Pipeline Construction
I'm working on an analysis for a new pipeline construction that will run through rural Minnesota. My intention is to break the analysis into two parts: the impacts of the construction project itself and the impacts of the construction worker spending on retail and hospitality industries. My questions are as follows:
1. I understand from speaking with the company that much of the construction work is done by contractors from outside the study area. For example, the engineering, environmental, and survey crews were all contracted from Minneapolis. The same is likely true of the construction general contractor. Should that portion of the project still be considered a direct impact? I imagine some of the purchases and potentially some employees would be local, but the majority of employees would be from outside the area.
2. To show the construction worker spending (lodging, restaurants, etc...) separately, I plan to use the directions from this thread (link below) as a guide. However, to avoid double counting, should I also adjust the construction industry spending pattern (sector 58) to reduce or remove retail and hospitality coefficients?
http://implan.com/index.php?option=com_kunena&view=topic&catid=84&id=19079&Itemid=1840#19828
3. Is there a preferred way to conduct this type of analysis that I am overlooking?
I appreciate any help or feedback.
Thank you!
-
Hi Monica! As you noted construction on these types of projects can be very difficult to tease out. Certainly relative to your report, you would want to discuss the entire value of the pipeline project, but as for how you go about modeling it, you may want to work with a different approach. 1. You noted in your post that the region is very rural. So the impacts of the construction spending pattern may be small anyway, but you may want to consider, what if any of the inputs needed will be supplied local for a pipeline project in addition to the Labor Income. In addition these types of activities "engineering, environmental, and survey crews " are included in that spending pattern. You thus may want to use the Analysis-by-Parts methodology and consider removing some of the purchases from the spending pattern or if you want to use the total value of the construction project for Activity Level you could set the LPP field on the Events you think will be entirely leaked to zero through the Event Options> Edit Event Properties>Set Local Purchase Percentage> User LPC field. 2. Yes. While most of the construction workers will be from outside of the Study Area and they are bringing back home with them their Labor Income; they are still spending some of their wages in the region. It would be worth capturing that portion of their spending to view the impact. In a separate Activity using the same Model as your Construction impact; if you have a breakdown as to what the per diems are or what the Construction workers spent their wages (restaurants, hotel, general merchandise, etc) and the amount; you can analyze the impact of the spending. As regards removing them or zeroing out the LPP field that would depend on whether or not you felt that the construction companies in the region would not yield any other per diems as a result of the project. Typically these are very small impacts, so you should be with either assumption. If you want to error on the side of being conservative you could remove/zero these out. 3. The one other thing we might mention off-hand is to be careful about reporting jobs. The jobs associated to the project are supported not created in that they are short-term temp jobs, and in your case mostly non-local sustained jobs. One of the most common criticisms of these types of projects is that people will talk about the jobs they create as a result of pipeline construction, which even if accidental, is inaccurate in its description. Thus while the value of the project may be discussed it is important to emphasize the portion of it that actually effects the local economy. One other possibility, dependent on your audience, your data and your interests since you know some of your impacts are in Minneapolis, you could do a secondary impact to that region if you have the data an it is of interest to you. Or possibly a linked MRIO study. We'd be happy to help further if you have any additional questions. -
I have a follow-up question to my previous post on the pipeline construction, much of which will be conducted by non-local workers in rural northern Minnesota. First, I have adjusted the employee compensation using the method suggested in previous posts: newEC = EC*[(1-userCR)/(1-samCR)] to account for the 50% non-local workforce. I have not been able to get a per diem estimate from the company, but I still would like to model the impacts of the non-local workers' spending. I found a previous post on the forum in which someone noted that a general rule of thumb is that non-local construction workers spend 20% of their income locally, and the remaining 80% at their home location (http://www.implan.com/index.php?option=com_kunena&view=topic&catid=84&id=19079&Itemid=1840#19884). Is there some source for this information that you are aware of? If not, do you have a better method for estimating local spending? I would like to use the 20% method, but I feel uncomfortable doing so without better documentation. Thanks, Monica -
Hello, We do not know where he found that information and can neither refute it nor vouch for it. Here is a website that shows per diem rates (separated into lodging and Meals & Incidentals) for government workers. We don't know how similar these rates are to private construction companies' per diem rates but they may serve as a decent ballpark and are at least documentable: http://www.gsa.gov/portal/category/100120 We hope this helps. Best regards, IMPLAN Group Staff -
Thank you for the information on the per diem rates. As I have mentioned previously, this is a large construction project located in northern Minnesota. It will take place over the course of a two-year period, with spending equally distributed between the two years. I have the total budget for the project (minus land acquisition costs) and some minimal spending information, but the categories that the company provided are very broad, so I don't feel that I have enough information to do a full analysis-by-parts. Rather, I modified the industry spending pattern for sector 58 with the little information I had - adjusting a few coefficients and LPPs to match the information given to me by the company. The company did not provide employee compensation estimates, or per diem rates. To calculate the labor income change, I used the IMPLAN employee compensation value for sector 58. I then adjusted that value based on the share of the workforce that is estimated to come from outside the study area. The total labor income change amount ended up being about half of what it would have been were the entire workforce. I modeled the per diem rates separately using the GSA estimates you shared, with an industry change activity. My remaining questions: 1. Does my methodology seem ok to you, or is there something I am missing? 2. Because I have modeled all of the construction results separately for Year 1 and Year 2, I know the company will be curious as to the TOTAL economic impact of the project (both years). I know that I should NOT sum employment numbers for construction projects, but is it acceptable to sum the other values - value added, labor income, and output? Or, should I state in the report that the impacts can not be summed? -
As a follow-up, I did receive the following spending categories from the company. As you can see, they are very broad. Should I model them separately as industry change activities and adjust the LPP values for each? So far my method has been to simply adjust coefficients in the industry spending pattern but using the total construction budget as the level. The two methods give very different results. Expense (% sourced locally within study area) Site Preparation - Construction (65%) Site Preparation - Project Management (65%) Procurement (25%) Engineering (90%) Environment (NA%) -
After more trial and error, as well as more research on the forum, I have decided to model the procurement, engineering, and environment budget items each as a separate industry change activity (sectors 395, 449, and 455), adjusting the LPPs for each based on what the company has indicated they expect to purchase locally. Then, I am creating an industry spending pattern for the site prep-construction budget item, removing sectors 395/449/455 from the spending pattern and using the site prep-construction total as the activity level. Finally, I'm modeling a separate labor income change for the construction income (reduced by about half to reflect commuting rates). I'm sort of stumped by the site prep-project management item, however. Originally, I had planned on including it in with the construction spending pattern, but maybe it would make more sense in its own industry change activity? If so, should I use sector 454 - management consulting services or 461 - Management of Companies? Neither really seems appropriate for construction project management. Thanks! -
Hello, Response to post #20333: [The company did not provide employee compensation estimates, or per diem rates. To calculate the labor income change, I used the IMPLAN employee compensation value for sector 58. I then adjusted that value based on the share of the workforce that is estimated to come from outside the study area. The total labor income change amount ended up being about half of what it would have been were the entire workforce. I modeled the per diem rates separately using the GSA estimates you shared, with an industry change activity.] - If you did a raw adjustment on the Labor Income value and did not consider the SAM payments to in-commuter, you may be slightly conservative in the local Labor Income results, assuming the region has a net-incommuting rate of workers (they live outside the study area) based on the information you have provided. 1. Your methodology is fine. With Analysis By Parts, you don’t always need detailed inputs. ABP is used for when an Industry doesn’t exist in the model region, or you want the more control over how the Industry spends. 2. The easiest way to report the jobs in construction across multiple years is to a is to sum the jobs per year and divide by the number of years, in this case two, to get the average jobs per year on the project, then add that to the total employment. \ - It's important to note in the report that these jobs are supported and short-term (unless this is an unusual pipeline that exist only for the period of construction and is not accompanied by operational jobs at the completion of the project). It may also be useful to try to report a local/non-local job count. The rest of the values are able to be summed, but it's important they are summed and expressed in a single year's values (you cannot sum 2016 Dollar Year for View with 2015 Dollar Year for View values). Typically the current year is the standard for reporting because those dollars are considered to be implicitly understood. You can use the DYFV to obtain the current year’s deflation. Response to post #20337 For the additional expenses, you can choose to edit your spending pattern values, or model them separately. You might want to ask the client what they prefer. -
Thanks, that is very helpful. I did adjust the labor income using the SAM payments to in-commuters. The rate is small, though, so the final result is still about half - 51%. As for jobs, you say to "sum the jobs per year and divide by the number of years, in this case two, to get the average jobs per year on the project, then add that to the total employment". What do you mean by "add that to the total employment?" Wouldn't the average per year BE the total employment? For the local/non-local job count, should I use the same equation that I used to calculate the SAM payments to in-commuters, or should I simply divide by half? According to the company, 50% of the jobs will be filled by local workers, but they are probably not considering the existing commuting rate. I assume all of the indirect and induced jobs are local, so the modification would only be for the direct jobs, right? My other question about the site-prep project management expenses still remains, as well. Thanks - this is a very controversial project, so I want to make sure I'm performing the analysis as accurately as I can. -
When presenting the findings, I have summed the construction impacts with the per diem spending to get the total impacts for the project. However, because I modeled the per diem spending as industry change, the results show up as direct impacts. They seem like they should be reported as induced impacts, as they represent employee spending. Should I sum the effects and add them to the indirect column? Or report them as direct? -
Hello, We have parsed your questions out to better organize and answer them. We apologize for the delay. [quote]As for jobs, you say to "sum the jobs per year and divide by the number of years, in this case two, to get the average jobs per year on the project, then add that to the total employment". What do you mean by "add that to the total employment?" Wouldn't the average per year BE the total employment?[/quote] - This was a mistake, you are correct. The Employment should be reported as jobs/year on the project. [quote]For the local/non-local job count, should I use the same equation that I used to calculate the SAM payments to in-commuters, or should I simply divide by half? According to the company, 50% of the jobs will be filled by local workers, but they are probably not considering the existing commuting rate. I assume all of the indirect and induced jobs are local, so the modification would only be for the direct jobs, right?[/quote] - Great catch, but that really only applies to the Labor Income because we are looking adjust the expenditures for what the software is already removing. The software will not make any changes to the Employment count, and this note was simply for reporting purposes. All jobs are defined as local in IMPLAN because they are considered onsite, but in terms of expressing this, people will likely appreciate knowing that the Employment split is estimated to be 50:50. To your point however, it sounds like your calculation for the net commuting rate was ~1%. If this is correct then the entered Labor Income value should be 51% of the expected or 49% less. (Just to confirm.) [quote]My other question about the site-prep project management expenses still remains, as well. - "I'm sort of stumped by the site prep-project management item, however. Originally, I had planned on including it in with the construction spending pattern, but maybe it would make more sense in its own industry change activity? If so, should I use sector 454 - management consulting services or 461 - Management of Companies? Neither really seems appropriate for construction project management."[/quote] - Without knowing more about this, it’s extremely hard to determine what Sector to run this through if you don't just run it through the construction spending pattern. Would there be an emphasis on architectural engineering at the site prep phase, or landscaping, or is this purely administrative? If you have a NAICS code or company name for this that could help to determine a Sector as well. You are also correct that these should be run with LPP, unless it's know that all these Activities are performed by local service providers. [quote]When presenting the findings, I have summed the construction impacts with the per diem spending to get the total impacts for the project. However, because I modeled the per diem spending as industry change, the results show up as direct impacts. They seem like they should be reported as induced impacts, as they represent employee spending. Should I sum the effects and add them to the indirect column? Or report them as direct?[/quote] - We typically would recommend reporting them separately. In addition to the project and local labor external workers are estimated to spend X resulting in Y impacts. If you do need to sum them then you would want to do it by using the Total Direct Labor Income local and non-local back into the Labor Income value and then adding the Indirect and Induced Effects together into the Induced Effects of the results. -
I have completed the project and submitted the results to the company, but they brought up a question that I was not sure of. Of course, I excluded the land acquisition costs from my analysis, but they actually indicated that the land acquisition they are referring to is in the form of easements - where they are paying households to use their land, but not actually purchasing the land from the individuals. In addition, they pay those households an additional 10% premium on the land value. Knowing this, should those land acquisition costs be included in the analysis as an increase in household income? Thanks, Monica -
Hi Monica, Here are some additional considerations: [ul] [li] Make sure to account for displaced activity. For example, if the easement is displacing an alternative use of the land, e.g. farming, there may be a loss of farming production impact, but a gain in rental/easement income. For another example, the payer of the easement might already be in the study area, and could redirect spending on something else to spending on the easement, in which case you would want to consider the net effect. [/li] [li] Note that landowners may not live in the study area. If that’s the case, there would be no impact. In general, ask whether the transfer of money is from outside the region to someone inside the region (positive impact) or vice versa (negative impact). [/li] [li] Consider who the payer of the easement is. Government transfer from inside the region to a property inside the region means that money is transferred from many taxpayers' pockets to a single property owner. A corporate transfer from within the region to a local individual would have an impact, as I-O treats corporate profits (an easement payment to a household would be paid from Other Property Income) as a leakage (so now there would be less leakage).[/li] [li] If you assume the recipient will spend the money rather than re-invest, you will probably want to model this as a Household Income Change activity type, as this will allocate some of the funds to personal taxes and savings. If you instead model this using an imported household spending pattern, you are making the assumption that every cent of the funds will be spent, with Local Purchase Percentage initially set to SAM value. - Note that the spending of easement income may differ from standard household income spending patterns. Households might treat it as windfall income and spend it on luxury goods, or save it, etc. - - This can be corrected by using the SAM to determine the payments to savings, taxes, etc to reduce the income value, and then the Household Spending Pattern can be modified to reflect items that would be purchased with expendable income.[/li] [li] If the recipient of the easement payment incurs significant legal or banking costs related to granting the easement, those costs should be modeled in the appropriate sectors. Likewise, if your estimate of the easement value includes legal and/or banking fees plus the actual amount transferred to the recipient, you should model those components in the appropriate sectors.[/li] [/ul] Best regards, IMPLAN Group Staff -
If we have the Employee compensation for the period of 15 months and total number of estimated employees. Assuming 50% of the employees are local and rest from outside. What sectors are we considering to run the analysis. If we use the construction sector are we over stating/estimating because pipeline construction involves more of highly skilled workers at a very high salaries when compared normal construction worker? We have the operating and maintenance budget so will be using sector 413 Pipeline transportation? And also they have provided the property taxes how do we consider them? Thanks, -
Hello Sunny, Thank you for your forum post. For measuring the impact of the construction of the pipeline, we recommend using Sector 58: Construction of other new nonresidential structures- which includes: reservoirs, pump stations, and water pipeline construction and pipeline construction other than sewer and water. We certainly encourage you to customize this Sector based on the information that has been provided to you. In your first paragraph you mention having Employee Compensation (EC) and an Employee estimate. If that is related to the construction of the pipeline; you can over -ride IMPLAN's estimates. We recommend entering your Employee Compensation value first and then editing the Employment number. This bases the Output estimate on your provided Employee Compensation value; EC tends to be a better predictor of Output since Employment counts can include part-time and seasonal workers in addition to full-time workers. Labor Income is spent at the point of residence. If you know that 50% of your construction employees live outside the region, you will need to account for this by altering your Employee Compensation value. You can find information about this process at the link below. Note that you will want to use this method after you've used your full Employee Compensation value to estimate Output (discussed in the paragraph above). http://support.implan.com/index.php?option=com_content&view=article&id=254:254&catid=230:230 Also, if the construction firm is utilizing only employees (i.e., no contractors), you will want to zero out Proprietor Income. For measuring the impact of the operations of the pipeline, sector 413 is the correct choice. However, if your budget only includes the cost of intermediate expenditures for the operations of the pipeline, you will need to estimate Output using the Study Area Data ratio of Output to IE. You can find this by navigating to Social Accounts > Balance Sheets > View By: Industry Balance Sheet > Select your sector > Commodity Demand Tab. The Total Commodity Demand Gross Absorption percentage is the percent of Total Output that goes to Intermediate Expenditures. Finally, taxes are not spent through the IMPLAN Model but rather set aside and reported in the Direct Tax Results tab. You can export the tax results to Excel and over-ride the tax information with the information you were provided. You will need to shift the other types of Taxes on Production and Imports (TOPI) to compensate for your changes. For example, if your Property Tax value is higher than IMPLAN's, you will need to reduce the values of the other TOPI types (Sales, Motor Vehicle Licenses, etc.) accordingly so that the sum of them all still equals the total Direct TOPI. Regards, IMPLAN Staff
Please sign in to leave a comment.
Comments
14 comments