

-2w

mo-iiia , W_

2@1

4i*

3 6668 00003 6451