IMPLAN Pro: Analysis-by-Parts Spending Pattern Basic Example

Example Analysis-by-parts:

We want to demonstrate a typical example of an ABP. In this example, we will show the annual operational impact of a 500 megawatt wind farm. Sector 31 Electric Power Generation and Distribution has a national average production function for all types of power generation (wind, natural gas, oil, nuclear, hydro) as well as distribution.

I wish to take advantage of detailed spending patterns available in the NREL worksheets (http://www.nrel.gov/analysis/jedi/download.html ) to perform an analysis-by-parts to analyze the operational impacts with data specific to wind power.

Based on the data found in the worksheet, I have created the spending pattern shown below which is on a per 100 MegaWatt basis. The NREL spreadsheet indicates that the values are in $2008.

Using table C1 to the right and the Washington County model created in task A, we can perform the analysis by parts.

Steps:

1. Under setup activities, choose New Activity > Industry spending Pattern. Name the new activity "Wind Power spending per MW". Set the Activity Level to 32,955,194. Click "Save".
2. Create a new event by clicking the "New Event" button and then choosing commodity 3031 (the primary commodity produced by industry 31).
3. Before entering the coefficient value change the event year to 2008 (Event Options > Edit Event Properties > Event Year).
4. Enter the Coefficient value 0.00105
5. Change the LPP: Event Options > Edit Event Properties > Local Purchase Percentage > Set to SAM Model Value. We know that wind power is generated 100% locally, but we don't know from where the goods and services are purchased.
6. Create new events for each of the other sectors 3039 through 3386. Note that each of the subsequent events will have event year 2008 and an LPP that is not 100% (ie, will be based on the SAM model value). Also, one or more of the commodities will not exist locally, so that value will not generate any local activity and the LPP will be 0. Check that the Sum of Event Values is 0.27.
7. Now we can create the Labor income activity. Choose New Activity > (type of activity) Labor Income Change. Name the activity "Wind Power Labor per 100 MW" and leave the Activity Level at 1.
8. The only new event will be 5001 Employee Compensation. Set the event year to 2008 before plugging in $446,504 for the Labor Income Value.
9. Run the analysis with a new scenario named "Wind Power 500 MW", setting the Scenario level to 5. Choose the two activities: "Wind Power spending per 100 MW" and "Wind Power Labor", then click "Analyze Single Region".
10. View the results with 2012 selected as the Dollar Year for View.

Discussion of Results:

We have indirect and induced effects. We still need to calculate the direct effects and add them to the table (based on values found in Table C1 and deflators found in the software).

• Employment: 6 per 100 MW * 5 gives direct employment of 30.
• Output: 32,955,194 per MW * 5 * 1.018 (2008 to 2012 output deflator) gives direct output of 167,741,937.
• Labor Income: 446,504 per 100 MW * 5 * 1.088 (2008 to 2012 GDP deflator) gives direct of 2,428,982.
• Value Added: 24,208,533 * 5 * 1.088 gives a direct value added of 131,694,420.
• Table C3, below, adds the direct effects to our final results.

As an analyst, it is important to look at what turns out to be the most import indirect effect - interest. Is this a local corporation borrowing from a local bank (total loan annual repayment of over $20 million), or is it a large outside corporation that happens to be locating here and borrowing from an out of region bank? Setting the LPP to 0 for interest would cancel the largest component of the indirect impact.