Introduction
Most base closing activity results in shifting of jobs from one region to another region. A base closing results in a loss of employment from one region, while other surviving bases gain employment.
Five case studies are provided, each building on the information and instruction that occurred in the case before it. Therefore, a sequential work-through will provide the most thorough understanding of the various aspects of the potential impacts of base closures.
Detailed Information:
Each Case Study below relies on a unique set of assumptions that drives how we view the impact.
-
-
Case Study 1
- Only the value of Employment change is known.
- All military operational/investment purchases (supplies) are made through Government Services Agency (GSA) or equivalent type of purchasing agency.
- Federal procurement is disconnected to any base activity, thus we do not assume that if your region produces tanks or paperclips for the military that those jobs leave as well.
- With this assumption, we are limiting ourselves to only payroll impacts.
-
Since there is an assumption that we will have no local Intermediate Expenditure impact for this case study, we can use the specialty Sectors in the IMPLAN Sectoring scheme (536 scheme: 535, 536; 440 scheme: 439, 440) that are specific to payroll and Employment only. The key take aways from this example are:
-
-
- The government payroll Sectors do not generate any Indirect Effects because these special Sectors have no associated spending patterns.
- For these unique Sectors, Value Added = Output.
- If non base Employment is included in the Employment values we need to remove these jobs to avoid double-counting.
- Federal reports can be used to make splits if the split between military and civilian Employment is unknown.
-
Case Study 2
- All military operational/investment purchases (supplies) are made through GSA or the equivalent type of purchasing agency except for known local contracts. Thus there will be no Indirect Effects outside of those modeled as contract losses.
- Information on base closures is only available in terms of Employment.
- Only base Employment is provided. Thus we don't have any jobs included that could double-counting jobs and impacts in the Induced Effects.
- The known local contracts are part of base operations and purchases and; therefore, will not be double counted by payroll.
-
The key take aways from this example are:
-
-
- Indirect Effects that are reported when the specialty payroll Sectors and contractors are combined reflect the purchasing rounds of these private Sector contracts ONLY. Other military base expenditures are not included in the Indirect Effects, because the payroll Sectors have not associated spending pattern as discussed in Case Study 1.
- When a value of a contract is provided the Event Year field must match the value of the contract year.
-
Case Study 3
- Both Employment and payroll information are provided
- All military operational/investment purchases (supplies) are made through GSA or equivalent type of purchasing agency
-
- Employee count includes only base employees
- Payroll is equivalent to Employee Compensation, so wage and salary values will need to be converted before they are entered into the Model.
-
The key take aways from this example are:
-
-
- Payroll estimates can easily be adjusted in the Event to match known values of military Labor Income
-
Employee Compensation includes all wages and salaries plus benefits (monetary and non-monetary) plus employer paid taxes, thus while the Federal government may be considered to pay less based on worker skills, the benefits packages are usually excellent.
-
Many times reported earnings data are simply wages and salaries, excluding benefits. For impact analysis we have to convert wage and salary data to Employee Compensation.
- Payroll values are entered first, then Employment, this provides the best estimate of Industry Sales.
- The Employee Compensation values are entered in that field, not Industry Sales.
-
Case Study 4
-
All military operational/investment purchases (supplies) are made through GSA or equivalent type of purchasing agency.
-
Employment values include only Military or Civilian base Employment.
-
-
The key take aways from this example are:
-
-
- BEA Data can be used to create a split of Employment for proxying the impact of base closures when only a total budget is provided.
-
Case Study 5
- Our military base acts like a local Industry making purchases of goods and services from the local economy to meet its operational needs.
- Employment counts include only base employees.
- If construction and equipment purchases also occur locally rather than on a GSA schedule, we will need separate spending patterns to consider these additional impacts.
- The local military purchases will occur at the same rate of local availability as all other commodity purchases in the Model.
-
The key take aways from this example are:
-
-
- We can also use regional data from our Model to make estimations for our analysis along with the federal information. This allows us some additional regional specification.
- Because of definitional differences in Institutions, results are easier to interpret when payroll is analyzed separately from local goods and services purchases.
- Employee Compensation is good proxy for Employment and Sales/Production values when these are unknown.
- Depending on the type of government activity involved there are multiple spending pattern options.
- It is important to consider the contractor when considering construction impacts in your analysis.
- Equipment and other none-structural capex purchases should use the LPP set to SAM Model Value to avoid overestimation of local impacts.
-
Usage:
Case 1: Only Employment Information is Known
When only Employment is known there are three (3) potential ways that we could look at the Employment change:
- Total Employment loss is known and there is an available split for military and civilian jobs.
- The loss if some civilian employees work in on-base services.
- Only total Employment is known, we have no information about the breakout of military or civilian jobs.
When a Breakout of Military and Civilian Jobs is Provided
We are provided with the following information for our impact:
-
- Federal Military: -2,000 jobs
- Federal non-military: -1,500 jobs
In IMPLAN, we'll create an Activity for the Military Base Closure and populate it with two Events one of each of our two Employment impacts.
We can then create a Scenario for our Activity, and view our results.
Considerations of the Results
Why does Output = Value Added?
Sectors 535 and 536 (439 and 440 in the 440 scheme) are unique in that their Output equation typically Output = Intermediate Expenditures + Value Added is simplified by the removal of Intermediate Expenditures, thus Output = Value Added. In addition to this, the typical Value Added equation is streamlined because governments do not pay Proprietor Income or Taxes on Production & Imports. Thus the Direct Effect is Value Added = Employee Compensation + Other Property Income.
What is Other Property Type Income for Government?
Other Property Type Income represents the Capital Consumption Allowance since profits are not part of administrative government activity. Other Property Type Income is treated as a leakage and has no impacts, so while it is part of the total Output and Value Added factors it does not drive any Induced Effects.
Why is there no Indirect Effect?
As noted in the Detailed information, Sectors 535 and 536 (439 and 440 in the 440 scheme) have no production function, thus there will be no Intermediate Expenditures resulting from the use of these Sectors. This is fine for our example, because in this case study we are assuming that government expenditures are GSA driven and thus independent of the base's activities. Therefore, our results are specifying jobs, which generate Induced Effects (payroll), but there are no other operational (Indirect) effects to this impact.
Removing Employment that Results from Base Activity but Are Non-federal Employees
Many military bases may have non-military Employment that is indirectly attached to the base. These are jobs at businesses that reside on or near the base in order to provide goods and services to base employees such as grocery stores, barbershops and other retail and service needs. We want to ensure that the base Employment information provided excludes employees providing these services. Why? Because these retail and service purchases are already accounted for in the Induced Effects. Thus if these employees are included in our analysis, we will double count their jobs and expenditures.
Thus ideally, we will either have a number exclusive of retail and service Employment on the base or an estimate by which we can reduce our total Employment count.
For this case we are told that 100 civilian jobs in our original count are from these services.
Federal Military Lost: 2,000 jobs
Federal non-military (civilian) Lost: 1,400 jobs (1,500-100)
We can then create a Scenario for our Activity, and view our results.
Considerations of Results
Because of definitional differences, it can be best to avoid using the terms Direct, Indirect, Induced if these will not follow your audiences expectations. For example, if most people knew the loss of civilian jobs to be 1,500 you can reframe your report to match expectations and avoid confusion without fundamentally altering the analysis. Thus, depending on your audience, it may be helpful to include the 100 jobs that are reported in IMPLAN as part of the Induced Effects in the Direct by removing those jobs from the Induced row and adding them to the Direct Effect making clear that those 100 jobs were supported by the military and civilian Employment. The remaining Direct value would not be altered because this would change the actual base figures and the Labor Income, Value Added and Output associated to these retail and service workers properly belongs in the Induced Effects.
If a Single Number Representing both Civil and Military Employment is Provided
If a breakdown isn't available, it is still best to try to determine a breakdown between military and civilian Employment. The simplest solution to this scenario is to use the study area data employment values for the two different federal government payroll sectors.
With a given loss of 3,500 jobs, the breakout would be as shown below:
| U.S. 2009 | Employment Value | Ratio | Base Employment Loss |
| Military | 1,088,465 | 0.65466 | 2291 |
| Civilian | 709,265 | 0.39434 | 1209 |
| Total | 1,797,730 | 1.00 | 3,500 |
Case 2: Employment plus Local Contracting
What if we have knowledge of local contracting by the base to perform services or provide goods related to the operations of that base?
Since more local information about local contracting done by the base, we can now look at additional impacts that will occur as a result of the base closure. There are three potential ways this can be addressed in IMPLAN.
- If only contract jobs are known.
- If contracts are provided in terms of annual value rather than in terms of jobs.
- When both jobs and contract value are provided.
When Military, Civilian and Contracted Jobs Are Known
Loss of base jobs:
-
- Federal Military: -2,000 jobs
- Federal non-military: -1,500 jobs
In addition to our military and civilian jobs, we now have contract losses of:
-
- 40 jobs in Military armored vehicle, tank, and tank component manufacturing; Sector 366 (293 in the 440 scheme)
- 100 jobs in Travel Arrangement and Reservation Services; Sector 466 (383 in the 440 scheme)
- 50 jobs in Fluid Milk Manufacturing – sector 84. (55 in the 440 scheme)
We can then create a Scenario for our Activity, and view our results.
We do have Indirect Effects in this set of results, but these are due to inclusion of the private Sector activities and do not reflect any purchasing of goods and services beyond those contracts specified in our Direct Effect. Thus they will appear smaller than would be expected if other base purchases were included.
Just as in previous examples, we would not want to double-count any jobs that might be attributed to employee purchases (see Case Study 1 Example 2).
Value Added is somewhat smaller than Output because of the addition of private Sector impacts.
When Military, Civilian Jobs Are Known but Contracts Are by Value
One key thing to remember is that the value of the contract is based on the dollar year of the award, thus Event Year should match the contract dollar year.
In this example our military and civilian jobs are unchanged but we now have the following contract values:
-
- $10,000,000 in Military armored vehicle, tank, and tank component manufacturing; Sector 366 (293 in the 440 scheme)
- $15,000,000 in Travel Arrangement and Reservation Services; Sector 466 (383 in the 440 scheme)
- $5,000,000 in Fluid Milk Manufacturing – sector 84. (55 in the 440 scheme)
We can then create a Scenario for our Activity, and view our results.
In comparing the results we can see that our starting value can have significant impact on our results, so ideally we might try to get both of these values.
Applying Both Employment and Value for Contracts
In our final Scenario, we have both the value of the contract and Employment, in addition to our military Employment loss.
-
- $10,000,000 in Military armored vehicle, tank, and tank component manufacturing; Sector 366 (293 in the 440 scheme)
- $15,000,000 in Travel Arrangement and Reservation Services; Sector 466 (383 in the 440 scheme)
- $5,000,000 in Fluid Milk Manufacturing – sector 84. (55 in the 440 scheme)
Note that because we modified the underlying relationship between Output and Employment all the private Sector fields show a customized value warning.
We can then create a Scenario for our Activity, and view our results.
Case 3: Employment and Payroll Information Are Both Available
When Employment and Payroll Are Known
When we have both the base Employment and Payroll information we can provide the most accurate estimate of the impacts of base closure, assuming that contracts are GSA based and thus no local operational impacts occur.
For the base analysis we are given the following information:
-
- Military job loss: 2,725
- Military payroll loss: 129,543,775
- Civilian job loss: 775
- Civilian payroll loss: 64,989,175
Our Employment Compensation value should be entered first, then Employment is edited.
We can then create a Scenario for our Activity, and view our results.
If Wage and Salary Rather than Payroll is Provided
We are told that the payroll data represents wage and salary data only, so we need to convert this data to Employment Compensation. Using the convenient national average compensation to wage and salary ratios (Table 3-5) we can make this conversion:
The following information is provided:
-
- Military job loss: 2,725
- Military wage and salary loss: $85,000,000
- Civilian jobs loss: 775
- Civilian wage and salary loss: $40,000,000
IMPLAN provides a conversion chart for modifying wage and salary values to Employment Compensation. These ratios are at a national level, so if you have local ratios available you may want to use these instead. The IMPLAN ratios for converting wage and salary to Employment Compensation are as follows for the 536 Sectoring scheme:
-
- Sector 535: 1.424881649
- Sector 536: 1.42985376
Since we know Employment Compensation is always the larger value, we will multiply these conversions by our wage and salary values to determine our new Employment Compensation.
| Sector | Jobs | Wage and Salary Value | Ratio | Employment Compensation Value |
| Military | 2,725 | $85,000,000 | 1.42985376 | $121,537,569.60 |
| Civilian | 775 | $40,000,000 | 1.424881649 | $56,995,265.94 |
Our final consideration would be to confirm that our payroll values are in 2015 dollars if that will be our Event Year. Event Year needs to match the value of the payroll dollars.
We are now ready to analyze the Scenario.
Case 4: Only Budget is Provided
When Budget but No Employment or Payroll is Provided
We are only told that the budget of the base being closed has an operational cost of $100,000,000. Since we are assuming that all Intermediate Expenditures are sourced through GSA and have no local impact, we will need to find a way to convert this budgetary value to Employment. we will use national relationships to convert the $100,000,000 to military and civilian employment.
2015 Quarter I Values ($billions )
-
- Gross Output of Defense: $638.3
- Employee Compensation $237.0
- Goods and services $233.8
- Investment $141.5
- Structures $4.6
- Equipment & Intellectual Property Products $136.8
- Gross Output of Defense: $638.3
It is important that we make this split because of expenditures, because the Employment and payroll portion can be derived only from the portion the non-investment portion of operations. Thus we will need to develop ratios to help us see create a breakdown for estimating Employment Compensation and Employment.
| Total Expenditures | Ratio of Total Expenditures | Ratio of Subcategory | |
| Gross Output of Defense | $638.3 | 0.818543 | 100% |
| Value Added | $404.5 | 0.518723 | 63% |
| Employment Compensation* |
$237.0 | 0.303924 | 58.5909% |
| Goods and services | $233.8 | 0.29982 | 37% |
| Investment | $141.5 | 0.181457 | 100% |
| Structures | $4.6 | 0.005899 | 3.2509% |
| Equipment, Software & Intellectual Property Products | $136.8 | 0.17543 | 96.6784% |
-
- * Employment Compensation is a subset of Value Added representing 58.6% of the Total Value Added
- Output = Intermediate Expenditures (Goods & Services) + Value Added
- The sum of Investment and Gross Output are used to derive the base value for creating the Ratio of Total Expenditures.
The next step is to remove the portion of $100,000,000 that corresponds to Investment. Since Investment represents 18.1457% we'll reduce our $100 million by that amount. Our adjusted military operating budget is $81,854,322.
Now we can combine the study area split of military and civilian jobs with our operational budgets to complete our splits.
| Values | Budget per Job | Number of Jobs per Budget | |
| Operational Budget | $638.3 | (Budget/Budget per Job) | |
| Total DOD Jobs | 1797730 | $355,059 | |
| Military Jobs | 1,088,465 | $586,422 | 140 |
| Civilian Jobs | 709,265 | $899,946 | 91 |
We can now create our Activity and Events based on our estimated jobs.
Case 5: A Military Base with Local Purchases
What if the Military Base Operations Function like a Normal Industry?
Payroll Component
Typically, because of the prevalence of GSA scheduling, we assume that base closures do not affect the local economy, but in some instances you may want to show impact of the base as if it were a normal 'Industry' in the economy.
We are given a 2015 base budget of $100,000,000. We can use the ratios developed in the previous example expanding out the Investment aspects of the data:
| Total Expenditures | Ratio of Total Expenditures | Ratio of Subcategory | Value of Our Base Operations | |
| Gross Output of Defense | $638.3 | 0.818543 | 100% | $81,854,322 |
| Value Added | $404.5 | 0.518723 | 63% | $51,872,275 |
| Employment Compensation* |
$237.0 | 0.303924 | 58.5909% | $30,392,408 |
| Goods and services | $233.8 | 0.29982 | 37% | $29,982,047 |
| Investment | $141.5 | 0.181457 | 100% | $18,145,678 |
| Structures | $4.6 | 0.005899 | 3.2509% | $589,895 |
| Equipment | $70.9 | 0.090921 | 50.1060% | $9,092,075 |
| Intellectual Property Products | $65.9 | 0.084509 | 46.5724% | $8,450,885 |
Based on these values we can determine how much of our $100,000,000 is used for operations of the base verses the amount going to construction and investment. The operation value can be determined by summing the ratios of our budget for Goods and Services (i.e. Intermediate Expenditures) and Value Added. Thus of our total budget can be derived as ($100,000,000/(0.518723 + 0.29982)) = $81,854,300
Examining the Operational Component
Now that we know what the total budget is we can import a Institution Spending Pattern for the military base to show us the impact of purchases of both goods and services and Employment and Employment Compensation based on a regional breakout.
To do this we need to start by importing the Institution Spending Pattern for Military Defense,
Step 1: Setup the Import
Step 2: Select the Federal Government Defense Spending Pattern and Import.
Step 3: Select the Institution Spending Pattern for Federal Government Defense.
Note that there are already local splits of Employment and Employment Compensation in the spending pattern for this Sector. If we have local data we may want to modify these, but if not these are a good local proxy, since they are based on locally reported military activity.
At this point we have two options:
- Modify the spending pattern to include only the goods and services purchases.
- Leave the spending pattern as is and model our Activity.
The advantage of leaving it as is (option 2), is that we have no other work prior to analyzing the results that we need to do.
The advantage of modify the current spending pattern is that the results of the analysis are clearer and easier to interpret.
Either way we should obtain the same final results.
Option 1
The included coefficients are the coefficients of total spending on our two Events payments to Commodity 3535 and 3536, thus we can multiply these directly against our budget to determine Value Added needed for running these Events in a different Activity and Scenario.
Thus our Value Added for these Sectors is as follows:
-
- 535= $11,859,542
- 536= $40,954,980
Since Value Added is equal to Output (since these special Sectors do not include any Intermediate Expenditures) and Output is equal to Industry Sales:
We will create a separate Scenario for these and determine our results.
Once we confirm that Sectors 3535 and 3536 are removed from the spending pattern, we can edit the Activity Level of the spending pattern to show the full budget value.
Because the payroll component has been removed, note that the spending pattern no longer sums to 1.00. This means that of the $81,854,300 entered as Activity Level only 35% of the value entered as Activity Level will be spent. This makes sense because the rest was spent in our payroll Activity.
We can now edit the Activity to set the value of the spending pattern Activity to $81,854,300. This window is opened through the Edit Activity button at the top of the Activity page or by right-clicking on the Activity to be edited.
We can now build a Scenario and view our results.
Let's note a few things about our results:
- Because of definitional difference between Institutions and Industries, the Direct Effect shown here actually represents the first-round of Institutional spending, not the military base's payroll Value Added, Output and Employment.
- If we wish to format this like Industry results we will need to sum the Direct Effect row into the Indirect Effect row and then add the actual miltary base values to the Direct row and resume the Totals.
- This is where having the separated Event comes in handy, our Direct Effects are basically defined for us by the Value Added impact.
- Output equals our total starting budget so our Direct Output equals our Activity Level, $81,854,300
- Value Added is provided by the 536 and 537 Military Employment Activity. So our Direct Value Added = $52,814,522
- Our military Employment and payroll are also defined by the Military Employment Activity: Employment = 309 and Employment Compensation = $29,191,124
- Our results then sum to:
In addition two new fields are populated as highlighted in yellow. In the context of the Federal Government Defense Spending Pattern:
-
- Direct Institution Change: This field describes the value of production that came from Institutional production in the Model and thus was leaked from the first-round Multiplier spending.
- LPP Imports: This shows the value of the spending pattern that was leaked from the first-round impacts as a results of regional availability to meet demand.
Option 2
Since we are not making any adjustments to the spending pattern with this option we can simply Edit the Activity level to our $81,854,300.
And create our Scenario and Results
Note that the values are higher than in the last set of Summary Results but the values of Direct Institution Change and LPP Imports have not changed. This indicates that the difference lies in the fact that the first-round Indirect expenditures in our spending pattern are now mixed with the actual base Employment, Labor Income and Value Added figures. We can demonstrate this by viewing our Detail Results tab
View By: Output. These results are sorted by Direct Effect.
In order to format the results as we did above we need to:
- Subtract the impacts of 536 and 535 for Employment, Labor Income, Value Added and Output.
- Follow the steps provided in Option 1 to sum the separated rounds of Indirect Effects (step 2).
- When we finish all the formatting our results will be identical:
Since IMPLAN is a linear Model the closure of a base or the opening of base of the same budget will yield the same results with opposite signs. So if this was a base closure, the results would be identical but negative.
Construction Components
Since we have a percentage of our total budget that is for structures, we can use the ratio developed in our first chart to estimate the impacts of construction on the base.
Equipment and Other Capex
We can use the remaining ratios to capture the potential impacts of Equipment and R&D purchases.
Comments
0 comments
Article is closed for comments.