Aquaculture Economics & Management
ISSN: 1365-7305 (Print) 1551-8663 (Online) Journal homepage: http://www.tandfonline.com/loi/uaqm20
Effects of regulations on technical efficiency of U.S.
baitfish and sportfish producers
Jonathan Van Senten, Madan M. Dey & Carole R. Engle
To cite this article: Jonathan Van Senten, Madan M. Dey & Carole R. Engle (2018): Effects of
regulations on technical efficiency of U.S. baitfish and sportfish producers, Aquaculture Economics
& Management
To link to this article: https://doi.org/10.1080/13657305.2018.1454539
Published online: 20 Jun 2018.
Submit your article to this journal
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=uaqm20
AQUACULTURE ECONOMICS & MANAGEMENT
https://doi.org/10.1080/13657305.2018.1454539
REGULAR RESEARCH ARTICLE
Effects of regulations on technical efficiency of
U.S. baitfish and sportfish producers
Jonathan Van Sentena
, Madan M. Deyb, and Carole R. Englec
a
Virginia Seafood AREC, Virginia Polytechnic Institute and State University, Hampton, VA, USA;
Department of Agriculture, Texas State University, San Marcos, Texas, USA; cEngle-Stone
Aquatic $LLC, Strasburg, VA, USA
b
ABSTRACT
KEYWORDS
The stringency of the regulatory environment has been shown
to negatively affect the growth of aquaculture. A technical efficiency analysis of baitfish/sportfish production in the United
States was performed using a stochastic production frontier
model and a jointly estimated maximum-likelihood procedure
(Frontier 4.1). Determinants of inefficiency were assessed for
their relationship to farm efficiency. Mean technical efficiency
for U.S. baitfish and sportfish producers was found to be 77%.
Several regulatory variables were found to be significant in
explaining the variation in levels of efficiency, including the
number of annual renewals of permits and licenses and the
amount of manpower required to comply with regulations.
Results support the hypothesis that the current regulatory
environment in the United States has reduced efficiency and
economic competitiveness of baitfish and sportfish producers.
Policy analysis; production
economics; United States
Introduction
Over the last decade, aquaculture in the United States has experienced an
overall decline, driven by rising feed prices, increased pressure from imports,
and a complex regulatory environment (Engle and Stone, 2013). A growing
body of literature has provided evidence that the regulatory environment can
result in excessive costs to aquaculture producers (Abate Nielsen, & Tveterås,
2016; Dresdner & Estay, 2016). More specifically, a recent survey of baitfish
and sportfish producers in the United States found that the average annual
cost of regulations was $150,000 per farm that represented costs of $7400 per
ha and 25% of total annual costs (van Senten & Engle, 2017). National industrywide regulatory cost for baitfish and sportfish was estimated to be in
excess of $12 million per year. Moreover, substantial restrictions in access to
markets were documented that were attributed directly to the regulatory
CONTACT Jonathan van Senten
jvansenten@vt.edu
Virginia Seafood AREC, Virginia Polytechnic Institute
and State University, 102 S. King Street, Hampton, VA 23669, USA.
ß 2018 Taylor & Francis
2
J. V. SENTEN ET AL.
environment; more than 60% of respondents from that study reported lost or
foregone sales as a result of regulations or regulatory complexity.
The U.S. baitfish and sportfish industry satisfies recreational demand for
anglers and pond owners by producing a variety of fish, including golden
shiners (Notemigonus crysoleucas), fathead minnows (Pimephales promelas),
goldfish (Carassius auratus), largemouth bass (Micropterus salmoides), smallmouth bass (Micropterus dolomieu), crappie (Pomoxis sp.) and other
Centrarchid species (Stone & Thomforde, 2001; USDA, 2014). The 2013 U.S.
Census of Aquaculture reported 166 baitfish producers and 282 sportfish
producers, with sales in excess of $52 million (USDA, 2014). While farm size
is highly variable across the sector, extensive pond production is the primary
method of production. This means that climate and season are important
factors for production and directly affect supply and demand of baitfish and
sportfish in many regions of the United States. Interstate transport of live
fish is a common practice within the industry, and necessary to reach wholesalers, retailers, and end users across the country. However, interstate transport requires compliance with various regulations, mostly focused on aquatic
animal health. The exact number of enforcement agencies regulating the permitting and licensure of baitfish and sportfish varies by state, as do the number of permits and licenses, and the specific requirements of those permits
and licenses. Most states require a permit or license to propagate fish; in
some cases, specific permits or licenses are required for specific species of
fish. In most cases, a certificate of health is required for fish to be shipped
across state lines, or in some cases between watersheds. The major cost of
these permits and licenses is not the direct cost of the permits, but rather the
manpower involved in maintaining compliance (van Senten & Engle, 2017).
Manpower on the farm is tasked with identifying which regulations are relevant in each state, how to obtain those licenses or permits, completing compliance activities to be eligible for those licenses and permits, completing
applications for licenses and permits, and filing records of licenses and permits once obtained. Reporting requirements for permits and licenses also
vary by state, with some states requiring all parties involved in the process to
maintain a copy of the documentation on hand. The number of annual permit and license renewals for baitfish and sportfish producers average 13, as
reported by van Senten and Engle (2017); however, individual farms reported
as many as 203.
Beyond simply increasing costs for producers, there is reason to suspect
that the current regulatory environment may be affecting the efficiency of
baitfish and sportfish producers. The study by van Senten and Engle (2017)
demonstrated that regulations primarily increased annual fixed costs but
because the regulatory environment also restricted access to markets, farmers were not able to increase the scale of their operation to spread the
AQUACULTURE ECONOMICS & MANAGEMENT
3
increased fixed costs over greater volumes of production. Asche and Roll
(2013), using a stochastic frontier approach, found that improvements in
governmental regulations resulted in more permanent improvements in
efficiency of Norwegian salmon producers. These recent studies contrast
with work completed more than 23 years ago (Thunberg Adams, & Cichra,
1994) that concluded that financial and marketing barriers were more problematic in the state of Florida than were regulatory barriers. Thunberg et al.
(1994) did not consider baitfish and sportfish in their analysis; however,
reports show that the regulatory burden in the United States has increased
over time (Crews, 2017). However, no quantitative analysis has been found
in the literature to quantitatively test whether the regulatory environment
does affect technical efficiency on aquaculture farms or which types of
regulatory metrics have significant effects.
A number of factors have been identified that affect technical efficiency of
agriculture in general. For example, the greater scale and scope of farms in the
U.S. Corn Belt were found to increase farm efficiencies and decrease competitiveness of small family farms (Paul, Nehring, Banker, & Somwaru, 2004;
Fengxia, Hennessy, Jensen, & Volpe, 2016). Education and outmigration
(Sauer, Gorton, & Davidova, 2015) have also been found to reduce technical
efficiency, while efficiency-enhancing investments (Mekonnen, Spielman,
Fonsah, & Dorfman, 2015) have been found to increase technical efficiency.
More specific to aquaculture, age, experience and education have frequently been found to significantly affect the technical efficiency of aquaculture farms, although whether the effect is positive or negative varied
among reviewed studies (Iliyasu et al., 2014). A variety of other variables,
such as farm size appear as significant in some studies (Arita & Leung,
2014) but were not significant in others (Iliyasu et al., 2014). In more
recent studies, Sandvold (2016) found that older producers were somewhat
more efficient at producing salmon smolts in Norway, and water use was
found to affect efficiency of catfish and red tilapia production in tanks in
Malaysia (Iliyasu & Mohammad, 2015).
A high level of technical efficiency, producing the maximum output for a
given level of inputs, is important for farms to be both competitive and profnoz, & Br€
ummer, 2017).
itable (Murova & Chidmi, 2013; Lakner, Brenes-Mu~
Recent research has suggested that the regulatory environment may reduce
farm-level competitiveness of aquaculture. For example, based on an analysis
of 95 developed and developing countries, Abate et al. (2016) showed that
the stringency of environmental regulations was negatively associated with
the rate of growth of aquaculture. A number of studies have concluded that
the fragmented nature of regulations for aquaculture and the resulting redundancies need to be addressed in developed countries (Engle & Stone, 2013;
Kite-Powell, Rubino, & Morehead, 2013; Abate et al., 2016; Engle, 2016;
4
J. V. SENTEN ET AL.
Knapp & Rubino, 2016). On the other hand, Rahman, Hatha, Selvam, &
Thomas (2016) discussed the importance of increasing the stringency and
enforcement of regulations in developing countries. For example, numerous
studies have shown that weak enforcement systems in countries such as
China and Vietnam contribute to continued widespread use of antibiotics
that are banned from use in livestock feeds in the U.S. (Broughton &
Walker, 2010; Rico et al., 2012, 2013).
It should be noted that concerns related to the regulatory environment in
the U.S. and the European Union (EU) do not stem from a desire to eliminate all regulations. Regulations are necessary to internalize production
externalities to maximize social welfare. However, there is increasing
evidence of inefficiencies in the regulatory environment specifically related
to aquaculture that result from redundancy, unnecessary duplication,
and overlap (van Senten & Engle, 2017; Osmundsen, Almklov, &
Tveterås, 2017).
The effect of regulations on technical efficiency in agriculture generally is
not well defined. For example, neither Paul, Johnston, and Frengley (2000)
or Yang, Hsiao and Yu (2008) found significant effects from regulations on
technical efficiency of sheep and beef farms in New Zealand and swine
farms in Taiwan, respectively. On the other hand, van der Vlist, Withagen
and Folmer (2007) reported results confirming Porter’s hypothesis (1995)
that stricter environmental policy resulted in more technically efficient
Dutch horticulture farms.
This study contributes to previous technical efficiency studies by estimating the technical efficiency of U.S. baitfish and sportfish producers, with
specific attention to whether variables that reflected regulatory compliance
on farms had significant effects on baitfish/sportfish production. Specific
objectives include: (1) to estimate technical efficiency on U.S. baitfish and
sportfish farms; (2) to identify whether there are regulatory variables that
contribute significantly to farm efficiencies; and (3) to estimate potential
effects on farm efficiencies of streamlined regulatory processes.
This paper proceeds by first describing the survey used to collect data
and the variables of interest and then presenting the production and inefficiency functions of the stochastic frontier model used. The empirical specification and sensitivity analysis are then described, followed by results,
discussion, and conclusions.
Materials and methods
Data and variables
A survey was conducted in 2015 to capture data on farm production, costs,
and the type, nature, and compliance costs of the total set of regulations that
AQUACULTURE ECONOMICS & MANAGEMENT
5
affect U.S. baitfish and sportfish producers (van Senten & Engle, 2017). The
survey targeted the 13 major baitfish and sportfish producing states in the
U.S.; budget limitations precluded a national survey. This targeted population
composed 81% of the total volume of production (USDA, 2014). The survey
was conducted as a census; thus, the data reported are population and not
sampling data. The survey was designed to capture line-item quantities and
costs of farm inputs, farm production volumes and sales, general farm characteristics, and regulatory compliance activities and costs.
Survey reliability was evaluated through the test-retest method and validity through consultations with experts and pre-testing of the survey instrument (Litwin, 1995). The survey responses covered 74% of the national
production volume with responses from 34% of the baitfish/sportfish farms.
Detailed information on regulations that affected U.S. baitfish/sportfish
farms can be found in van Senten and Engle (2017). Survey data were
coded and entered into a spreadsheet (Microsoft ExcelV) prior to transformation for efficiency analysis using Frontier 4.1 (Coelli, 2011).
Since the baitfish/sportfish industry is composed of a variety of species, production systems, and marketing strategies, production data (both
inputs and outputs) were recorded in U.S. dollar values. As an example
of the variation in the way various baitfish/sportfish farms measured output, some farms measured production by weight (pounds) while others
used inches or “head” of fish (individual units). While it is customary
for production functions to be expressed in terms of quantities, it is
possible to utilize values as a proxy for quantities (Grieco, Li, & Zhang
2016). In this case inputs were sourced from a competitive market where
baitfish and sportfish producers are price takers and subject to the same
prices for given inputs (Grieco et al., 2016). Thus, value is a proxy for
quantity in this analysis. A description of the variables used and their
mean values are presented in Table 1.
There were a total of 60 observations (farms) in the dataset, with over
20 production input categories. All data (output and inputs) were transformed to a per-hectare (ha) value and the natural log taken. Stata 11
(StataCorp, 2009) was used to test for homoscedasticity (RV plot and
Breusch-Pagan test), multicollinearity (VIF test), model specification
(Lowess test), and normality (Kdensity, Pnorm, Qnorm, and
Shapiro–Wilk test).
R
Stochastic Frontier Model
The stochastic frontier production model (Aigner, Lovell, & Schmidt, 1977)
can be expressed as:
lnYi ¼ f ðb; Xi Þ þ e
(1)
6
J. V. SENTEN ET AL.
Table 1. Definition of variables and average values.
Parameter
Symbol/Coefficient
Description
Average value
y
Total sales (output)
$878,136
Feed
b1
Feed used in production
$84,413
Amm
b2
$28,648
Rep
b3
Ins
b4 =d6
Fertilizer, herbicides, pesticides, disinfectant, and
other chemicals (amendments)
Repairs and maintenance of
facilities, equipment, and
infrastructure
Insurance for the
farm business
Total labor for the farm
production (minus the cost
of manpower for regulatory compliance)
Other production activities:
pumping water, electricity,
interest on loans,
office expenses
Total cost of regulatory compliance
16.11
$16,008
Production function
Sales
$25,000
$31,321
Labor
b5
Other
b6
RegCost
b7
Inefficiency function
Regulatory variables
Renew
d10
ChangeR
d3
Number of annual permit/
license renewals
Changes due to regulations
Fhealth
d4
Fish health activities
$7,250
Insurance for the
farm business
Lost/foregone sales due to
regulations
Manpower to comply with
regulations
Number of state regulations
$31,321
Ins
b4 =d6
Lost
d2
ManR
d5
Nstate
d11
Nfed
d12
Nship
d9
Farm characteristic variables
d1
Size
Bait
d7
Sport
d8
Glregion
d13
Seregion
d14
$222,709
$54,218
$148,554
$85,039
$15,948
7.12
Number of federal
regulations
Number of states
shipped to
1.18
Farm size (ha)
156.36
Raising baitfish only
(dummy variable)
Raising sportfish only
(dummy variable)
Farm is located in the Great
Lakes region
(dummy variable)
Farm is located in the
Southeast region
(dummy variable)
0.25
9.70
0.38
0.40
0.15
where lnYi represents the stochastic frontier for observation i, b is the parameter estimate of Xi , Xi represents the explanatory variable, i, (input i),
and e is the error term. The error term (e) consists of two components; Vi
AQUACULTURE ECONOMICS & MANAGEMENT
7
a firm specific white-noise stochastic error and Ui the firm specific inefficiency (Asche & Roll, 2013). The frontier production function represents
the case where there is no inefficiency in the model (Ui ¼ 0) (Coelli, Rao,
O’Donnell, & Battese, 2005). The gap between the stochastic frontier and a
firm’s observed production output, then, represents that specific firm’s inefficiency. Technical efficiency is commonly represented as a ratio (Coelli
et al., 2005), where 1 represents complete efficiency and 0 complete inefficiency. Expressed in terms of inputs Xi and firm specific output Yi as
(Murova & Chidmi, 2013); technical efficiency is:
TE ¼
Yi
f ðbXi Ui Þ
¼
f ðbXi Þ
f ðbXi Þ
(2)
Assuming a half-normal distribution of the error term allows for a more
simplified approach to the determinants of inefficiency (Asche & Roll,
2013). A separate function represents the inefficiency of each firm (r2), as a
function of explanatory variables Zi, which may be the same as Xi from the
stochastic frontier function (Asche & Roll, 2013):
r2i ¼ expðZi ; WÞ
(3)
where W is a vector of those same explanatory variables (Zi).
Battese and Coelli (1995) argued that two-stage estimation of efficiency
was inconsistent in its assumption of the independence of inefficiency
effects. Therefore, they developed a model to perform a joint estimation of
the firm-level efficiencies and regress those predicted efficiencies over firmspecific variables (Coelli, 2011):
Yit ¼ bXit þ ðVit Uit Þ
i ¼ 1; :::N; t ¼ 1; :::T
(4)
where Yit is the dependent variable, Xit the explanatory variable, b is the parameter estimate of Xit explanatory variable, Vit are random variables assumed to be
independent and identically distributed N(0, rv2), and Uit are non-negative random variables assumed to account for technical inefficiency (Coelli, 2011). The
Battese and Coelli specification also assumes that Uit is independently distributed as truncations at zero of N(mit, ru2), where (Coelli, 2011):
mit ¼ zit d
(5)
where Zit is a vector of the variables influencing inefficiency and d is a vector
of the parameters to be estimated (Coelli, 2011). Through replacing rv2 and
ru2:
r2 ¼ r2V þ r2U and c ¼ r2U =ðr2V þ r2U Þ
(6)
this model can be estimated through a maximum-likelihood procedure
(Coelli, 2011). The Frontier 4.1 program utilizes a three-step estimation
8
J. V. SENTEN ET AL.
procedure to obtain maximum likelihood estimates of the parameters in the
stochastic frontier production function as follows (Coelli, 2011):
1. an ordinary least squares (OLS) estimate of the function is performed
where, with the exception of the intercept, all parameter estimates (b)
are unbiased;
2. a grid search of c with parameters (b) set to the OLS estimates; and
3. an iterative maximization procedure using first-order partial derivatives
to obtain final maximum likelihood estimates (Davidon–Fletcher–Powell
Quasi-Newton method).
Empirical specification
A number of production variables were consolidated to correct for issues of
multicollinearity. For example, fertilizer, herbicides, disinfectants, and
chemicals were grouped into a variable named “amendments.” Similarly,
on-farm energy inputs (electricity and pumping water), fuel inputs, and
other miscellaneous inputs were grouped into an “other” variable. The final
stochastic frontier production model was:
lnSales ¼ b0 þ b1 lnFeed i þ b2 lnAmmi þ b3 Repi
þ b4 lnInsi þ b5 lnLabori þ b6 lnOtheri
þ b7 lnRegCosti þ ðVi Ui Þ
(7)
where bi is the parameter estimate of explanatory variables of feed
(lnFeedi), amendments (lnAmmi), repairs and maintenance (lnRepi), insurance (lnInsi), labor (lbLabori), other miscellaneous inputs (lnOtheri), and
regulatory costs (lnRegCosti) expressed as natural logs. Vi are random variables assumed to be independent and identically distributed N(0, rv2), and
Ui are non-negative random variables assumed to account for technical
inefficiency (Coelli, 2011). A likelihood ratio test value of 59 exceeded the
upper bound Wald criteria for joint testing of the equality and inequality
restrictions at the 5% level (Kodde & Palm, 1986), and therefore the null
hypothesis of a trans-log model over a Cobb-Douglas model was rejected;
the Cobb-Douglas specification was used for the analysis.
The inefficiency function focused on descriptive and regulatory cost variables obtained from the survey. These regulatory variables were selected
based on findings from van Senten and Engle (2017) who identified manpower to comply with regulations, fish health testing, changes due to regulations, and lost and foregone sales as substantial elements of the regulatory
compliance cost on farms. Lost and foregone sales contributed the largest
portion of regulatory cost (57%), followed by changes due to regulations
(22%), manpower to comply with regulations (11%), and fish health testing
costs (5%). It should be noted that the costs of farm-level changes due to
AQUACULTURE ECONOMICS & MANAGEMENT
9
Table 2. Hypothesis testing for stochastic frontier and inefficiency parameters.
Null hypothesis
H0:
H0:
H0:
H0:
H0:
H0:
H0:
c ¼ d0 ¼ d1 ¼ … ..¼ d14 ¼ 0
c¼0
d0 ¼ d1 ¼ … ..¼ d14 ¼ 0
d0 ¼ d2 ¼ d3 ¼ d4 ¼ d5 ¼ d9 ¼ d10 ¼ d11 ¼ d12 ¼ 0
d0 ¼ d2 ¼ d3 ¼ 0
d0 ¼ d5 ¼ d10 ¼ 0
b1 þ b2 ¼ þ … ..þ b7 ¼ 1
Log likelihood statistic
25.22
39.49
59.18
58.44
43.05
58.33
6547.85
Critical value
Decision
5.14 –24.384
5.14–32.07
24.99
16.92
7.82
7.82
3.84
Rejected
Rejected
Rejected
Rejected
Rejected
Rejected
Not rejected
regulations did not include permit/license renewals or manpower; only the
costs of implementing the change, to avoid double counting regulatoryinduced costs. While the direct cost of permits and licenses may only have
accounted for 1% of total regulatory costs on average, the relationship
between the number of annual license and permit renewals (in excess of 100
in some cases) and the manpower compliance costs of producers was also
noted. Hence, these regulatory variables were believed to be likely contributors to inefficiency on farms and were included in the inefficiency function.
In addition to these variables, dummy variables for the type of fish raised on
farms and the region in which farms were located were added to the function. The function as entered into Frontier 4.1 was specified as:
mi ¼ d0 þ d1 Sizei þ d2 Losti þ d3 ChangeRi þ d4 FHealthi
þ d5 ManRi þ d6 Insi þ d7 Baiti þ d8 Sporti þ d9 NShipi þ d10 Renewi
þ d11 NStatei þ d12 NFedi þ d13 GlRegioni þ d14 SeRegioni þ ei
(8)
where mi is the mean of the non-negative random variables (Ui) assumed
to account for technical inefficiency, di is the parameter estimate of farm
size (Sizei), lost/foregone sales (Losti), changes due to regulation (ChangeRi),
fish health activities (FHealthi), manpower cost of compliance (ManRi),
insurance for the farm (Insi), dummy variable for raising baitfish only
(Baiti), dummy variable for raising sportfish only (Sporti), number of states
shipped to (NShipi), number of annual permit/license renewals (Renewi),
number of state regulations (NStatei), number of federal regulations
(NFedi), dummy variable for the Great Lakes region (GlRegioni), dummy
variable for the Southeast region (SeRegioni), and ei is the error term. The
base scenario modelled was for a farm raising both baitfish and sportfish in
the South Central region. The technical efficiency of the ith sample farm
ðTEi Þ is obtained as TEi ¼ expð Ui Þ:
Hypothesis testing
Three hypotheses related to the validity of the estimates developed were
tested with a generalized likelihood ratio test (Hassan & Ahmad, 2005)
10
J. V. SENTEN ET AL.
Table 3. Average efficiency scores by state/grouping.
State/Grouping
Alabama
Arkansas
New York
North Carolina
Ohio
Pennsylvania
Wisconsin
Other States
All Observations
Florida, Illinois,
“Other States”.
Average efficiency score
Min
Max
0.929
0.822
0.795
0.927
0.553
0.462
0.925
0.865
0.766
0.833
0.431
0.621
0.905
0.008
0.181
0.909
0.596
0.008
0.967
0.955
0.939
0.959
0.934
0.765
0.955
0.961
0.967
included
under
Kansas,
and
Texas
are
(Table 2). The first tested for the absence of inefficiency effects from the
model as follows:
H0: c ¼ d0 ¼ d1 ¼ d2 ¼ d3 ¼ d4 ¼ d5 ¼ d6 ¼ d7 ¼ d8 ¼ d9 ¼ d10 ¼ d11 ¼ d12
¼ d13 ¼ d14 ¼ 0.
Secondly, the inefficiency effects of the model were tested for not being
stochastic as follows:
H0: c ¼ 0.
The third hypothesis tested was to determine if the 14 variables used in
the inefficiency equation (11), had a significant effect on the inefficiency,
with the following null hypothesis:
H0: d0 ¼ d1 ¼ d2 ¼ d3 ¼ d4 ¼ d5 ¼ d6 ¼ d7 ¼ d8 ¼ d9 ¼ d10 ¼ d11 ¼ d12
¼ d13 ¼ d14 ¼ 0.
A hypothesis of whether regulatory variables in the inefficiency function
contributed to technical efficiency of the model was tested with the following null hypothesis:
H0: d0 ¼ d2 ¼ d3 ¼ d4 ¼ d5 ¼ d9 ¼ d10 ¼ d11 ¼ d12 ¼ 0.
Separate hypothesis tests were performed on the effect of the regulatory
variables that were significant on technical efficiency with the following
null hypotheses:
H0: d0 ¼ d5 ¼ d10 ¼ 0.
Positive coefficients (number of annual permit/license renewals and manpower to comply with regulations)
H0: d0 ¼ d2 ¼ d3 ¼ 0.
Negative coefficients (changes due to regulations and lost/foregone sales)
Constant returns to scale were tested with the following hypothesis:
H0: b1 þ b2 þ b3 þ b4þ b5 þ b 6þ b7 ¼ 1.
Results
Technical efficiency estimates were obtained for all 60 farms, with an average efficiency of 77% across all participants (Table 3). In order to preserve
confidentiality, efficiency estimates presented in Table 3 are summarized by
AQUACULTURE ECONOMICS & MANAGEMENT
11
Table 4. Parameter estimates for jointly estimated stochastic frontier model.
Parameter
Production function
Intercept (b0 )
Amendments
Repairs and maintenance
Labor
Feed
Insurance
Regulatory cost
Other
Inefficiency function
Intercept (d0 )
Regulatory variables
Annual renewals of permits/licenses
Changes due to regulation
Fish health regulations
Insurance
Lost/foregone sales
Manpower to comply with regulations
Number of federal regulations
Number of state regulations
Number of states shipped to
Farm characteristic variables
Farm size
Baitfish only (dummy)
Sportfish only (dummy)
Great Lakes region
Southeast region
r2
c
Coefficient estimate
Standard error
T-ratio
3.166
0.126
0.333
0.211
0.174
0.178
0.018
0.153
0.719
0.061
0.086
0.088
0.140
0.102
0.060
0.043
4.403
2.048
3.893
2.406
1.245
1.756
0.296
3.547
3.216
0.856
3.756
0.007
0.187
0.093
0.640
0.099
0.220
0.061
0.043
0.005
0.003
0.057
0.066
0.176
0.046
0.082
0.095
0.032
0.017
2.339
3.289
1.414
3.635
2.178
2.669
0.643
1.334
0.281
0.002
0.109
0.753
0.121
1.658
0.255
0.365
0.001
0.557
0.382
0.515
0.924
0.070
0.189
1.918
0.196
1.969
0.235
1.795
3.647
1.930
Significant (p <0.05)
state (with Florida, Illinois, Kansas, and Texas included under
“Other States”).
The null hypothesis for the first hypothesis test related to the validity
of the estimates developed (H0: c ¼ d0 ¼ d1 ¼ d2 ¼ d3 ¼ d4 ¼ d5 ¼ d6 ¼ d7 ¼ d8
¼ d9 ¼ d10 ¼ d11 ¼ d12 ¼ d13 ¼ d14 ¼ 0) was rejected (test statistic of 25.22;
critical value range of 5.14 to 24.38), indicating that the model contains
inefficiency effects. The null hypothesis for the second null hypothesis
related to the validity of estimates developed (H0: c ¼ 0) was also rejected
(test statistic of 39.49 exceeding the critical value range of 5.14 to 32.07).
Therefore, the technical inefficiency effects of the model are understood to
be random (Hassan & Ahmad, 2005). The third hypothesis tested related to
the validity of the estimates developed (H0: d0 ¼ d1 ¼ d2 ¼ d3 ¼ d4 ¼ d5
¼ d6 ¼ d7 ¼ d8 ¼ d9 ¼ d10 ¼ d11 ¼ d12 ¼ d13 ¼ d14 ¼ 0) was rejected (exceeded
the critical value of 24.99). Thus, the 14 variables specified were found to
have an effect on technical inefficiency.
The parameter estimates for each variable from the maximum-likelihood
estimation procedure are listed in Table 4. In the stochastic production
function (8) estimated, the b coefficients reveal that “amendments”, “repairs
and maintenance”, “labor”, and “other” (included costs to pump water,
electricity, and interest expenses on loans), variables were significant
(asymptotic absolute value t-ratio >1.96). The negative coefficient for the
12
J. V. SENTEN ET AL.
“amendments” variable indicated that production output decreased as
amendment usage increased. Variables denoting feed, insurance, and regulatory costs were not significant.
In the inefficiency function, the following variables were significant
(asymptotic absolute value t-ratio >1.96): “permit/license renewals”,
“changes due to regulations”, “insurance”, “lost/foregone sales”, and
“manpower to comply with regulations”. Of the farm characteristic variables, only the dummy variable for “Sportfish” was significant.
The hypothesis of whether regulatory variables in the inefficiency function contributed to technical efficiency of the model was rejected (test statistic of 58.44 exceeded the critical value of 16.92), and the regulatory
variables specified were found to contribute to technical inefficiency in the
model. Both null hypotheses tested on the effect of the regulatory variables
that were significant on technical efficiency were rejected, suggesting that
these respective variables (positive coefficients and negative coefficients) did
have an effect on technical inefficiency of baitfish and sportfish producers.
The null hypothesis for constant returns to scale was not rejected (test statistic of 6,547.85, lower than the critical value of 3.84), indicating that the
model as specified exhibited constant returns to scale.
Discussion
Production function
The base scenario modelled by the estimation was for a farm raising both
baitfish and sportfish in the U.S. South Central region; with dummy variables accounting for farms that raised only baitfish or only sportfish, and
farms located in the Great Lakes region and Southeast region. Results of
the stochastic production frontier estimation for U.S. baitfish/sportfish
demonstrated that feed was not a significant input for baitfish and sportfish
producers in this study in contrast with foodfish production (Lacewell,
Nichols, & Jambers, 1973; Nerrie, Engle, Hatch, & Smitherman, 1990;
Losinger, Dasgupta, Engle, & Wagner, 2007; Engle, 2010). Baitfish and
sportfish producers more commonly use fertilizers to promote algal blooms
and natural food with supplemental, rather than intensive feeding of commercial diets. In some instances, producers rearing both baitfish and sportfish have used surplus baitfish as forage to supplement feeding of sportfish.
Indeed, the “amendments” variable that included fertilizer, was significant.
The negative coefficient on this variable may reflect the need to fertilize at
higher rates in ponds with water with high calcium hardness (Brunson,
Hargreaves, & Stone, 2000), typical of the quality of water found on many
baitfish/sportfish farms in Arkansas, where the greatest amount of baitfish/
sportfish are raised in the U.S. The increasing difficulty of controlling
AQUACULTURE ECONOMICS & MANAGEMENT
13
problematic aquatic vegetation in older ponds, such as those used for baitfish and sportfish production may result in weedy ponds that tend to have
less phytoplankton and zooplankton for baitfish and sportfish to feed on
(Jones et al., 2016).
The variables “repairs and maintenance”, “labor”, and “other” inputs of
production were also significant in the production function, with positive
coefficients. Thus, as the value of repairs and maintenance and labor
increased, so did production output. This result aligns well with common
knowledge of the management of many aquaculture businesses, in which
labor, fuel, and costs to pump water, among others are important costs of
aquaculture businesses. Given the relatively high level of fixed costs as a
percentage of total annual costs in baitfish and sportfish production, repair
and maintenance costs would be expected to be relatively higher to maintain intermediate and long-term assets such as equipment, buildings, and
pond infrastructure. Labor is a major cost of production in baitfish/sportfish (Stone, Kelly, & Roy, 2016) as is the combined costs of pumping, electricity, and interest costs included in the “other” category.
The regulatory cost variable was not found to be significant in the production function, signifying that the cost of regulations did not significantly
explain the quantity of baitfish and sportfish produced. While the overall
cost of compliance was found to be a major economic burden on producers, van Senten and Engle (2017) demonstrated that 60% of the total
regulatory cost was the opportunity cost of lost or foregone sales. The direct cost of regulations, as measured in the survey, is composed of the direct
costs and fees of the permits and licenses. The costs of the permits and
licenses required for U.S. baitfish/sportfish was found to compose only 1%
of the total regulatory cost. A similar finding, that the cost of permits and
licenses composed only a very small part of the total farm-level cost of regulations, was previously reported by Hurley and Noel (2006) for California
agricultural producers.
Inefficiency function
In the inefficiency function, several regulatory variables significantly contributed to farm inefficiency, including: (1) the number of permit/license
renewals each year; (2) manpower to comply with regulations; (3) changes
due to regulations; and (4) lost/foregone sales. The insurance variable,
although not a regulatory variable, was also significant. Other regulatory
variables such as the numbers of state and federal regulations, the number
of states shipped to, farm size, and fish health regulations were not found
to contribute significantly to farm efficiency.
14
J. V. SENTEN ET AL.
The positive coefficient for the “number of annual permit and license
renewals” in the inefficiency equation, was significant. Survey respondents
reported a high number of annual renewals of permits and licenses. Onethird of respondents reported between 10 and 203 annual renewals a year.
While the majority of these permits and licenses (90%) were obtained from
state agencies, approximately 33% of the state regulations had been developed
to enforce federal statutes. Respondents reported being required to annually
renew multiple permits from multiple agencies, sometimes within the same
states, that request the same information. The significance of the number of
annual renewals indicates the complexity of the regulatory environment.
Manpower spent to comply with regulations was also found to be a significant determinant of inefficiency on farms. Thus, as more manpower was
used to comply with regulations, the technical efficiency of the farm was
reduced. This is likely a reflection of the fact that diverting manpower
towards regulatory compliance activities reduces the time available for
productivity-enhancing innovations or to develop new markets and results
in a reduction in production output.
Other studies have reported negative effects on U.S. aquaculture farms
that result from the time and manpower required to comply with the complexity created by the total set of regulations with which aquaculture businesses contend (Engle & Stone, 2013; Kite-Powell et al., 2013). Often, it is
the farm owner or manager who must spend time to ensure that the business is fully compliant with provisions, monitoring, and compliance with
all required permits and licenses. The overlapping and redundant nature of
many of the state regulations faced by baitfish/sportfish producers in the
U.S., who ship most of their product to other states, has been cited in existing literature as a major part of the regulatory burden on aquaculture businesses in the U.S. (Engle & Stone, 2013; Kite-Powell et al., 2013). As
revealed by the data in this study, significant manpower is required at the
farm level to attend to these overlapping and redundant regulatory requirements, which very often have different reporting formats and deadlines. In
other types of agriculture, labor productivity was found to be a significant
determinant of technical efficiency, for example, in the Greek food and beverage market (Rezitis & Kalantzi, 2016). Hurley (2004) reported a 40%
increase in manpower time spent by agricultural producers in California on
regulatory compliance, with the manpower costs constituting the greatest
cost of regulatory compliance. In the data used for the present baitfish and
sportfish study, the cost of manpower to comply with regulations was only
11% of the total regulatory cost (van Senten & Engle, 2017); but it was
noted that this value was likely underestimated. Producer respondents in
the survey did not keep records of the time spent on every telephone call
or contact to request application forms, to identify changes in regulations
AQUACULTURE ECONOMICS & MANAGEMENT
15
from the previous year, or in a number of cases, to identify the appropriate
individual in charge of the permit for that year.
The variable “insurance” was also found to be a significant determinant
of inefficiency. The survey results documented relatively high costs associated with a variety of insurance products on baitfish and sportfish farms,
which is why insurance was included as a variable in the analysis even
though it was not a regulatory cost, nor a requirement for producers.
Insurance is a management option to reduce various types of financial risks
in a farm business. The significance of insurance in the model likely reflects
attempts by farmers to manage the risk associated with baitfish and sportfish businesses and may reflect a greater degree of risk associated with baitfish/sportfish production as compared to foodfish production.
Lost or foregone sales were significant but with a negative sign on the
coefficient. While a literal interpretation would indicate that lost sales
increased farm efficiency, the data show that a number of respondents
reported few or no foregone sales; moreover, farms reporting higher quantities of foregone sales, were more efficient on average. It appears that farms
that operated at high levels of technical efficiency also had better records,
including those of sales that had been lost due to regulatory changes or
increased complexities. The importance of foregone sales as an unintended
consequence of regulatory compliance requirements was also identified by
Dresdner and Estay (2016) in their assessment of potential effects on the
Chilean salmon industry of various biosecurity measures proposed in
response to disease outbreaks. To avoid substantial foregone sales that would
have been based on production limits to Chilean salmon farms, Dresdner
and Estay (2016) proposed an approach that would result in estimates of
optimal levels of biosecurity regulation, given both the need for disease control and the need to avoid excessively costly levels of production limits that
would restrict sales of salmon.
The cost of changes on the farm due to regulations also had a negative
coefficient, implying that increased costs of the changes improved farm efficiencies. Farms that reported the highest costs for changes due to regulations
were also those that scored very high on estimated technical efficiency. Those
were likely also the farms with more comprehensive records, and were those
that were able to remain in business because they incurred the expense to
comply with regulatory changes. Those farms had a better understanding of
how regulatory changes had resulted in infrastructure and management
changes in their business and how this added additional costs. Management
changes reported to comply with regulations tended to include hiring additional personnel for record-keeping and hiring additional drivers for more
numerous, smaller, but more expensive trucks. Such changes all increased
costs, but larger farms that have stayed in business were those that made the
16
J. V. SENTEN ET AL.
changes. Smaller farms unable to make such changes likely exited the industry. van Senten and Engle (2017) showed that 23% of the total farm costs of
regulations were due to changes made to be in compliance.
Farm size was not found to be significant in the inefficiency equation. In
other studies of the technical efficiency of aquaculture, farm size generally
was not found to explain economic efficiencies (Iliyasu et al., 2014). This was
also true in analysis across counties in a specific country (Tan et al., 2011) or
across countries (Dey, Paraguas et al., 2005; Dey, Rab et al., 2005), although
results are variable. In U.S. baitfish/sportfish production, 29% of small-sized
farms exited the industry between 2005 and 2013 (USDA, 2014). van Senten
and Engle (2017) attributed this decrease in part to the substantially greater
costs/ha imposed by regulations on smaller, as compared to larger, farms.
Quantitative identification of regulatory variables that were significant
determinants of inefficiency, confirms that suggested by the descriptive data
discussed by van Senten and Engle (2017) in their summary of regulatory
costs on U.S. producers of baitfish and sportfish. Nearly, one-third (30%) of
survey respondents indicated that their top challenges (first or second) were
related to issues related to regulatory compliance. The current study affirms
that several components of regulatory compliance have increased farm-level
inefficiencies for U.S. baitfish and sportfish producers.
The problem that has emerged over the years is not necessarily the laws
and regulations themselves, but how the permits are written and enforced.
Osmundsen et al., (2017) referred to the dynamic nature of aquaculture
technologies that poses problems for regulators who frequently have little
formal training in aquaculture nor the means to remain current with its
rapidly evolving technologies and management practices. Osmundsen et al.,
(2017) called for a more adaptive regulatory system that would avoid
restricting new productivity-enhancing technologies on aquaculture farms.
Abate, Nielsen and Nielsen (2018) discussed the effects of inter-agency
rivalry, ideological perspectives of regulatory personnel, and the subsequent
effects on the regulatory environment for industries such as aquaculture.
Further analysis investigating effects of regulatory variables on technical
efficiency revealed that decreasing the number of permits/licenses would
result in an increase in efficiency on farms (Table 5). Similarly, a decrease
in the cost of manpower for compliance would result in increased efficiency, providing further evidence for the idea that the diversion of manpower to compliance tasks has a negative effect on production.
Conclusions
Regulatory variables in the inefficiency function were found to be significant determinants of farm inefficiency on U.S. baitfish/sportfish farms,
AQUACULTURE ECONOMICS & MANAGEMENT
17
Table 5. Sensitivity of technical efficiency score to the changes in number of permits and manpower costs.
Variable (X)
X level
Decrease in X
No. of permits
Now
16.000
0.000
10% reduction
14.400
1.600
20% reduction
12.800
3.200
30% reduction
11.200
4.800
40% reduction
9.600
6.400
50% reduction
8.000
8.000
Manpower cost (in natural logarithms)
Now
9.677
0.000
10% reduction
9.572
0.105
20% reduction
9.454
0.118
30% reduction
9.320
0.134
40% reduction
9.166
0.154
50% reduction
8.984
0.182
Represents reduction in absolute values of manpower cost.
U
TE
0.267
0.255
0.244
0.233
0.222
0.211
0.766
0.775
0.783
0.792
0.801
0.810
0.267
0.243
0.241
0.237
0.233
0.226
0.766
0.784
0.786
0.789
0.792
0.797
affirming that the U.S. regulatory environment has affected the competitiveness of baitfish and sportfish producers by reducing farm-level efficiency.
The significant effect of variables such as the number of annual permit/
license renewals and manpower to comply with regulations demonstrates
that the time farmers spend attempting to navigate a complex and convoluted business environment takes time away from productivity-enhancing
innovations and new market development. The farmers who participated in
the study did not argue that there should be no regulations; in fact, several
pointed out the need for regulations to maintain and protect the social
quality of life they desire and the natural systems they not only rely on, but
personally value and enjoy. However, this study, combined with the
descriptive results reported by van Senten and Engle (2017) point to a regulatory environment that is characterized by redundancy across agencies and
in reporting of compliance. The determinants of inefficiency identified in
this analysis demonstrate the excessive time burden on family businesses
whose owners are attempting to comply with the law, and, as a result, are
operating less efficiently than possible.
There clearly is a strong need to identify effective and practical ways to
streamline monitoring and compliance reporting activities across local,
state, and federal agencies to reduce the time burden and inefficiencies that
are introduced at the farm level and for prompt notification to farmers of
deadlines for renewals and of changes in the regulatory requirements. This
study demonstrates that the time burden resulting from compliance activities for producers is significant and that poor communication between and
from agencies results in additional, sometimes unexpected, costs at the
farm level. Specifically, it identifies areas of indirect regulatory cost that
have had a negative effect on farm technical efficiency. Study findings suggest that reducing the time burden at the farm level, resulting from
18
J. V. SENTEN ET AL.
frequent permit and license renewals and duplicative reporting requirements, as well as reducing unexpected changes at the farm level, resulting
from poor communication and poorly accessible information, could
improve farm efficiency. One possible solution would be for policy makers
to work closely with industry to understand how some regulations result in
unintended consequences at the farm level. Improved communication and
information sharing between agencies may also help to reduce the manpower required for reporting compliance at the farm level. Likewise, development of uniform reporting standards and forms, and easily accessible
information regarding regulations and compliance requirements could also
aid in reducing the time burden for producers. In the end, both regulators
and producers share the common goal of achieving compliance; working
together to identify solutions that are feasible, satisfy both sets of needs,
and reduce complexity benefits all parties involved.
Acknowledgements
The authors thank all those who contributed to the planning, implementation, and completion of this project; especially to the producers of baitfish and sportfish who participated
and trusted us to maintain the confidentiality of their data. We would also like to thank all
of the extension specialists, state aquaculture coordinators, and state agencies who contributed in developing the lists of baitfish and sportfish farms for the 13 study states.
Funding
This study was funded, in part by USDA, APHIS [Cooperative Agreement Award No.149200-0403-CA].
ORCID
Jonathan Van Senten
http://orcid.org/0000-0002-3513-7600
References
Abate, T. G., Nielsen, R., & Tveterås, R. (2016). Stringency of environmental regulation and
aquaculture growth: A cross-country analysis. Aquaculture Economics and Management,
20, 201–221. doi:10.1080/13657305.2016.1156191
Abate, T., Nielsen, R., & Nielsen, M. (2018). Agency rivalry in a shared regulatory space and
its impact on social welfare: the case of aquaculture regulation. Aquaculture Economics
and Management, 22, 27–48. http://dx.doi.org/10.1080/13657305.2017.1334243.
Aigner, D. J., Lovell, C. A. K., & Schmidt, P. (1977). and estimation of stochastic frontier
production function models. Journal of Econometrics, 6, 21–38. doi:10.1016/03044076(77)90052-5
Arita, S., & Leung, P.-S. (2014). A technical efficiency analysis of Hawaii’s aquaculture
industry. Journal of the World Aquaculture Society, 45, 312–321. doi:10.1111/jwas.12124
AQUACULTURE ECONOMICS & MANAGEMENT
19
Asche, F., & Roll, K. (2013). Determinants of inefficiency in Norwegian salmon aquaculture.
Aquaculture Economics & Management, 17, 300–321. doi:10.1080/13657305.2013.812154
Battese, G. E., & Coelli, T. J. (1995). A model for technical efficiency effects in a stochastic
frontier production function for panel data. Empirical Economics, 20, 325–332.
doi:10.1007/BF01205442
Broughton, E. I., & Walker, D. G. (2010). Policies and practices for aquaculture food safety
in China. Food Policy, 35, 471–478. http://www.sciencedirect.com/science/article/pii/
S0306919210000606.
Brunson, M., Hargreaves, J., & Stone, N. (2000). Fertilization of fish ponds. In R. R.
Stickney (Ed.), Encyclopedia of Aquaculture (pp. 360–363). Chichester, UK: John Wiley &
Sons, Inc.
Coelli, T., Rao, D. S. P., O’Donnell, C. J., & Battese, G. E. (2005). An introduction to efficiency and productivity analysis. New York: Springer-Verlag.
Coelli, T. J. (2011). A guide to FRONTIER Version 4.1: a computer program for stochastic
frontier production and cost function estimation. Centre for Efficiency and Productivity
Analysis Working Papers. No. 7/96. Armidale, Australia: The University of New England.
Crews, C. W. (2017). Ten thousand commandments: An annual snapshot of the federal regulatory state. Washington, DC: Competitive Enterprise Institute.
Dey, M. M., Paraguas, F. J., Srichantuk, N., Xinhua, Y., Bhatta, R., & Thi, C. D. L. (2005).
Technical efficiency of freshwater pond polyculture production in selected Asian countries: Estimation and implication. Aquaculture Economics and Management, 9, 39–63.
doi:10.1080/13657300590961528
Dey, M. M., Rab, M., Paraguas, F. J., Piumsombun, S., Bhatta, R., Alam, M. F., …
Ahmed, M. (2005). Status and economics of freshwater aquaculture in selected countries
in Asia. Aquaculture Economics and Management, 9, 11–38. doi:10.1080/
13657300590961609
Dresdner, J., & Estay, M. (2016). Biosecurity versus profits: a multiobjective model for the
aquaculture industry. Journal of the World Aquaculture Society, 47, 61–73. doi:10.1111/
jwas.12256
Engle, C. R. (2010). Aquaculture Economics and Financing: Management and Analysis.
Ames, Iowa: Blackwell Scientific.
Engle, C. R. (2016). Sustainable growth of aquaculture: the need for research to evaluate the
impacts of regulatory frameworks. Journal of the World Aquaculture Society, 47, 461–463.
doi:10.1111/jwas.12340
Engle, C. R., & Stone, N. (2013). Competitiveness of U.S. aquaculture within the current
U.S. regulatory framework. Aquaculture Economics and Management, 17, 251–280.
doi:10.1080/13657305.2013.812158
Fengxia, D., Hennessy, D. A., Jensen, H. H., & Volpe, R. J. (2016). Technical efficiency, herd
size, and exit intentions in U.S. dairy farms. Agricultural Economics: The Journal of the
International Association of Agricultural Economists, 47, 533–545. doi:10.1111/agec.12253
Grieco, P. L. E., Li, S., & Zhang, H. (2016). Production function estimation with unobserved
input price dispersion. International Economic Review. 57, (2), 665–690. Retrieved from
https://onlinelibrary.wiley.com/doi/abs/10.1111/iere.12172, May, 2016.
Hassan, S., & Ahmad, B. (2005). Stochastic frontier production function, application and
hypothesis testing. International Journal of Agriculture and Biology, 7 (3), 427–430.
Hurley, S. P. (2004). A cross comparison between California and its domestic and international competitors with respect to key labor issues. (Report). San Luis Obispo,
California: California Polytechnic State University, Retrieved from http://digitalcommons.
calpoly.edu/cgi/viewcontent.cgi?article=1061&context=agb_fac, January, 2016.
20
J. V. SENTEN ET AL.
Hurley, S. P., & Noel, J. (2006). An estimation of the regulatory cost on California agricultural producers. (Report). Long Beach, California: American Agricultural Economics
Association. Retrieved from http://digitalcommons.calpoly.edu/cgi/viewcontent.cgi?article=1045&context=agb_fac, January, 2016.
Iliyasu, A., Mohamed, Z. A., Ismail, M. M., Abdullah, A. M., Kamarudin, S. M., & Mazuki,
H. (2014). A review of production frontier research in aquaculture (2001–2011).
Aquaculture Economics & Management, 18, 221–247. doi:10.1080/13657305.2014.926464
Iliyasu, A., & Mohamed, Z. A. (2015). Technical efficiency of tank culture systems in peninsular Malaysia: an application of data envelopment analysis. Aquaculture Economics and
Management, 19, 372–386. doi:10.1080/13657305.2015.1082118
Jones, S., Stone, N., Kelly, A., Selden, G., Timmons, B., Whisenhunt, J., & Oliver, M.
(2016). Farm pond management for recreational fishing. MP360, Cooperative Extension
Program, University of Arkansas at Pine Bluff, Pine Bluff, Arkansas.
Kite-Powell, H., Rubino, M. C., & Morehead, B. (2013). The future of US seafood supply.
Aquaculture Economics and Management, 17, 228–250. doi:10.1080/13657305.2013.812691
Knapp, G., & Rubino, M. C. (2016). The political economics of marine aquaculture in the
United States. Reviews in Fisheries Science and Aquaculture, 24, 213–229. doi:10.1080/
23308249.2015.1121202
Kodde, D. A., & Palm, F. C. (1986). Wald criteria for jointly testing equality and inequality
restrictions. Econometrica, 54, 1243–1248. doi:10.2307/1912331
Lacewell, R. D., Nichols, J. P., & Jambers, T. H. Jr. (1973). An analysis of pond raised catfish production in Texas. Southern Journal of Agricultural Economics, 5, 141–145.
doi:10.1017/S0081305200010943
Lakner, S., Brenes-Mu~
noz, T., & Br€
ummer, B. (2017). Technical efficiency in Chilean
agribusiness industry: a meta-frontier approach. Agribusiness: An International Journal,
33, (3), 302–323. doi:10.1002/agr.21493.
Litwin, M. S. (1995). How to measure survey reliability and validity. Thousand Oaks,
California: Sage.
Losinger, W., Dasgupta, S., Engle, C., & Wagner, B. (2007). Economic interactions between
feeding rates and stocking densities in intensive catfish Ictalurus punctatus production.
Journal of the World Aquaculture Society, 31, 491–502. doi:10.1111/j.17497345.2000.tb00901.x
Mekonnen, D. K., Spielman, D. J., Fonsah, E. G., & Dorfman, J. H. (2015). Innovation systems and technical efficiency in developing-country agriculture. Agricultural Economics:
The Journal of the International Association of Agricultural Economists, 46, 689–702.
doi:10.1111/agec.12164
Mujeeb Rahiman, K. M., Mohamed Hatha, A. A., Gnana Selvam, A. D., & Thomas, A. P.
(2016). Relative prevalence of antibiotic resistance among heterotrophic bacteria
from natural and culture environments of freshwater prawn, Macrobrachium
Rosenbergii. Journal of the World Aquaculture Society, 47, 470–463. doi:10.1111/
jwas.12287.
Murova, O., & Chidmi, B. (2013). Technical efficiency of US dairy farms and federal government programs. Applied Economics, 45, 839–847. doi:10.1080/00036846.2011.613772
Nerrie, B. L., Hatch, L. U., Engle, C. R., & Smitherman, R. O. (1990). The economics of
intensifying catfish production: a production function analysis. Journal of the World
Aquaculture Society, 21, 216–224. doi:10.1111/j.1749-7345.1990.tb01026.x
Osmundsen, T. C., Almklov, P., & Tveterås, R. (2017). Fish farmers and regulators coping
with the wickedness of aquaculture. Aquaculture Economics and Management, 21,
163–183. doi:10.1080/13657305.2017.1262476
AQUACULTURE ECONOMICS & MANAGEMENT
21
Paul, C. J. M., Johnston, W. E., & Frengley, G. A. G. (2000). Efficiency in New Zealand
sheep and beef farming: the impacts of regulatory reform. The Review of Economics and
Statistics, 82, 325–337. doi:10.1162/003465300558713
Paul, C., Nehring, R., Banker, D., & Somwaru, A. (2004). Scale economies and efficiency in
U.S. agriculture: are traditional farms history? Journal of Productivity Analysis, 22,
185–205. doi:10.1007/s11123-004-7573-1
Porter, M. E., & van der Linde, C. (1995). Green and competitive. Harvard Business Review.
September–October 1995, 120–134. Retrieved from https://hbr.org/1995/09/green-andcompetitive-ending-the-stalemate
Rahman, K. M. M., Hatha, A. A. M., Selvam, A. D. G. & Thomas, A. P. (2016). Relative
prevalence of antibiotic resistance among heterotrophic bacteria from natural and culture
environments of freshwater prawn, Macrobrachium Rosenbergii. Journal of the World
Aquaculture Society, 47, 470–480.
Rezitis, A. N., & Kalantzi, M. A. (2016). Investigating technical efficiency and its determinants by data envelopment analysis: an application in the Greek food and beverages
manufacturing industry. Agribusiness: An International Journal, 32, 254–271. doi:10.1002/
agr.21432
Rico, A., Satapornvanit, K., Haque, M. M., Min, J., Nguyen, P. T., Telfer, T. C., &
van den Brink, P. J. (2012). Use of chemicals and biological products in Asian aquaculture and their potential environmental risks: a critical review. Reviews in Aquaculture, 4,
75–93. doi:10.1111/j.1753-5131.2012.01062.x
Rico, A., Phu, T. M., Satapornvanit, K., Min, J., Shahabuddin, A. M., Henriksson, P. J. G.,
… Van den Brink, P. U. (2013). Use of veterinary medicines, feed additives and probiotics in four major internationally traded aquaculture species farmed in Asia. Aquaculture,
412–413, 231–243. doi:10.1016/j.aquaculture.2013.07.028
Sandvold, H. N. (2016). Technical inefficiency, cost frontiers and learning-by-doing in
Norwegian farming of juvenile salmonids. Aquaculture Economics and Management, 20,
382–398. doi:10.1080/13657305.2016.1224659
Sauer, J., Gorton, M., & Davidova, S. (2015). Migration and farm technical efficiency: evidence from Kosovo. Agricultural Economics: The Journal of the International Association
of Agricultural Economists, 46, 629–641. doi:10.1111/agec.12159
StataCorp (2009). Stata Statistical Software: Release 11. College Station, TX: StataCorp. LP.
Stone, N. M., Kelly, A. M., & Roy, L. A. (2016). A fish of weedy waters: Golden shiner biology and culture. Journal of the World Aquaculture Society, 47, 152–2000. doi: 10.1111/
jwas.12269.
Stone, N. M., & Thomforde, H. (2001). Common farm-raised baitfish. Southern Regional
Aquaculture Center. SRAC 120. Retrieved from http://www.srac.tamu/edu/serveFactSheet/7
Tan, R. I., Garcia, Y. T., Dator, M. L., Tan, I. M. A., & Permsl, D. E. (2011). Technical efficiency of Genetically Improved Farmed Tilapia (GIFT) cage culture operations in the
lakes of Laguna and Batangas, Philippines. Journal of the International Society of
Southeast Asian Agricultural Sciences, 17 (1), 194–207.
Thunberg, E. M., Adams, C. M., & Cichra, C. E. (1994). Economic, regulatory, and technological barriers to entry into the Florida aquaculture industry. Journal of Applied
Aquaculture, 4, 3–14. doi:10.1300/J028v04n02_02
USDA (2014). Census of Aquaculture (2012). Retrieved from https://www.agcensus.usda.
gov/Publications/2012/Online_Resources/Aquaculture/aquacen.pdf, December, 2015.
van der Vlist, A. J., Withagen, C., & Folmer, H. (2007). Technical efficiency under alternative environmental regulatory regimes: the case of Dutch horticulture. Ecological
Economics, 63, 165–173. doi:10.1016/j.ecolecon.2006.10.013
22
J. V. SENTEN ET AL.
van Senten, J., & Engle, C. R. (2017). The cost of regulations on U.S. baitfish and sportfish
producers. Journal of the World Aquaculture Society, 48, 503. doi:10.1111/jwas.12416.
Yang, C. C., Hsiao, C. K., & Yu, M. M. (2008). Technical efficiency and impact of environmental regulations in farrow-to-finish swine production in Taiwan. Agricultural
Economics: The Journal of the International Association of Agricultural Economists, 39,
51–61. doi:10.1111/j.1574-0862.2008.00314.x
Carole Engle