Modelling water cost changes

Comments

21 comments

  • Avatar
    IMPLAN Support
    Hi Roger. You are correct in how you might want to model the impacts of a region-wide increase in water costs on households. That is, you would treat an increase in water costs as a decrease in discretionary income, inputting the total cost change to the household/Institution Sector 432. For industries and institutions, however, the increase in water costs would represent an increase in their production costs as you say. Without modifying the production functions of each industry that produce this commodity, you could apply the amount of the rate increase to Sector 33 (for private industry) to capture the output, income, and employment effects of a water cost hike on industries and institutions. I hope this answers your question. Please let us know if we can be of further help.
    0
    Comment actions Permalink
  • Avatar
    TCW Economics
    Thank for the response. Just to clarify what you're suggesting, wouldn't the model interpret the inputting of the water cost change to Sector 33 as a change in final demand for water rather than as a cost increase? Or would the results be the same? Also, the water in question is being produced by a public utility. So should the change in water costs be input to Sector 33, which I understand includes private utilities, or the "Other state/local enterprises" sector? Thanks again!
    0
    Comment actions Permalink
  • Avatar
    IMPLAN Support
    Hi Roger. In response to your follow up questions and after further consultations with another colleague here at IMPLAN, we think another alternative (and perhaps better) might be to import and “Institutional Spending Pattern for household income group $35- $50K. With this spending pattern, you would look at the water rate increase and how you would use the Activity Level to examine changes in overall household income or spending on water. To determine what the effect of this reduced spending will have on the Study Region. 1. Return to the Setup Activities screen. 2. Activity Options>Import>Institution Spending Pattern>Select $35- $50K from the library. • The Activity will successfully import. Click OK in the pop-up screen. • An Institution Spending Pattern of households in the $35- $50K” range will appear in the Activities screen. • Click on the Activity. The Events list will display. 3. Edit the spending pattern to remove expenditures like rent and utilities, which will not change from the rate increase. Select Edit Activity from the top of the screen. Change the Activity Level to the amount of decrease in household discretionary income due to the rate increase. The value entered here should be negative to reflect a loss. Create a Scenario to analyze the effect of this proposed mill levy. • Click New Scenario. Name the Scenario "Water Rate Increase", leave Level at 1.00 and Save. Select the Activity Institution Spending Pattern. 6. Analyze for Single Region 7. View Results. To model the impact of a water rate increase on local industries, you could use an “Industry Spending Pattern” for Sector 33. Import the Industry Pattern for Sector33 (Water, sewage, and other treatment and delivery services) to model the effects of the water rate increase on local industries. To determine what the effect of this reduced spending will have on the Study Region. 1. Return to the Setup Activities screen. 2. Activity Options>Import>Industry Spending Pattern>Select Sector 33 from the library. • The Activity will successfully import. Click OK in the pop-up screen. • An Industry Spending Pattern for Sector 33 will appear in the Activities screen. • Click on the Activity. The Events list will display. 3. Edit Activity Level to show the investment value for the project. Select Edit Activity from the top of the screen. Change the Activity Level to the amount of decrease in proprietor income (increase in cost due to the rate increase). The value entered here should be negative to reflect a reduction in income. Create a Scenario to analyze the effect of this proposed mill levy. • Click New Scenario. Name the Scenario "Water Rate Increase", leave Level at 1.00 and Save. Select the Activity Industry Spending Pattern. 6. Analyze for Single Region 7. View Results. Note: That the total Event value is less than 1.0, and that LPPs are set to the SAM model values. • The Event value indicates the percent of the increased cost of water in Sector 33 that goes to purchasing goods and services which require water. The remaining difference therefore is the Value Added portion of the impact. For your project, you would account for this remaining difference by using a Labor Income change to capture the impact, which is explained in the next section of our reply. Industry Spending Patterns can also be imported from IMPLAN's library using the Activity Options. Like custom created spending patterns, Labor Income and direct effect will still need to be figured and added to the total impact. To create a Labor Income Activity to model the remaining effects. 7. Return to the Setup Activities screen. 8. Create a Labor Income Activity to model the impact of the remaining difference of Sector 33 of the water rate increase on local industries. • Click New Activity. Select Labor Income from the Activities list, set level to the difference between one and the sum of the Event Level and name the Activity "Water Rate Increase LI" and save. You are choosing to model using a Labor Income impact because the Study Region has not specified what the average annual income of their employees will be. Also they have not indicated where there workers reside, so for the LI scenario you will want to assume that all workers in Sector 33 for this project live within the Study Area. 10. Choose Sector 6001 Proprietor Income, and set the Labor Income Value to the negative value of the decrease in proprietor income as a result of the water rate increase. 11. Create a Scenario to model the impact of the Proprietor Income. • Click New Scenario. Name the Scenario "Water Rate Increase LI", leave Level at 1.00 and save. Select the " Water Rate Increase LI" Activity. 12. Analyze for Single Region 13. View Results. In this scenario of Labor Income impacts, we are tracking the lost spending of Proprietor Income, so we are tracking only induced effects. Remember that you still need to add in the effect of the actual output and employment of Sector 33. In this instance, if you do not have actual employment for your Study Region, you could estimate the direct employment impact by entering the negative amount of lost Proprietor Income into Sector 33 and use IMPLAN's employment estimate, which will automatically populate in the Employment Field once the negative value of proprietor income is entered. We hope this provides more clarity on how you might set up this scenario.
    0
    Comment actions Permalink
  • Avatar
    TCW Economics
    Thank you very much for the detailed response. This does seem like a better approach. I will give it a try and let you know if I have further questions. I greatly appreciate your assistance.
    0
    Comment actions Permalink
  • Avatar
    TCW Economics
    After trying to work through the methodology you provided for modeling water cost effects, I have a few follow-up questions. For modeling household effects (the first section of your response), #3 says to edit the spending pattern in the events list to remove expenditures such as rent, utilities, etc. When I go to the events list, I get a lengthy list of sectors with coefficients. I assume that there may be several sectors for which spending may not change due to a water cost increase. Do I need to set the coefficients for all these sectors to zero? And wouldn't the event values then sum to less that 1.0? For modeling labor income effects (the third section of your response), #10 says to "choose Sector 6001 Proprietor Income and set the Labor Income Value to the negative value of the decrease in proprietor income." Two questions: 1) When I highlight the labor income activity that I've created, nothing appears in the events box. So where do I choose Sector 6001 to set the Labor Income Value? and 2) Is the "decrease in proprietor income" value that I'm supposed to enter represented by the total proprietor income change from the industries assessment (part two of your response)? Finally, for the task of adding in the output and employment effects for Sector 33 (the last paragraph of your response), is the "lost Proprietor Income" that I'm to enter into Sector 33 also represented by the total proprietor income change from the industries assessment (part two of your response)? Thanks again for your help.
    0
    Comment actions Permalink
  • Avatar
    IMPLAN Support
    Hi Roger. We apologize for the delayed reply, but our office has been closed for the last day and a half due the snow and ice we received during this period. The detailed set of instructions sent to you earlier were meant to help familiarize you with using and editing Institutional Spending Patterns and doing Analysis-By-Parts (ABP) methodology, while providing some guidance in framing the problem. We will review your questions and build an example model to help illustrate our points and make them a little easier to follow. We will send you a reply to your questions as soon as possible.
    0
    Comment actions Permalink
  • Avatar
    IMPLAN Support
    Hi Roger. Our response to your questions are embedded below in your Forum Post, which is included below. After trying to work through the methodology you provided for modeling water cost effects, I have a few follow-up questions. For modeling household effects (the first section of your response), #3 says to edit the spending pattern in the events list to remove expenditures such as rent, utilities, etc. When I go to the events list, I get a lengthy list of sectors with coefficients. I assume that there may be several sectors for which spending may not change due to a water cost increase. Do I need to set the coefficients for all these sectors to zero? And wouldn't the event values then sum to less that 1.0? (Yes. You would enter a zero in those sectors that you feel would not be impacted by a water rate increase such as rent, utilities, etc). For modeling labor income effects (the third section of your response), #10 says to "choose Sector 6001 Proprietor Income and set the Labor Income Value to the negative value of the decrease in proprietor income." Two questions: 1) When I highlight the labor income activity that I've created, nothing appears in the events box. So where do I choose Sector 6001 to set the Labor Income Value? Roger, if you select the Labor Income Activity that you created and then click on it, in the right corner of the cell you should see a dropdown selection to allow you to select “Proprietor Income.” and 2) Is the "decrease in proprietor income" value that I'm supposed to enter represented by the total proprietor income change from the industries assessment (part two of your response)? It is to reflect the loss/decrease in proprietor income from the water rate increase. Finally, for the task of adding in the output and employment effects for Sector 33 (the last paragraph of your response), is the "lost Proprietor Income" that I'm to enter into Sector 33 also represented by the total proprietor income change from the industries assessment (part two of your response)? Yes. Thanks again for your help.
    0
    Comment actions Permalink
  • Avatar
    TCW Economics
    Thank you for the additional details. I'll give it try and see how it works out. Hope you're thawing out back there.
    0
    Comment actions Permalink
  • Avatar
    IMPLAN Support
    Thanks Roger. We are glad to help.
    0
    Comment actions Permalink
  • Avatar
    TCW Economics
    I've been working through the modeling process you laid out in your previous responses, and I have one final question for you. For the task of adding in the output and employment effects for Sector 33 (the last paragraph of your Feb 11th response), is the "lost Proprietor Income" that I'm to enter into Sector 33 the same as the proprietor income change [u]for Sector 33[/u]from the industries assessment (part two of your response), or is it the total proprietor change (across all sectors)? Thank you.[b][/b]
    0
    Comment actions Permalink
  • Avatar
    IMPLAN Support
    Hi Roger. We may have responded to and earlier Forum Post from you and not your most recent Post. We apologize for the oversight. However, in our reply to you below, we may have still answered your question. If not, please let us know. We embedded my reply to your post in with your questions. Hopefully, this is not confusing. After trying to work through the methodology you provided for modeling water cost effects, I have a few follow-up questions. For modeling household effects (the first section of your response), #3 says to edit the spending pattern in the events list to remove expenditures such as rent, utilities, etc. When I go to the events list, I get a lengthy list of sectors with coefficients. I assume that there may be several sectors for which spending may not change due to a water cost increase. Do I need to set the coefficients for all these sectors to zero? Yes. If you feel that those sectors also would not be impacted by the rate increase. And wouldn't the event values then sum to less than 1.0? Yes. If you feel there are other Sectors that would not be impacted by this rate increase. We forgot to mention in earlier post but you would need to rebalance for event values to sum to one. You can navigate to Setup Activities>Edit Options>Change All>Normalize Events>. The Normalize Events function balances or redistribute the deleted Sectors’ coefficient values proportionately across all the remaining Sectors so that the Sum of Event value sums to one. For modeling labor income effects (the third section of your response), #10 says to "choose Sector 6001 Proprietor Income and set the Labor Income Value to the negative value of the decrease in proprietor income." Two questions: 1) When I highlight the labor income activity that I've created, nothing appears in the events box. So where do I choose Sector 6001 to set the Labor Income Value? [b]To the immediate left of “Labor Income” field you should see a down arrow key. Click on the down arrow and you should see a selection for “Proprietor Income”, Click and space over to “Labor Income Column” and enter the negative Proprietor Income value. 2) Is the "decrease in proprietor income" value that I'm supposed to enter represented by the total proprietor income change from the industries assessment (part two of your response)? Yes. It is whatever the increase in water cost to businesses in the Study Area. We are assuming that the businesses impacted by the rate increase will not pass along the increase cost to consumers and that they will simply consider it as reduce/lost proprietor income, at least in the short to immediate run. [/b] Finally, for the task of adding in the output and employment effects for Sector 33 (the last paragraph of your response), is the "lost Proprietor Income" that I'm to enter into Sector 33 also represented by the total proprietor income change from the industries assessment (part two of your response)? [b]Yes. It is whatever the increase in water cost to businesses in the Study Area. When you do this, not only will you get the employment associated with the lost proprietor income but output as well. [/b]
    0
    Comment actions Permalink
  • Avatar
    TCW Economics
    Thanks for your response. It was helpful. As I now try to better understand what I'm getting from the modeling process that you've provided, another question came to mind. For the following part of the methodology that you provided, as follows: >>To model the impact of a water rate increase on local industries, you could use an “Industry Spending Pattern” for Sector 33. Import the Industry Pattern for Sector33 (Water, sewage, and other treatment and delivery services) to model the effects of the water rate increase on local industries.<< Does the modeling output from this represent the direct change in industries that purchase water from Sector 33? If so, is that change happening because they're reducing their industrial output in response to increased water costs? If not, what is the model assuming that results in the reported impacts? Thanks again.
    0
    Comment actions Permalink
  • Avatar
    IMPLAN Support
    Hi Roger. The answers to the first two questions of your Forum Post are yes, which the third question a non-issue. Please remember, in reporting your results to state what your assumptions were in arriving at your results. Initially, I think we said that you could treat the water rate increase on households as a loss in Household Income and the rate increase on businesses in the Study Area as a loss of Proprietor Income, choosing not to pass along the rate increase to their consumers in the region. This is one possible scenario. Another possible scenario, and it sounds like the direction you are taking is to treat the rate increase on households as lost income and the rate increase on businesses as increased cost of doing business that will be pass along to consumers. That is certainly a realistic option, and we are not sure how businesses would respond. Again, we suggest you clearly state your assumptions in reporting the results. This will help others reviewing the results to understand how you derived your results. We hope this helps.
    0
    Comment actions Permalink
  • Avatar
    TCW Economics
    Thank you again. I appreciate you hanging in with me as I try to sort out this complicated impact assessment. So to summarize, and to make sure I'm understanding how to explain my modeling results in my report: 1. The modeling assumption underlying the results from using an Institutional Spending Pattern for households is that an increase in water costs will decrease the amount of income available to spend on other goods and services in the region. 2. The modeling assumption underlying the results from using an Industry Spending Pattern for Sector 33 is that an increase in the cost of water produced by Sector 33 will decrease proprietor income in all industries that purchase water from Sector 33. (This one confuses me a bit, because I thought an Industry Spending Pattern reflects what a specific industry spends on inputs for producing its products. So it seems to me that inputting the increased water cost to an Industry Spending Pattern for Sector 33 would generate only backward-linked impacts. In other words, the results would reflect the decrease in spending by Sector 33 on inputs other than water as a result of the increased cost of the water input. Am I wrong here?) 3. The modeling assumption underlying the results from using a Labor Income Activity is that reduced economic activity caused by a reduction in proprietor income for all water-dependent industries (calculated in Step 2)would result in lost labor income in all of those industries, which wasn't picked in the results of Step 2. If you could verify or revise what I've said above, as well as respond to my parenthetical questions under Step 2, I'd greatly appreciate it!
    0
    Comment actions Permalink
  • Avatar
    IMPLAN Support
    Hi Roger. The main take away here is that your project and assumptions you make help determine how you frame the problem(s) and what might be the appropriate methods to use in addressing these issues with the IMPLAN model. Here is a brief reply to your Forum Post. Our responses are shown below each question. 1. The modeling assumption underlying the results from using an Institutional Spending Pattern for households is that an increase in water costs will decrease the amount of income available to spend on other goods and services in the region. Reply That is not an underlying assumption of the Institutional Spending Pattern itself. The assumption(s) are made on the part of the analyst. Given the change/impact of a water rate increase on households, one assumption is that this increase in water cost will result in a decrease or loss in household income because they are now having to pay more for water. If this is in fact try, you are further assuming that they will respond by reducing their demand for goods and services by a similar amount. If the amount of the rate increase is fairly nominal, then households my not change their spending habits at all. Or if the rate increase fairly significant, they may still spend the same amount by drawing down savings, investments, charging on credit cards, and/or borrowing to maintain their spending. It really depends on how significant the rate increase is for both households and businesses. The Institutional Spending Pattern allows you to model any or all of these scenarios by allowing you customize it to your need. Much of what we said about households apply to businesses as well. It really depends on how significant the rate increase is on businesses as to how they might respond. Obviously, you might want to run several scenarios to illustrate the possible actions that businesses and household may take. 2. The modeling assumption underlying the results from using an Industry Spending Pattern for Sector 33 is that an increase in the cost of water produced by Sector 33 will decrease proprietor income in all industries that purchase water from Sector 33. (This one confuses me a bit, because I thought an Industry Spending Pattern reflects what a specific industry spends on inputs for producing its products. So it seems to me that inputting the increased water cost to an Industry Spending Pattern for Sector 33 would generate only backward-linked impacts. In other words, the results would reflect the decrease in spending by Sector 33 on inputs other than water as a result of the increased cost of the water input. Am I wrong here?) Reply Again, almost everything that I said in the response to question #1 can apply to question #2. This is not an underlying assumption of the Industry Spending Pattern itself. The assumption(s) are made on the part of the analyst. Based on the analyst assumptions and what we know about your project, the Industry Spending Pattern appear to be an appropriate method to use in modeling your situation by allowing you customize it to your need. As far what the industry spending pattern reflects, you are correct in that it reflects what a specific industry spends on inputs for producing its products. For your project, you are assuming that Sector 33 best reflect the activities water and water produced by water systems. You are also correct in that if you enter a negative value in the event field for this sector and model it, you should see a decline in output, employment, etc., the size depending on how significant the change is. The decline in output, in your case, water sales, is intended to reflect a higher cost of obtaining this commodity. Finally, yes. The effects would be backward linkages. If you knew exactly how firms might respond to the rate increase and how this might translate to other firms who purchase products from the firms experiencing the rate increase, then you could potentially model the impact of forward-linked industries. This would not be easy but it is something that could potentially be explored. 3. The modeling assumption underlying the results from using a Labor Income Activity is that reduced economic activity caused by a reduction in proprietor income for all water-dependent industries (calculated in Step 2) would result in lost labor income in all of those industries, which wasn't picked in the results of Step 2. If you could verify or revise what I've said above, as well as respond to my parenthetical questions under Step 2, I'd greatly appreciate it! Reply We recommended a Labor Income Change Activity because of the Analysis-of-Parts would allow you to capture the labor income and employment effects of reduced proprietor income of affected businesses in your Study Region. This is not an underlying assumption of the Labor Income Activity itself. But based on your assumptions and knowledge of your project, the Labor Income Activity would appear to be an appropriate method to use in modeling this part of your project and given the fact that Analysis-by-Parts allows you to split an impact into smaller parts as you are attempting to do. We want to emphasize again that you be clear in your report exactly which assumptions you are making and why.
    0
    Comment actions Permalink
  • Avatar
    TCW Economics
    Thank you for the response. I understand that I'll need to clearly state my assumptions for the assessment. However, I'm trying to better understand what the model is actually doing when I'm using the Industry Spending Pattern for Sector 33 to model the impacts of a water cost increase on regional industries. When I change the activity level for the Sector 33 Industry Spending Pattern to reflect the cost increase, and then analyze effects, is the model generating impacts on industries that purchase water [u]from[/u] Sector 33, or is it generating impacts on industries that [u]supply[/u] inputs to Sector 33? Thank you.
    0
    Comment actions Permalink
  • Avatar
    IMPLAN Support
    Hi Roger, We apologize about the delay in responding, but in reviewing your posts we are concerned that there is some confusion between what we are understanding the problem to be and what you are actually trying to do. We are now stepping back and trying to clarify. When you use the spending pattern for Sector 33 you are basically showing the increased expenditures that would be associated to an increase in production for Sector 33. This would be translating the value of the increased cost into increased production of inputs needed for water. Looking over your previous posts, we aren't sure that that is what you are trying to model, and we are afraid there might be some confusion in all this setup, likely because the model is really not designed to be used for the type of problem you may be trying to setup. It sounds like you may be trying to see how industries and institutions would be effected in your region if the cost of water increased. To do this you would need to do something similar to what you did with Households for each and every Industry and Institution that you feel will be impacted by the increase in water costs. But to do that you need to make some basic assumptions. The IMPLAN model is only able to look at changes in production assuming that technology and product mix remain static. So you would need to determine outside the software if: [ol] [li]The change would actually impact Industry spending, rather than just come from Industry profits. If the difference would be from profits. There is nothing to model in IMPLAN, as profits are considered leakages unless you know how they are spent, and/or that they are spent locally.[/li] [li]If the increase would actually affect production, i.e. will other Sectors purchase less water because of the cost? Will they decrease production? The change negative change in Labor Income would be appropriate if you thought that people would be losing jobs or working less hours because of decreased production.[/li] [li] If the consumer will be asked to absorb the cost, then it would be a situation of the spending pattern changing and the potential adjustment of other coefficients or the ratios of Intermediate Expenditures and Value Added changing, which as we described above will only have economic impact if it affects the total spent on Intermediate Expenditures (in which case seeing the increased production of water by the spending pattern for 33 would be appropriate). [/li] [li]Although it is unlikely in this case, if rate increases are high enough, it could be possible that companies would leave the region, or substitute another product instead of purchasing a specific commodity. You would then model the assumption of decreased production in the Industries that you think would be leaving the region.[/li] [li] Finally all these same principles would apply in a general sense to Institutional purchases, you would need to determine how you think the Institution would respond and whether or not it would affect their 'production' of a policy or program. Typically though Institutions don't have the same cause/effect responses to price changes, and also sometimes have contracts to restrict these changes, so they may not really see an impact that IMPLAN can model. [/li] [/ol] Unfortunately, we don't have a behavioral model that will allow us to predict how Industries or Institutions will actually respond to price changes, and these are very difficult responses to measure with an I-O Model, as you have acknowledged. Sometimes IMPLAN data is used in CGE models because these models have built in assumptions to try to predict behavioral responses. We are certainly happy to continue to assist you in trying to work with the model, but we want to ensure moving forward that we are clear in what both sides are describing, understanding and working towards. Again we apologize for what we see may be potential confusion here in the past posts.
    0
    Comment actions Permalink
  • Avatar
    TCW Economics
    Thank you so much for your thoughtful response and for your kind offer to provide continuing help. But before I get into that, I have an unrelated question for you: Are county-level data files still available for 2010? If so, how would we purchase them. (Your website only offers 2011 and 2012, and we're trying to be consistent with related analyses that use a 2010 database.) Concerning the development of my modeling approach, you're correct that what I'm trying to do is assess how households, industries, and institutions would be affected if the overall cost of water increases in my region. Without getting into too much detail, our project could temporarily affect water prices for all customers groups in a large urban region. For households, I think the assumption that increased water costs would lead to decreased discretionary spending on other goods and services is reasonable. So using an institutional spending pattern for a household income group makes sense. But for commercial and industrial water customers, the situation is more complex. As you noted, commercial and industrial customers could react in several ways, including 1)temporarily accepting reduced profits, 2)decreasing production, 3)raising prices, or 4)somehow changing their mix of inputs to reduce non-water-related costs. In reality, businesses in the region would likely take a combination of these actions. However, after reading your message, my sense is that trying to model any of these actions would be difficult. My thinking now is that the simplest approach may be to assume that most business owners would absorb reduced profits due to the temporary nature of the water-cost increase. This then would result in reduced personal discretionary spending, both by business owners and corporate shareholders (the ultimate recipients of corporate profits). If I treat the water-cost increase this way, then I could model it like I would for households, using an institutional spending pattern for households (presumably upper-end households). I realize this is an oversimplification of a complex situation, but do you think this is reasonable from a modeling perspective? For institutional water users, the situation, as you noted, is not straightforward. I don't really see a good way to model effects on these users other than to treat them similarly to businesses, by assuming that higher water costs will ultimately lead to lower discretionary consumer spending, either because of reduced program spending on other inputs, reduced wage payments, or reduced employment levels. So without a better option, I may model using an institutional spending pattern for households. Your thoughts on this simplified approach would be appreciated.
    0
    Comment actions Permalink
  • Avatar
    IMPLAN Support
    Hi Roger Thank you for your questions. You can still purchase 2010 data but you will need to call in and place the order, we apologize if this is an inconvenience. Please let our staff know if you have the same county data for the current year's data release as well. You can contact our sales staff at 651.439.4421. They are in the office from 8am Eastern time to 6.30pm Eastern time. The one consideration you might want to consider is the possibility of a threshold for Households for where it would affect their spending on other items, or if there is a different threshold for different income levels. You certainly can assume that all Households will respond in like manner and at the same threshold, but it would seem logical to us that certain Household income groups would be more affected and might be pressed to decrease their water usage, before higher income Households would notice. Thus you might want to run a couple variations of your edited spending pattern to be able to express a range of potential responses. Please also remember for all these, considerations to remove purchases of water from your decreased purchases. You might find the information at this link helpful as well: http://implan.com:80/v4/index.php?option=com_kunena&func=view&catid=84&id=13274&limit=6&limitstart=12&Itemid=35#13576 Your commercial response is also correct in that most businesses probably will respond with a mix of changes. The issue with reduced spending by private business owners and shareholders is determining whether or not these individuals are actually located within you study region, so one option here as well, unless you happen to have raw data that can point you to this information, this might be another situation where if you wanted to model this type of spending reduction you could do so in a range 50% local ownership, 25% local ownership 10% etc, or whatever seems logical to you as the analyst. Another option might be to try to find some information from regions that have experienced a change like the one you are describing to see if you can gain any insight from those who have already experienced it. One other potential solution we've seen is based on the assumption that Industries change how they produce their output. In this case you could: 1. Capture the impact of current Output 2. Edit production function to reflect changes, you feel would be made to the Intermediate Expenditures 3. Calculate impact of the new output using new production function. The impact is the difference in impacts between steps 1 and 3. However, since you are looking at the entire spread of Industries, this would be cumbersome and laborious. There is the possibility of rolling Sectors up to a more manageable number, but this comes with issues of Aggregation Bias, as well. If you were to attempt something like this you would want to use a different model that you used for the Institutional and Household spending models so you can keep the full level of Sectoral integrity for those Models. One other thought in regards to the Industry portion of the analysis, the Model will tell you what Industries and Institutions produce Commodity 3033 in your region at Explore> Social Accounts> Balance Sheet (Tab), and select View By: Commodity Balance Sheet and the Industry-Institutional Production and Industry Demand tabs. Here you can see if other Industries or Institutions provide these services in your region. If that is the case, there may also be a shift in Market Share that you could Model that would result from a price increase. Additionally, you could look at the Industry and Institutional Demand tab to try to identify Industries that have a larger Gross Absorption of the Commodity 3033, as the ones most likely to experience a change, and isolate those Industries as being the ones experiencing the greatest effects, even without doing the analytical side or anything past you described method or the method described above, only impacting the Industries you consider to be truly impacted, based on their Gross Absorption requirements. One note to this, the Gross Absorption sort can be somewhat deceptive, because of a bug we just found in the software, it sorts the values that have been adjusted with a negative exponent to the top of the list. So you will want to expand the Gross Absorption column, and scroll down to the items that don't include the exponents to find those Industries experiencing the greatest impacts. On Institutional spending here are a couple of additional thoughts. You could use these same sheets to identify Institutional Demand and see what Institutions are most impacted (Households as well as Governments). You might also want to affect the Institution Spending Patterns less the 437-440 Sectors since most government entities are reluctant to reduce payroll or Employment. You might also consider if the impact would have a greater effect on government capital expenditures than program expenditures, in which case you shift a bigger removal of funds to that spending pattern. We will run this by a couple of other people here, to be sure no one else has any further suggestions, and we'll let you know if anything else comes up. Please let us know if you have any additional questions. IMPLAN Support Team
    0
    Comment actions Permalink
  • Avatar
    IMPLAN Support
    Hi Roger, In getting back, the only other option we know of is to use an actual CGE Model. If this is something that interests you to pursue please let us know, and we can try to get you pointed in the right direction. --Implan Support Staff
    0
    Comment actions Permalink
  • Avatar
    TCW Economics
    Thank you for your suggestions and detailed response. I don't think I'm going to go in the direction of a CGE model, but I'll keep that in mind. I first want to test some of the options you suggested to see how they work and what kind of results I get. I may have other questions when I do this, but in the meantime, I appreciate the thought and effort put into your multiple responses to my questions.
    0
    Comment actions Permalink

Please sign in to leave a comment.