Direct Effect Multiplier

Thanks. I'm using employment, payroll, and purchase/contracts values (it's dollar value but not employment) as input to evaluate the impact. Am I double counting the effect? If so, how can I avoid it?

You will not be double-counting. You can run this analysis one of two ways: a. If you know the line item purchases, you can run two activities: 1) a Labor Income Change with your known payroll figure (New Activity > Labor Income Change) and 2) an Industry Change Activity with an Event for each purchase. b. If you don't know exactly what was purchased (i.e., you just have the total purchase amount), you can use IMPLAN's built-in military spending pattern (Activity Options > Import > Institution Spending Pattern > Federal Government Defense). Click on the Activy to highlight it then click Activity Options and set the Level to the total sum of payroll + purchase/contracts. You are using the entire expenditure amount in this case (and not using a separate Labor Income Change Activity) because the spending pattern includes a line item for payroll (commodity 3440). You may find this paper helpful: https://www.google.com/url?q=http://implan.com/v4/index.php%3Foption%3Dcom_docman%26task%3Ddoc_download%26gid%3D201%26Itemid%3D60&sa=U&ei=syb_T_y9CpGYhQe23_TDBg&ved=0CAUQFjAA&client=internal-uds-cse&usg=AFQjCNHpYvWgemjBbcLAGQNxwy9-ZtnF_Q Regarding the question in your first post, how were you running the analysis that you found the direct employment effect to be greater than the employment in the base?

thank you, Jenny. In the first post, what i was doing is exactly the same as the first method you just recommend (since I know each purchase item). The only difference is I define employment and payroll under industry change activity rather than labor income change. I guess this may be the reason that I got weird result. I will redo it following your method.

I'm changing the activity setup following the first method you recommended. After I choose new activity > Labor Income Change, I find that there are only two sectors to choose to create the New Event, 5001 Employee Compensation or 6001 Propriet Income. Does it mean that the industry where employees earn income does not matter in the impact study? [quote="Jenny Thorvaldson" post=13362]You will not be double-counting. You can run this analysis one of two ways: a. If you know the line item purchases, you can run two activities: 1) a Labor Income Change with your known payroll figure (New Activity > Labor Income Change) and 2) an Industry Change Activity with an Event for each purchase. b. If you don't know exactly what was purchased (i.e., you just have the total purchase amount), you can use IMPLAN's built-in military spending pattern (Activity Options > Import > Institution Spending Pattern > Federal Government Defense). Click on the Activy to highlight it then click Activity Options and set the Level to the total sum of payroll + purchase/contracts. You are using the entire expenditure amount in this case (and not using a separate Labor Income Change Activity) because the spending pattern includes a line item for payroll (commodity 3440). You may find this paper helpful: https://www.google.com/url?q=http://implan.com/v4/index.php%3Foption%3Dcom_docman%26task%3Ddoc_download%26gid%3D201%26Itemid%3D60&sa=U&ei=syb_T_y9CpGYhQe23_TDBg&ved=0CAUQFjAA&client=internal-uds-cse&usg=AFQjCNHpYvWgemjBbcLAGQNxwy9-ZtnF_Q Regarding the question in your first post, how were you running the analysis that you found the direct employment effect to be greater than the employment in the base?[/quote]

I just got the results following the method a. This time the direct employment number equals the number if I just use purchase data as the inputs. This means that the income and employment in the military base doesn't create any job. Does it make sense?

Neither of these two approaches includes the direct employment (i.e., the employment at the base itself). Both approaches give you the indirect employment (from the input purchases) and the induced employment (from the spending of wages). (Note that in the case of option a, the reported "direct" effects are truly the first round of indirect effects). So in either case, you will need to add the direct employment to the total impacts. Ditto for direct Labor Income and direct expenditures.

Clear. Thanks. In the method a. If I know the employment and payroll for civilian and military employees respectively, can I model then by using Industry Change rather than Income Change?

There is no military industry sector to model as you might an auto plant or service sector. There is a "Federal Government Defense" institution that you could import, setting the activity level equal to the overall military budget for the base. Note that this institution "buys" its payroll as a payment to sectors 439 and 440. The "direct" employment will not only indclude 439 and 440 employment but also the employment associated with the institution's spending pattern, which is more correctly considered indirect employment. My preference is to add the direct employment and payroll, which you know, to the results of method a.

Where can I find Federal Government Defense Institution and how can I import it? [quote="Doug Olson" post=13381]There is no military industry sector to model as you might an auto plant or service sector. There is a "Federal Government Defense" institution that you could import, setting the activity level equal to the overall military budget for the base. Note that this institution "buys" its payroll as a payment to sectors 439 and 440. The "direct" employment will not only indclude 439 and 440 employment but also the employment associated with the institution's spending pattern, which is more correctly considered indirect employment. My preference is to add the direct employment and payroll, which you know, to the results of method a.[/quote]

Discard the question I just posted. I found it. I imported the spending pattern. Under the Events window, I see many industrial sectors with their corresponding coefficients, event year, and local purchase percentage. Now my question is: how do I use this table given that I know the civilian & military employment, civilian & military payroll, military base contracts and purchases (itemized) measured in dollar value?

As was mentioned in post #13381, you need to set the Activity Level to the total budget (intermediate expenditures + payroll). You can do this by going clicking the Activity to highlight it and then clicking the Edit Activity button. When you create the Scenario for this Activity, you will want to leave the Scenario Level at 1.00. The resulting impacts represent the economic activity that occurs as a result of those expenditures. They will not include the employment at the base.

[quote="Jenny Thorvaldson" post=13389]As was mentioned in post #13381, you need to set the Activity Level to the total budget (intermediate expenditures + payroll). You can do this by going clicking the Activity to highlight it and then clicking the Edit Activity button. When you create the Scenario for this Activity, you will want to leave the Scenario Level at 1.00. The resulting impacts represent the economic activity that occurs as a result of those expenditures. They will not include the employment at the base.[/quote] thanks. Your instruction is very useful.