sector comparison of two different business types

Comments

15 comments

  • Avatar
    jenny
    I believe your best bet will be to use the technique known as analysis-by-parts (ABP), which allows you to edit a sector's spending pattern. There is a tutorial for ABP (http://implan.com/V4/index.php?option=com_docman&task=doc_download&gid=108&Itemid=7) and numerous forum discussions. This short video will show you how to take advantage of the search features on our website and forum: http://implan.com/v4/index.php?option=com_content&view=article&id=700:search-the-forum&catid=266:faq-videos&Itemid=50 Basically, for each dairy type you would: a. Import the Industry Spending Pattern for the dairy sector (Activity Options > Import > Industry Spending Pattern), then go to Event Options > Change All > Normalize Events. At this point, the coefficients sum to 1.00 and represent the proportion of the non-payroll budget that is spent on each input. Edit these coefficients as necessary. After making the edits, you'll need to normalize again: go to Event Options > Change All > Normalize Events. Finally, click Activity Options and set the Activity Level to the non-payroll portion of the expenditures. This will give you the indirect effects. b. Run a Labor Income Change Activity for the payroll portion of the budget. This will give you the induced effects. c. The dairy sales and associated employment and labor income are the direct effects.
    0
    Comment actions Permalink
  • Avatar
    jeffreyohara
    Thanks for that response. I also know the amount that was paid in real estate taxes from the budget. So, in addition to excluding labor costs from the non-payroll budget, am I correct by not including real estate taxes also? I can then modify the "tax impact" section of the results to provide a more accurate indication of the tax impact of the sector.
    0
    Comment actions Permalink
  • Avatar
    jenny
    That is exactly correct.
    0
    Comment actions Permalink
  • Avatar
    jeffreyohara
    Hi Jenny Thanks for your help. I think I am understanding this better. I wanted to confirm three aspects of your answer to part a to the response below (calculating indirect impacts from analysis-by-parts). Question 1 -- margining. Has margining for this already occurred? When I go to Event Options => Edit Event Properties, the "Margins" tab is grayed out. I don't see an ability to incorporate margining in any other tab under Event Options. Is this because margining has already occurred and I don't need to do anything else to reflect this? Question 2 -- normalizing. When I look at the non-payroll coefficients before normalizing they sum to a number of less than one. I trust this is because labor is not included. However, I need to normalize because an "Industry Spending Pattern" change is strictly reflecting the impacts of relative shifts in non-payroll expenditures. So, if I did not normalize the indirect impacts would be underreported? Question 3 -- purging imports. It appears that the default is for the "local purchase percentages" to be set at SAM model values, so they already incorporate that some input purchases are from out of state. So, there is no further work required to make sure import purchases are purged from the calculation after these "local purchase percentages" are established?
    0
    Comment actions Permalink
  • Avatar
    DougO
    1) Yes, since the Social Accounts Matrices data is in producer prices, industry spending patterns and institutional spending patterns have already been margined. You do not need to margin. 2) Yes, the spending pattern does not include value added(labor income, profit and taxes on production). If you are specifying only the goods and services needed for production then you would need to margin. 3) Again, correct. If you knew how much spending for a product that was produced locally, then you could change the LPP, otherwise accept our default.
    0
    Comment actions Permalink
  • Avatar
    jenny
    1. Yes - Industry Spending Patterns are already margined. No further action related to margining is required. 2. That is correct. The Sum Of Event Values is less than one because the rest goes to Value-Added (which is comprised of Employee Compensation, Proprietor Income, Other Property Income, and Indirect Business Taxes). If you don't re-balance after editing some of the coefficients, the Sum Of Event Values will be higher or lower than the original, depending on the changes you made. If you are okay with that, then you do not need to rebalance; however, in this case, you would need to set the Activity Level to the [u]entire[/u] budget because if you set it to just the non-payroll portion of the budget, it will spend only 63% (or whatever proportion is your Sum of Event Values after making your edits) of that non-payroll budget, when you really want it to spend 100% of the non-payroll budget amount. In summary - if the Sum of Event Values is 1.00, set the Activity Level to just the non-payroll portion of the budget because the software will spend that entire amount on the spending pattern items. If the Sum of Event Values is less than 1.00, set the Activity Level to the entire budget becuase the software will spend just the Sum of Event Values proportion of that value on the spending pattern items. 3. Correct - there is no further work required to make sure import purchases are purged from the calculation once LPP has been set to SAM Model Value (the default when you import an Industry Spending Pattern).
    0
    Comment actions Permalink
  • Avatar
    jeffreyohara
    Thanks Doug and Jenny I guess the issue is that the financial accounting industry data that I have is essentially unmargined whereas the Industry Spending Patterns are already margined. So, making adjustments to margined coefficients with unmargined data seems like it might raise an issue. Is this something I need to be cognizant or careful about?
    0
    Comment actions Permalink
  • Avatar
    jenny
    Yes, you will want to margin your data so that the coefficients are comparable with IMPLAN's. The attached Excel spreadsheet contains two tab - one has the margins for every IMPLAN sector, while the other contains the codes for the various types of margin (i.e., household, industry, etc.) You will probably want to use the Industry margins (Type 2). The "Sector" column is the sector from which the purchase is made, while the "MarginSector" column lists all the sectors that get a piece (i.e., margin) of that purchase value. The margins for any given Type and any given Sector should sum to 1.00. [attachment=287]Margins.xlsx[/attachment]
    0
    Comment actions Permalink
  • Avatar
    jeffreyohara
    I am currently trying to compare Industry Spending Patterns by sector using "analysis-by-parts". I had planning on aggregating prior to calculating the indirect effects by changing Industry Spending Patterns. However, I am having problems aggregating. For example, I have financial data on expenditures on utilities, but I don't know the relative percentage that was spent on water-electricity-natural gas. So, I attempted to aggregate these sectors into one sector by going to Options=>Aggregation and then create a new "utilities" sector. When I then click on "Aggregate Model", however, everything under "Analyze" becomes grayed out and I lose the ability to calculate the indirect effects by changing industry spending patterns. If instead I click on "Close" (instead of "Aggregate Model")and then look at industry spending patterns, nothing is aggregated. Any guidance on the correct way to aggregate would be greatly appreciated. Thanks very much.
    0
    Comment actions Permalink
  • Avatar
    DougO
    After you aggregate you need to re-build the multipliers (you have changed the picture of the economy and there should be a note in red in the lower left corner noting that). Go to Options > Construct > Multipliers. Creating impacts will be available once the multipliers are there. I would prefer to distribute the utilities based on a proxy industry spending pattern (Explore > Social Accounts > Balance Sheet > View by: Industry Balance sheet then choose the industry) rather than aggregating the model.
    0
    Comment actions Permalink
  • Avatar
    jeffreyohara
    I noticed that in the margin excel spreadsheet that you sent me (see email message below) that for some type-sector combinations that the margins sum up to less than 1. Specifically, for Type 1 (household), the retail sectors between 319 and 331 are all less than 1. For example, if for Type 1 Sector 324 I see a margin of 0.282 for Sector 324, how is this supposed to be interpreted? Does this mean that 71.8% flows out of state? Yes, you will want to margin your data so that the coefficients are comparable with IMPLAN's. The attached Excel spreadsheet contains two tab - one has the margins for every IMPLAN sector, while the other contains the codes for the various types of margin (i.e., household, industry, etc.) You will probably want to use the Industry margins (Type 2). The "Sector" column is the sector from which the purchase is made, while the "MarginSector" column lists all the sectors that get a piece (i.e., margin) of that purchase value. The margins for any given Type and any given Sector should sum to 1.00.
    0
    Comment actions Permalink
  • Avatar
    DougO
    The retail and wholesale trade sectors are margin sectors. That is, the output of those sectors = Gross margin (rather than gross sales). Ideally, you would apply any expenditure to the manufacturing sector that satisfies that demand, that way the producer, the transportation and the margin sectors (retail and wholesale trade) can be assigned a value and the coeficients sum to 1.0. If you apply a value to a retail sector, the software does not know what commodities are being purchased. In this case it is assumed that the goods being purchased are being imported and only the average gross margin for the retail sector will be assigned. Hence there is only a single margin coeficient and it is less than 1. Assigning an expenditure to a retail sector would occur if: 1) You only want the impact of the retail sector 2) It is a package of goods (eg, souvenirs, gifts, or groceries) so you don't know who the manufacturers are and it is not worth the effort to itemize. 3) It is a rural region and you can safely assume that any producer value would be imported.
    0
    Comment actions Permalink
  • Avatar
    mrmoose64
    Doug, I'm working with Jeff O'Hara on this question, and we've got a Michigan state database we're working with. Where in the model could I check what the current margins are for local vs. nonlocal goods sold at, say, a grocery store (Sector 324), and how could I change them? Thanks,Michael Shuman
    0
    Comment actions Permalink
  • Avatar
    DougO
    You would have to look up the individual goods (eg, Frozen Food Mfg or Beet sugar Mfg) to see the margins associated with each of the margins. We do have a vector that represents a collection of "Groceries", but they are already in producer prices - ie, they do not need to be margined - Setup Activities > Activity Options > Import > From a Model - then browse to the appliance:\Implan User Data\Utilities directory and select the dummy model: PCE_Patterns_for_NIPACategory_MIG_Lib440.impdb and choose the activity: "Food for off-premise consumption".
    0
    Comment actions Permalink
  • Avatar
    DougO
    Unless you are asking to see what the average margin is for the grocery sector itself. In that case, create an event for the retail sector (eg 324) plug in a value, choose gross sales. Then under Event Options > Edit Event Properties > Margins > Edit. The software will then show you what ratio of the gross sales is gross margin.
    0
    Comment actions Permalink

Please sign in to leave a comment.