Sampling Procedures Developed for Data Collection Projects
Effort continues on developing an appropriate sampling methodology for ongoing regional economic data collection projects. Because the majority of gross revenue within each harvesting sector comes from a few number of boats, a simple random sampling of boats would only include a small portion of the total exvessel values, and therefore, would be misleading. Therefore, an unequal probability sampling (UPS) method without replacement will be used. The objective of implementing the sampling task is to estimate the employment and labor income information for each of three disaggregated harvesting sectors using the exvessel revenue information provided by the Alaska Commercial Fisheries Entry Commission earnings data.
Since each sector will be used as a separate economic sector in the IMPLAN model (a regional economics model), we face three separate problems for three different sectors in sampling. For each sector, we will implement a UPS without replacement. In the literature, many methods exist for conducting UPS without replacement. One critical weakness with most of these methods is that the variance estimation is very difficult because the structure of the 2nd order inclusion probabilities is complicated.
One method that overcomes this problem is Poisson sampling. However, a problem with Poisson sampling is that the sample size is a random variable, which increases the variability of the estimates produced. An alternative method similar to Poisson sampling but that overcomes its weaknesses is Pareto sampling (which yields a fixed sample size).
In this project, there are two tasks that we need to accomplish to estimate the population parameters using the UPS. First, the optimal sample size needs to be determined. Second, once the optimal sample size is determined, the population parameters and confidence intervals need to be estimated.
For the first task, we will use the Poisson variance (not Poisson sampling). For the second task, we will use a Pareto sampling method. In determining the optimal sample size, we will use information on an auxiliary variable (ex-vessel revenue). To estimate the population parameters, we will use actual response sample information on the variables of interest (employment and labor income).
With inputs from experts in UPS sampling, a document detailing these sampling procedures has been completed and an Excel program has been developed to show these procedures using an example data (2002 exvessel value data for small boat sector).
By Chang Seung
Estimating Interregional Economic Effects of Vessels in Alaska and West Coast Fisheries
Many vessels operating in Alaska fisheries are owned and crewed by residents of Washington and Oregon. Some of these vessels also participate in West Coast fisheries during the year. While much of the income earned by these vessels leaves Alaska, expenditures made elsewhere will generate positive economic impacts for that region and may also have spillover effects. It has been demonstrated that assuming all commodities and services are locally supplied will significantly overestimate regional impacts. Understanding the location of expenditures made by the vessels, both in Alaska and elsewhere, will enhance our understanding of the overall economic impacts of Alaska fisheries.
Standard regional economic models focus on a single region. These models generally fail to capture economic impacts transmitted outside that region and also do not account for spillover effects in the study region resulting from events occurring outside. An interregional or multiregional model can more fully measure the impacts of a region’s fisheries, including those impacts occurring in regions that supply commodities or factors of production to industries in the study region, or that demand the goods and services produced there. An inter-regional model would be especially useful in the case of Alaska, where most intermediate goods are imported and much of the factor income leaks out of the region to nonresident vessel owners and crew members.
This type of model could also be used to track the impact of expenditures by vessel owners and crew members who are also active in other regions’ fisheries. However, developing an inter-regional model involves the daunting task of estimating interregional flows of commodity inputs and factor services. Acquiring this information has traditionally been very challenging due to an absence of interregional trade flow statistics. Consequently, to date, only one published study employed an interregional or multiregional model to estimate economic impacts of fisheries.
Our project will estimate the distribution and magnitude of intra-regional and inter-regional economic impacts generated by vessels participating in both Alaska fisheries and in fisheries off the U.S. West Coast. Recently, Alaska Department of Labor and Workforce Development has created useable data for estimating nonresident labor use by Alaska industries, including seafood processing and some commercial fishing.
Data on the ownership of vessels used in Alaska and West Coast fisheries also exists, and annual catch by these vessels is available from PacFIN and NORPAC data systems. Information on the cost structures of vessels participating in the two fisheries will be gathered from the literature and key industry informants. Results of a cost and earnings survey of West Coast groundfish trawlers that is currently being administered by the NWFSC may be available to assist this project.
This project will develop a multiregional social accounting matrix (SAM) model of Alaska and West Coast fisheries using these data combined with IMPLAN regional models constructed for Alaska and the U.S West Coast (i.e., Washington and possibly Oregon). IMPLAN Version 3.0 (beta version) will include an inter-regional trade modeling capability that will facilitate the estimation of commodity trade flows between the two regions. The investigators in this project recently developed a single-region Alaska SAM model to examine the economic impact of Alaska fisheries. This project will build on that effort to develop an updated interregional SAM model.
The implementation of this project will include the following steps:
Gather data on the residence of owners and crews of vessels operating in Alaska and U.S. West Coast fisheries from NOAA permits databases and other sources.
Gather annual catch data by these vessels from PacFIN and NORPAC data systems.
Gather information on vessel cost structures and the locus of input purchases by vessels participating in the two fisheries. Major sources of data will include relevant literature and interviews with key industry informants. Results of a cost and earnings survey of West Coast trawlers currently being administered by NWFSC may be available to assist this task.
Generate regional economic models of Alaska and the U.S. West Coast (Washington and Oregon) economies using IMPLAN. The models will incorporate the latest representative economic data available for both regions.
Estimate the value of commodities, services, labor and capital flowing between Alaska and the West Coast using IMPLAN and the models developed in Step 4. The focus will be on those factors, commodities, and services of particular importance to commercial fisheries-related economic activity.
Develop a multiregional social accounting matrix (SAM) model of Alaska and West Coast economies using fisheries data, trade estimates and IMPLAN regional models developed in steps 1–5.
Use the multiregional SAM model to estimate economic impacts of commercial fishing and related activities on Alaska and the U.S. West Coast.
Understanding the location and magnitude of effects generated by these vessels, both in Alaska and elsewhere, will enhance our understanding of the overall economic impacts of Alaska fisheries. After this project is completed, the investigators will conduct (possibly jointly with a NWFSC economist) a potential follow-on project. In the follow-on project, a more comprehensive data gathering program will be implemented to resolve economic data issues identified in steps 1–3 above. These data will be used to validate the interregional SAM model and would be available for the development of an advanced model such as an interregional computable general equilibrium (CGE) model.
