AEDR ss Part B 052107

SUPPORTING STATEMENT
AMENDMENT 80 ECONOMIC DATA REPORT FOR THE CATCHER/PROCESSOR
NON-AFA TRAWL SECTOR
OMB CONTROL NO.: 0648-xxxx
B. COLLECTIONS OF INFORMATION EMPLOYING STATISTICAL METHODS
1. Describe (including a numerical estimate) the potential respondent universe and any
sampling or other respondent selection method to be used. Data on the number of entities
(e.g. establishments, State and local governmental units, households, or persons) in the
universe and the corresponding sample are to be provided in tabular form. The tabulation
must also include expected response rates for the collection as a whole. If the collection has
been conducted before, provide the actual response rate achieved.
The potential respondent universe will be a maximum of 28 non-AFA trawl catcher/processors
(see Table 1) operating in the waters of the BSAI and GOA. Non-AFA trawl catcher/processors
are a closed set that include those catcher/processors not listed as AFA catcher/processors at 50
CFR 679.4(l)(2)(i). Each catcher/processor would have one Amendment 80 QS permit and LLP
license holder required to collect and report all data on an Amendment 80 EDR. While a
maximum of 28 licensees could be required to report in this collection in the first year of the
proposed Amendment 80 program, owners of multiple licenses and associated vessels will be
required to submit one report for each license/vessel, reducing the respondent universe but not
the overall reporting burden. In addition, in subsequent years some consolidation in this sector
could occur, reducing the number of entities required to respond to the EDR.
This collection would require each QS holder for a non-AFA trawl catcher/processor to submit
an annual EDR, which is a single form design with identical fields. If the final rule for this
action is approved by the Secretary of Commerce in 2007, year 2008 would be the first full year
of data that would be required for the Amendment 80 EDR. The data (EDR and responses to
questions) for 2008 would be required by July 2009. Each subsequent year of catch and
production would require a new EDR.
Table 1. Non-AFA trawl catcher/processor sector, 2007
Vessel Name
F/V Alaskan Rose (Tremont)
F/V Arctic Rose (Sunk 2001)*
F/V Seafisher
F/V Alaska Juris
F/V Alaska Voyager
F/V Alaska Victory
F/V Alaska Warrior
F/V Alaska Ranger
F/V Alaska Spirit
F/V American #1
F/V US Intrepid
F/V Defender
F/V Enterprise
F/V Constellation
F/V Prosperity

Vessel Name
F/V Arica
F/V Cape Horn
F/V Rebecca Irene
F/V Unimak Enterprise
F/V Vaerdahl
F/V Alliance
F/V Legacy
F/V Bering Enterprise
F/V Harvester Enterprise
F/V Ocean Peace
F/V Seafreeze Alaska
F/V Ocean Alaska (Beagle)
F/V Golden Fleece

*The Arctic Rose may be replaced by one additional LLP that could be assigned to a new Vessel*

1

The sample selection method is an annual census of all 28 vessels, as any other sampling
methodology would produce too few observations to estimate representative levels of cost,
earnings, and other outputs required for this collection.
As this program is a mandatory collection, and valuable fishing privileges will be withheld if an
EDR is not submitted, we anticipate a 100% response rate from QS holders. Quota shares in this
program will be issued to entities, rather than vessels, and specific provisions in the rule require
that each QS holder is responsible for including data from any acquired vessel in this sector.
Also, each of these vessels is classified as a large entity with greater than $4.0 million in annual
gross earnings. The organizations owning and managing these vessels routinely provide NMFS
extensive data on catch by location and weight as well as production data to both NMFS and the
State of Alaska through logbooks, catch account reports, and other collections. The long history
of this sector in providing mandatory data reinforces the expectation that this data will be
provided by all vessels in this program.
2. Describe the procedures for the collection, including: the statistical methodology for
stratification and sample selection; the estimation procedure; the degree of accuracy
needed for the purpose described in the justification; any unusual problems requiring
specialized sampling procedures; and any use of periodic (less frequent than annual) data
collection cycles to reduce burden.
Given that only 28 vessels will be participating in this fishery, it is not feasible to generate
enough observations on any one of the variables without applying this collection annually. And,
as discussed above, random sampling from this population is not a viable option for statistical
reasons. Based upon the degrees of freedom and number of observations required for estimating
the statistical relationship among the variables in this collection, data in the Amendment 80 EDR
may be pooled to create a time-series of cross-sectional data in order to generate sufficient
observations for economic and statistical analysis. Although the strata to be utilized in preparing
analyses (either deterministic or statistical) of management actions for this fleet will depend on
the specific questions of interest, vessels are commonly stratified by vessel length and the
distribution and amount of catch, by species.
a. Potential dependent variables and models developed with EDR data
Much of the data requested will be used to compute total or average quasi-rents (revenues less
variable costs) based on a census of catcher/processors in the years following implementation of
this rationalization program. To understand the relationships between the vessel quasi-rents and
the variables we collect that affect total or average quasi-rents, econometric models will be
required. Examples of some dependent and exogenous variables of interest are shown in the
following table.

2

Examples of some dependent and exogenous variables of interest
Estimating Dependent Variables that do not Require a Model
a) Distribution of average catch and processed revenue by vessel length class, or
type of operation (based on distribution and amount of catch by species)
b)
Distribution of average variable vessel costs by vessel length class, or type
of operation (based on distribution and amount of catch by species)

c)
Distribution of average quasi-rents by vessel length class, or type of
operation (based on distribution and amount of catch by species)

d) Seasonality of average catch and revenue by vessel class
e) Catcher processor vessel ownership & interest in QS
f) Level and distribution of harvesting and processing sector employment and
payments to labor (number of individuals, hours/days worked, and income)

g) Degree of involvement of non-AFA trawl catcher/processor sector in other AK
fisheries
h) Observer Costs in QS Fisheries (Impacts of Increased Observer Coverage)
i) Total fishing and processing taxes including fee collection
j) Changes in Fleet Composition (comparison of cost, revenue and compensation
structure of vessels exiting the fleet versus those staying, based on the measures
given in this section)
k) Product Recovery Rates (PRR) by species
l) Production
m) Consolidation
n) Observer costs

Data Required
Catch, production and revenue information, vessel information, and vessel owner information are required. COAR data would be used as the primary
source for providing data on gross revenues paid by processing product and species.
Data Required
Total variable costs, by vessel, vessel characteristics, landings records
Specific Measure
Annual Total Variable Costs = CDQ costs + QS costs + observer costs + fuel + lube and hydraulics + food and provisions + freight costs for landed fish +
lube and hydraulic fluid + crew payment or share payment + processing materials + labor costs for processing + packaging + freezing + captain's share
payment + fish taxes (including raw fish and local tax) + gear costs
Seasonal Variable Harvesting Costs = fuel costs + captain and crew costs + gear costs
Freight & Storage Costs = Freight costs of supplies to vessel + freight costs for landed fish + storage costs
Data Required
Total variable costs, by vessel, vessel characteristics, landings records; COAR data would be the primary source for providing data on gross revenues paid
by processing product and species
Specific Measure
Quasi-rents = Total revenue - (CDQ royalty payments + IFQ costs + fuel + lube and hydraulics + food and provisions + freight costs for landed fish + lube
and hydraulic fluid + crew share payment + captain's share payment + fish taxes + processing materials + labor costs for processing + packaging +
freezing)
Quasi-rents / pounds landed = QR per pound
Quasi-rents / days fished = QR per day
Data Required
Catch, processed revenue, vessel class and ownership.
Data Required
Processor, vessel and QS ownership data are required.
Data Required
Harvesting and processing sector employment and payments to labor data are required.
Specific Measures
Labor Income = Crew share payment + Captain's share payment + QS holder’s payments (where applicable) + processing labor payment + all other labor
payment
or
Labor Income = Crew share * (Total revenue - CDQ leases - QS leases - fuel - lube and hydraulics - bait - food and provisions - freight costs for supplies freight costs for landed & processed fish - fish taxes) + processing labor payment + all other labor payment
Where applicable
Averaged daily Wage = Labor Payment / # of Processing Days
$ per Hour = Labor Payment / Total Man-hours
Labor as % of Revenue = labor payment / value of product
Labor as % of variable costs = labor payment / variable costs
Labor Income Per Capita = Labor income / # of crew earning shares
Average number of harvesting crew per vessel by season (by geographic region of employee residence)
Average captain's share (%) & wages
Average crew share (%) & wages
Description of typical expenses deducted from crew wages
Data Required
Catcher Processor and vessel ownership data, as well as total catch, production, and revenue data are required.
Data Required
Cost per day-at-sea by individual. Number of days purchased per season from data collected by the observer program.
Data Required/Specific Measures
Taxes, use fees paid by catcher/processors
Data Required/Specific Measures
Cost, revenue, labor income, and compensation structure of vessels to construct the measures given in the above section.
PRR = Finished Pounds / Raw Pounds
Production per Day = Finished Pounds / # of Processing Days
Production Per Employee = Finished Pounds / # catcher/processor positions
Avg. Production per catcher/processor = total processed pounds / # of catcher/ processors producing groundfish.
Observer cost as percent of revenue= Observer costs / revenue
Observer cost per day = Observer cost / # of processing days

3

b. Estimating Dependent Variables that Require a Model
AFSC analysts will use the data contained within the completed and verified EDRs to construct
statistical models that characterize the determinants and factors affecting the costs and revenues
of vessels within each stratum. The benefit of using statistical models to characterize the
relationship between costs or revenues and the factors that influence them is that the models may
initially be used to analyze the way in which economic performance changes after the immediate
implementation of the program. If the Council makes adjustments to the program at a later date,
analysts will be able to observe the changes in quasi-rents not attributable to the factors that have
historically been the predominant statistical determinants to draw conclusions about the impact
of the adjustments. That is, these statistical techniques can be used to disentangle the influence
of particular economic variables on quasi-rents from “policy” or “management” variables that
change directly as a result of managers’ choices over policies or regulations. Examples of
economic variables would be the prices of fuel, materials, or other inputs used in fishing and
processing. Variables that can be altered directly by fishery managers or regulation are the
length of fishery openings by statistical area and species, the amount of allocation of a species to
a sector, or individual vessels or persons in a sector.
The data collected in the EDRs will be used to develop cost and quasi-rent (i.e., restricted profit)
functions that characterize the relationships between fishing and processing activities and their
economic impacts. In order to estimate such functions one needs vessel-level information on
variable costs of operation and gross earnings. These variables will form the basis for the
dependent part of the statistical model, while the other data collected on input quantities, catch,
and prices will be used as exogenous variables. The analysts will determine the exact
specification of the cost and quasi-rent functions based upon the questions desired by fishery
managers, the number of observations available, and the perceived quality or accuracy of the
collected data.
Econometric Methods. The primary and most common approach for estimating and specifying
cost and quasi-rent functions is with econometric methods. This approach examines the
multivariate statistical relationships between short- run costs or quasi-rents and exogenous
variables, using choices or decisions made by economic agents over target species and fishing
location. Observed behavior over time and strata may be merged with other data to infer how
management actions impact quasi-rents. This analysis would include data on catch by species
and area, data on the value of retaining catch of a given species, and data on species with lower
market value. Error and regression statistics may be generated from econometric models to
indicate the level of statistical significance of estimated parameters. Given the number of
variables that could be included in any of these models, we are not prepared at this time to
provide quantitative standards of accuracy for each parameter included in the EDR. The level of
accuracy required in any given independent data value for estimating a particular dependent
variable may vary greatly from one dependent variable to another.
Mathematical Programming Model. A second approach that could be used to characterize the
relationship between costs or quasi-rents and economic variables would be a mathematical
programming model. In this approach one makes an assumption about the way in which the
variables are related, and conducts non-parametric tests on how well it explains the variation in
quasi rents. Multilevel and multi-objective programming models have been used in fisheries to
evaluate management policies. They may involve linear or non-linear programming, and would
also generate uncertainty measures to evaluate the model accuracy.
4

3. Describe the methods used to maximize response rates and to deal with non-response.
The accuracy and reliability of the information collected must be shown to be adequate for
the intended uses. For collections based on sampling, a special justification must be
provided if they will not yield "reliable" data that can be generalized to the universe
studied.
Each of the QS holders operating a catcher/processor in this fishery will be required to submit an
annual EDR. All of these respondents will be applying for one or more QS. Because this is a
mandatory collection, and valuable fishing privileges will be withheld if an EDR is not
submitted, we anticipate a 100% response rate from QS holders.
Measures to verify the accuracy of the EDR data would be developed by NMFS economists and
analysts to ascertain anomalies, outliers, and other deviations from averaged variables. The
principle means to verify data would be consultation between NMFS and the submitter when
questions arise regarding data. NMFS would request oral or written confirmation of data
submissions and request copies of or review documents or statements that would substantiate
data submissions. The person submitting the EDR would need to respond within 20 days of the
inquiry for information. Responses after 20 days could be considered untimely and could result
in a violation and enforcement action.
NMFS would amend data in the EDR through this audit verification. NMFS could choose to
audit an EDR either through random selection or when circumstances require more thorough
review of the submissions. In instances where a random audit occurs or an audit is otherwise
justified, NMFS may retain a professional auditor/accounting specialist who would review and
request financial documents substantiating economic data that is questioned. NOAA guidelines
for the Data Quality Act will be followed and estimates without an adequate statistical basis will
not be used.
Enforcement of the data collection program will be different from enforcement programs used to
ensure that accurate landings are reported. It is critical that landings data are reported in an
accurate and timely manner, especially under a QS system, to properly monitor catch and
remaining quota. However, because it is unlikely that the economic data will be used for inseason management, it is anticipated that persons submitting the data will have an opportunity to
correct omissions and errors before any enforcement action would be taken. Giving the person
submitting data a chance to correct problems is considered important because of the complexities
associated with generating these data. Only if the agency and the person submitting the data
cannot reach a solution would the enforcement agency be contacted. The intent of this program
is to ensure that accurate data are collected without being overly burdensome on industry due to
unintended errors.
A discussion of four scenarios will be presented to reflect the analysts’ understanding of how the
enforcement program would function. The four scenarios are:
1. No information is provided on an EDR;
2. Partial information is provided on an EDR;
3. NMFS has questions regarding the accuracy of the data that has been submitted
on an EDR; and
4. A random audit to verify the data does not agree with data submitted in the EDR.
5

In the first two cases, the person would be contacted by NMFS (or a NMFS contractor) and
asked to fulfill his/her obligation to provide the required information. If the problem is resolved
and the requested data are provided, no other action would be taken. If that person does not
comply with the request, the collecting agency would notify enforcement that the person is not
complying with the requirement to provide the data. Enforcement would then use their
discretion regarding the best method to achieve compliance. Those methods would likely
include fines or loss of quota and could include criminal prosecution.
In the third case, questions may arise when, for example, information provided by one company
is much different than that provided by similar companies. These data would only be called into
question when obvious differences are encountered. Should these cases arise, the agency
collecting the data would request that the person providing the data double check the
information. Any reporting errors could be corrected at that time. If the person submitting the
data indicates that the data are accurate and the agency still has questions regarding the data, that
firm’s data could be audited. It is anticipated that the review of data would be conducted by an
accounting firm selected jointly by the agency and members of industry. Only when that firm
refuses to comply with the collecting agency’s attempts to verify the accuracy of the data would
enforcement be contacted. Once contacted, enforcement would once again use their discretion
on how to achieve compliance.
In the fourth case, a random audit reports different information than that contained in the EDR.
The audit procedure being contemplated is a verification protocol similar to that which was
envisioned for use in the pollock data collection program developed by NMFS and Pacific States
Marine Fisheries Commission (PSMFC). During the design of this process, input from certified
public accountants was solicited in order to develop a verification process that is less costly and
cumbersome than a typical audit procedure. That protocol involves using an accounting firm,
agreed upon by the agency and industry, to conduct a random review of certain elements of the
data provided.
Since some of the information requested in the EDRs may not be maintained by companies and
must be calculated, it is possible that differences between the audited data from financial
statements and EDR data may arise. In that case the person filling out the form would be asked
to show how his/her numbers were derived. If the explanation resolves the problem, there would
be no further action needed. If questions remained, the agency would continue to work with the
providers of the data. Only when an impasse is reached would enforcement be called upon to
resolve the issue. It is hoped that this system would help to prevent abuse of the verification and
enforcement authority.
In summary, members of the non-AFA trawl catcher/processor sector will be contacted and
given the opportunity to explain and/or correct any problems with the data, which are not willful
and intentional attempts to mislead, before enforcement actions are taken. Agency staff does not
view enforcement of this program as they would a quota monitoring program. Because these
data are not being collected in “real” time, there is the opportunity to resolve occasional
problems as part of the data collection system. The program will be developed to collect the best
information possible. Analyses of the Amendment 80 rationalization program will be conducted,
to minimize the burden on industry and minimize the need for enforcement actions.

6

4. Describe any tests of procedures or methods to be undertaken. Tests are encouraged as
effective means to refine collections, but if ten or more test respondents are involved OMB
must give prior approval.
The Council held two industry meetings in 2006 to review and recommend data to be collected
in the EDRs. While this did not result in a formal pretest of the data reports, several fields in the
data forms were significantly revised. In addition, some members of the non-AFA trawl
catcher/processor sector have voluntarily submitted individual comments on previous versions of
this data form.
5. Provide the name and telephone number of individuals consulted on the statistical
aspects of the design, and the name of the agency unit, contractor(s), grantee(s), or other
person(s) who will actually collect and/or analyze the information for the agency.
Ron Felthoven, Ph.D.
Economist
NMFS WASC Route: F/AKC3
PH: (206) 526-4114
Internet Address: ron.felthoven@noaa.gov
Brian Garber-Yonts, Ph.D.
Economist
NMFS WASC Route: F/AKC3
PH: (206)526-4114
Internet Address: Brian.garber-yonts@noaa.gov
Mark Fina, Ph.D.
Economist
North Pacific Fisheries Management Council
PH: (907) 271-2809
Internet Address: mark.fina@noaa.gov
Dave Colpo
Program Manager
Pacific States Marine Fisheries Commission
PH: (503) 595-3100
Internet Address: front_office@psmfc.org

7