Multi-Industry Contribution Analysis

Comments

9 comments

  • Avatar
    IMPLAN Support
    Hello Tom, We also agree that multi-industry contribution analysis is a good approach. In regards to question #1, this adjusts the model so that each industry in the contribution analysis is making 100% of its primary commodity and thus the commodity value is not being altered by production in another industry that may produce that primary commodity an a byproduct. (e.g. Since multipliers are industry based if the marble factory has a restaurant by-product, then we will get marble factory indirect whenever anyone wants restaurant food.) #2, for event year in a contribution analysis, you will want to use 2008 since you Model is 2008, then you can set Dollar Year for View to the year you are interested in viewing. This is because you are basing your numbers off of the models relationships, which would be 2008, and it also allows you to check your results because the Direct Sum of all factors (Employment, Labor Income, Value Added, Output) should equal the values in the Study Area Data for the same group of manufacturing sectors. 3#; BLS CEW data is not complete coverage of employment. It only covers wage and salary employment covered by UI and federal civilian jobs covered by UCFE. It does not include proprietors, military, elected officials, railroads, religious organizations, small farms, and some private elementary school employment. IMPLAN employment numbers will more closely resemble BEA REIS numbers. Please let us know if you have any further questions.
    0
    Comment actions Permalink
  • Avatar
    Tom
    Thank you. #1 and #2 make sense. I'm still a little unclear on #3. Manufacturing is almost entirely covered so BEA and QCEW align well. The most recent BEA employment is a litte over 31,000 vs. a little under 31,000 with QCEW. Does the employment impact total more closely align with 2008? Thanks, Tom
    0
    Comment actions Permalink
  • Avatar
    IMPLAN Support
    Hello Tom, Yes since the data you are using is 2008 data you would need to compare it to 2008 QCEW and BEA REA numbers, as these would be the numbers from which the data is derived. The Dollar Year for View field will inflate the values between the current year and model year for you, but it does not affect the underlying model relationships which will all be based on reported 2008 values. Sorry for the confusion. Please let us know if you have any additional questions
    0
    Comment actions Permalink
  • Avatar
    Tom
    The 2008 employment totals are comparable. Do the output and value added contributions also represent the 2008 economy (with inflated dollar values)? If so, can I get a more current contribution overview by customizing the study area data? Tom
    0
    Comment actions Permalink
  • Avatar
    IMPLAN Support
    Hi Tom, We apologize but we are having some issues with our forum today, so we are having to respond via email, and this has also delayed the response times. All the data in a 2008 model will reflect the 2008 data year. If you have newer data available and are comfortable doing so you can certainly modify you model by changing the numbers in the Customize menu. However, there are two additional considerations:The remainder of the model will still retain the 2008 relationships, so your analysis will be using one Sector updated to 2010 relationships with the remaining relationships in the model in 2008, including the trade relationships for you manufacturing Sector, which cannot be edited. If you are comfortable with this as an analyst, you can certainly make that modification. However, we definitely recommend making mention of any modifications that you do to the Study Area data in your report. Additionally, just for your information, the Employment data in IMPLAN does not drive any economic impacts. The Indirect sales are driven by the value of the Industry Sales for the Sector you select and the Induced impacts are driven by the Employment Compensation and Proprietor Income and then the successive rounds of income from the Indirect and Induced purchases. If you’d like to update your data to reflect the current years (2011) we would be glad to assist you with that as well. Thanks, Implan Support Staff
    0
    Comment actions Permalink
  • Avatar
    Tom
    Thanks. I would certainly mention any modifications. I’m having kind of a tough time with my region because paper manufacturing and wood products are two of the largest industry subsectors. Both have taken huge hits over the past five years. I’m comfortable making an assumption that the trade relationships have remained constant, but I feel like I need to adjust for the sharp declines in wood and paper. Can I run a multi-region contribution analysis if I aggregate some industries? I know I would lose precision, and it would average out the trade relationships and the multipliers. However, it might be my best option if I’m customizing the study area data. Along with producing a contribution estimate, I might use this to make an argument for getting the updated data. I’m a fairly new implan user, but it has been an incredibly valuable tool. Tom
    0
    Comment actions Permalink
  • Avatar
    IMPLAN Support
    Hello Tom, Unfortunately multi-region analysis is not possible when using any aggregated model. However, a multi-industry contribution analysis in a single region can be preformed by following the same instructions after the Model has been aggregated. It sounds like you are aware of [url=http://implan.com/v4/index.php?option=com_glossary&id=227]aggregation bias[/url], but if you are interested in a short definition of what can happen with aggregated industries, please see this glossary. We certainly are not trying to say that you have to update your data, however it seems that in your circumstance and with the analysis you are preforming, it might be easier to purchase a newer data sets. If these would be helpful to you, here are a couple of links that talk about when or why to update your data. http://implan.com/v4/index.php?option=com_content&view=article&id=752:how-often-do-i-need-to-update-my-data&catid=269:faq&Itemid=112 https://implan.com/v4/index.php?option=com_content&view=article&id=840%3Aimplan-annual-data-updates&catid=254%3AKB34&Itemid=2 Let us know if you have more questions.
    0
    Comment actions Permalink
  • Avatar
    schmi_ny
    Do you have additional details on the assumptions made for a contribution analysis? Perhaps the algebra behind the analysis to help explain the methodology more specifically within the general IO language. I want to ensure I'm interpreting the results correctly and can explain the assumptions made to generate the results. Thanks. Todd
    0
    Comment actions Permalink
  • Avatar
    IMPLAN Support
    Hi Todd, For the Multi-Industry Contribution analysis the procedure is actually very simple. By editing the RPC values to zero for all the primary commodities produced by the primary industry you are examining with a contribution analysis, and by editing Byproducts so that each industry creates 100% of its primary commodities production for the region we are forcing all Indirect and Induced purchases of these industries to be imported or leaked from the model. This results in the Total impact for each of the desired industries equaling only the Direct Effect specified in Industry Sales. Thus all sales of that industry for the region are constrained to the Direct Effects. But since each of these industries still needs to buy its full compliment of inputs and pay its workers for the total level of regional production all the local Indirect and Induced purchases are still captured for all products except those that you are Directly impacting and thus have constrained by setting the RPC to zero. Please let us know if this addresses your concern or if you have any additional questions.
    0
    Comment actions Permalink

Please sign in to leave a comment.