IMPLAN Pro: Analysis-by-Parts Manually Margining Spending Patterns

Introduction:

When you have a line-item budget available, you may want to create a spending pattern unique for your Industry. It is important in these cases to remember that Capital Expenditures should not be included in the spending pattern, but there may still be general expenditures that are purchased by businesses but are sourced through retail or wholesale venues. Since the prices paid to a wholesaler or retailer are purchaser prices and not producer prices, adjustments need to be made to the sales value.

Detailed Information:

Analysis-by-Parts can be an incredibly useful technique for digging deeper into your firm's budgetary expenditures, but when we look at retail or wholesale purchases, it is important that these purchases are dealt with carefully. The reason for this is that IMPLAN understands values only expressed in producer prices, but a retail or wholesale sales value (purchase price) includes not only the production component of the retail or wholesale venue but also the production of all the elements of the Value Chain that proceed it. Thus the sales value of a retail sale is a combination of the Retail Value, Wholesale Value, Transportation Value, and Producer Value. The wholesale sales value is actually a combination of the Wholesale Value, Transportation Value, and Producer Value. Thus, if we create a coefficient of spending for either a retail or wholesale component without Margining, or if we apply the entire value to the production Sector's coefficient, we will be mis-estimating the impacts on those respective Sectors.

There are two possible situations that might arise:

1. Purchases are known to be retail or wholesale, but the specific items purchased are unknown.
2. Purchases are known to be specific items, but the value associated to them is a wholesale or retail value.

These two situations will be handled in the Usage section in two different examples using simplified spending patterns as examples.

The other caveat to using Margins with spending patterns is that the normal Margining features are not available for spending patterns in IMPLAN, thus we have to use tables provided in the downloads section of the IMPLAN website to determine how we should deconstruct our sales value so that we can manually Margin our spending patterns for Analysis-by-Parts.

Usage:

Example 1: Retail or Wholesale Purchase Value Is Known, but the Item Purchased Is Unknown

Our flour milling Industry buys office supplies as part of its production function. We know that of their $15,000,000 Output value they spend $1,000 a year on office supplies at the local Office Retail Superstore and $21,600 on toiletry products through their Wholesaler. Our first inclination might be to simply create straight up coefficients based on the respective portions of total cost that these purchases represent. If we were to do so, we would represent these coefficients as (assuming the 536 Sectoring scheme):

Sector #	Sector Description	Dollar Value	Coefficient (Dollar Value/Output)
406	Retail- Miscellaneous Retail Stores	$1,000	0.00006667
395	Wholesale Trade	$21,600	0.00144000

However, as stated above, this would be attributing too much value to the Retail and Wholesale Sectors involved.

In order to determine what the Margin should be for each of these Sectors, we need to adjust the sales value further by determining what portion is kept by the retailer and wholesaler respectively. Note that in this circumstance, because we do not know the items purchased, the value that cannot be attributed to the retailer or the wholesaler is leaked from the spending pattern.

We will get the Margin value from the "2013 Common Margins" Spreadsheet that can be downloaded through the Downloads>536 Section of the Website. When we open the Excel Spreadsheet, we will see the following information:

For our example, we do not know exactly what is being purchased, so we are going to highlight Column A and use the Find function in Excel to search for 395-Wholesale Trade. We will then see the portion of the table captured below:

Note that Retail and Wholesale Sectors (395-407) have the same Sector designation in Columns A and B and that unlike the Sector 2 sample above, the coefficients will not sum to 1.00. This is again because we do not know what we are purchasing. Therefore, the only 'local' production we can account for is that of our local Retailer and Wholesaler.

With this new information, we will now reformulate our coefficient calculation to include the Margined values captured from the spreadsheet above.

Sector #	Sector Description	Dollar Value	Margin	Margined Sale Value (Dollar Value * Margin)	Coefficient (Margined Sale Value/Output)
406	Retail- Miscellaneous Retail Stores	$1,000	0.454	$454.00	0.00003027
395	Wholesale Trade	$21,600	0.174	$3758.40	0.00025056

Thus when we added these components to our spending pattern they would be represented as their Margined coefficients.

*Note that more than one component in a spending pattern will likely have Wholesale and Retail purchases. It is usually best, unless you want to create a single Event for each Wholesale and Retail Margin, to sum these coefficients into their total component. We will demonstrate this, in part, in Example 2.

Example 2: Item Purchased Is Known, but the Value Is Expressed in Retail or Wholesale Costs

Our flour milling Industry also buys grain as one of the primary components of its production. Of their $15,000,000 annual Output, they spend $8,400,000 on grain, but they purchase the grain through a wholesale food provider.

Just like above, we cannot directly attribute the 8.4 million dollars of production to Sector 2 Grains, but unlike in Example 1, we know exactly what our flour mill is purchasing in this circumstance. But now there is also an added caveat.

When we look at the splits for Sector 2 again, note that all elements of the Value Chain are included in the spreadsheet. Thus we see Margin splits for Retail Value, Wholesale Value, Transportation Value, and Producer Value. However, we said in our circumstances we know there is no retailer because the purchase was made through a wholesaler. Since the purchase was made through a wholesaler and not a retailer, we will need to recalculate the Margin splits by adjusting the provided values so that the portion of the Margin assigned to the retailer in the table is redistributed into the wholesale, transport, and production elements of the Margin.

To do this we will need to zero out the retail Margin and normalize the remaining Margin values.

We can normalize the remaining Margins as follows:

Margin Sector	Margin Type	Margin Value	Normalizing Calculation	Resulting Margin Adjusted for No Retailer
2	Producer	0.501761261	0.501761261/ 0.764493558	0.656331574
395	Wholesale	0.122895003	0.122895003/ 0.764493558	0.160753484
400	Retail	0.00
408	Transport	0.003169934	0.003169934/ 0.764493558	0.0041466449
409	Transport	0.056842672	0.056842672/ 0.764493558	0.074353369
410	Transport	0.032852038	0.032852038/ 0.764493558	0.042972289
411	Transport	0.046972651	0.046972651/0.764493558	0.061442835
	Sum of Margin Values:	0.764493558	Normalization Check Sums to:	1.00

Now that we have adjusted the Margins to account for the entire sales value but not include a retail component, we can apply the Margin values to the grain purchase to determine how much of the grain purchase goes to each of these respective Sectors.

Margin Sector	Margin Type	Dollar Value	Resulting Margin Adjusted for No Retailer	Margined Sale Value (Dollar Value * Margin)	Coefficient (Margined Sale Value/Output)
2	Producer	$8,400,000	0.656331574	$5,513,185.22	0.36754568
395	Wholesale	$8,400,000	0.160753484	$1,350,329.27	0.09002195
408	Transport	$8,400,000	0.0041466449	$34,831.82	0.00232201
409	Transport	$8,400,000	0.074353369	$624,568.30	0.04163789
410	Transport	$8,400,000	0.042972289	$36,096.72	0.02406448
411	Transport	$8,400,000	0.061442835	$516,119.81	0.03440799

If these were the three components of our spending pattern, we would then sum the wholesale components into a final spending pattern result that would include an aggregate value of wholesale purchases for all commodities.

395 from Toiletries	0.00025056
395 from Grain	0.09002195
Total	0.09027251

While this represents a simplified spending pattern, the same principles apply to Margining larger more complete spending patterns.

 

Some Additional Considerations

Are you having trouble using our Help>Sector Search feature?

The Sector search can be sorted by clicking the Industry Code column. You can then scroll down through the Industry codes until you find the codes you are interested in examining. The associated NAICS descriptions will show you the types of Activities that will report into each category. This is often helpful in determining Sectors.

What about the Local Purchase Percentage?

This should be set to SAM Model Value. This will allow the Model to make estimates of local purchasing ability of all commodities in your spending pattern. For spending patterns, we have an Event Options> Change All> Local Purchase Percentage> SAM Model Value option that will set all the Events to SAM Model Value at the same time.

Thinking on transport.

Typically there are two legs of transport: between the producer and the wholesaler and between the wholesaler and the retailer. The transportation expenses should be accrued into a single Sector. If you are using the IMPLAN Margin splits as outlined in this article, both 'legs' of transportation are included in the respective Margin components.

Zeroing out the retail margin.

In this case, we re-normalize across all Margin Sectors. However, this is not the only option, though it is typically the most conservative one. In contrast, you could argue that the wholesaler and transporters do not get a larger cut when there is no retailer. Perhaps it is more likely that the producer gets to keep the difference or splits the difference with the wholesaler. You can make these adjustments based on what you know about the Industry, and then state how you handled the normalization process in your results.