B-234488.2, Aug 7, 1989, 89-2 CPD 109

B-234488.2, Aug 7, 1989, 89-2 CPD 109

PROCUREMENT - Special Procurement Methods/Categories - In-house performance - Cost estimates - Cost evaluation - Fringe benefits DIGEST: 1. In a cost comparison conducted pursuant to Office of Management and Budget Circular No. A-76, an agency may use a mock reduction in force procedure to determine the propriety and amount of certain one time labor conversion costs, that is, severance pay, relocation costs and retraining costs, to be added to contractor's price. PROCUREMENT - Special Procurement Methods/Categories - In-house performance - Cost evaluation - Personnel 2. In a cost comparison conducted pursuant to Office of Management and Budget Circular No. A-76, an agency may include, as a one-time conversion cost to be added to a contractor's price, the reasonable costs of supplementary employees necessary to conduct a reduction in force in the event a determination is made to convert an in-house function to a contract, provided the agency's need for such personnel is established. PROCUREMENT - Special Procurement Methods/Categories - In-house performance - Cost estimates - Contract administration - Personnel 3. In a cost comparison conducted pursuant to Office of Management and Budget Circular No. A-76, an agency properly obtained a waiver permitting it to include the costs of six contract administration personnel, rather than the four assumed by the Circular, where the agency found that it needed the additional contract administration personnel in view of the technically specialized disciplines involved in the contract and the function's base-wide coverage as established by an operational audit conducted to support the waiver request. PROCUREMENT - Special Procurement Methods/Categories - In-house performance - Cost estimates - GAO review 4. In a cost comparison conducted pursuant to Office of Management and Budget Circular No. A-76, an agency properly included costs in its estimate to accomplish indefinite quantity work required by the statement of work. PROCUREMENT - Special Procurement Methods/Categories - In-house performance - Cost estimates - GAO review 5. In a cost comparison conducted pursuant to Office of Management and Budget Circular No. A-76, an agency estimate of overtime hours to be included in its cost estimate will be upheld where the overtime estimate is reasonable on its face.

Intelcom Support Services, Inc.:

Intelcom Support Services, Inc., protests a determination made by the Air Force pursuant to an Office of Management and Budget (OMB) Circular No. A- 76 cost comparison that it would be more advantageous to the government to maintain in-house performance of base operation services, rather than to contract for these services, at the Air Force Reserve Base, Minneapolis- St. Paul, Minnesota. Intelcom alleges that the Air Force improperly conducted the cost comparison that justified this determination under request for proposals No. F21611-87-R-0004, issued by the Air Force Reserve Headquarters, Robins Air Force Base (AFB), Georgia.

We deny the protest.

The RFP solicited fixed price offers to provide a variety of base operating services for 1 year with two yearly options. These services, encompassing line items 0001 through 0006 of the basic contract period, include the following functions: (1) supply; (2) motor vehicle management and maintenance; (3) traffic management; (4) administrative services; and (5) real property maintenance. Additionally, there was provision in the RFP for indefinite quantity work (IQW) including (1) maintenance, repair and minor construction/alteration (line item 0007), (2) training (line item 0008), and (3) other miscellaneous services, e.g., carpet repair, snow removal, asbestos removal, etc. (line item 0010).

Intelcom submitted the low acceptable proposal of $6,165,226.79. The government estimate was $8,055,416, and was based upon a most efficient organization (MEO) of 78 persons, a significant reduction from the 113 government employees who were performing the function at the time of the study.

The Air Force completed the cost comparison pursuant to OMB Circular A-76 Supplement Part IV-- Cost Comparison Handbook (August 1983) (Handbook) and Air Force Regulation (AFR) 26-1 volume 1 (Aug. 28, 1987). Under the cost comparison, the cost of federal income tax was deducted and the costs of contract administration, one-time conversion costs (including severance pay and relocation and retraining costs of government employees incurred if a contractor were selected that was calculated with the assistance of a mock reduction-in-force (RIF), as well as associated personnel costs and inventory costs), and a conversion differential equal to 10 percent of the government's in house personnel costs, were added to Intelcom's price, raising it to $8,124,179.98. Since this figure exceeded the total in- house costs of $8,055,416 by $68,763.98, the Air Force decided not to contract for the services.

Intelcom timely appealed this decision to the Robins AFB base cost comparison administrative appeal review team (base appeal), which found a number of mistakes involving wage escalation in the government estimate, which caused the estimate to increase by $3,845.60. Because that left the in-house estimate low by $64,918.38, the base appeal team affirmed the decision to retain the services in-house.

Intelcom then requested and received a second tier major command review of the cost comparison as contemplated by the Air Force procedures. See Dyncorp, B-233727.2, June 9, 1989, 89-1 CPD Para. ***; Intelcom Support Servs., Inc., B-234488, Feb. 17, 1989, 89-1 CPD review board (second tier appeal) found no additional errors and sustained the decision to retain the services in-house.

In reviewing an A-76 cost comparison, our decision turns on whether the agency complied with applicable procedures in selecting in-house performance over contracting and whether the comparison was faulty or misleading. Dyncorp, B-233727.2, supra. Intelcom alleges numerous errors in the cost comparison and asserts that the Air Force received unsupported and unreasonable waivers of ordinary OMB Circular A-76 cost comparison procedures. Our review, however, indicates that the Air Force committed no errors in the cost comparison that affect the decision to retain the services in-house.

A primary focus of Intelcom's protest is the one-time conversion costs added to Intelcom's price based upon the mock RIF of those incumbent government employees who the Air Force believes will be displaced if a contractor is selected. Intelcom asserts that the severance pay is $287,278 too high, and that the other costs associated with the mock RIF, that is, relocation costs ($99,677), retraining costs ($73,388), and the costs of supplementary personnel to process the RIF ($37,378) are not proper one-time conversion costs.

Although Intelcom has made detailed arguments disputing each of these elements of cost, Intelcom's basic argument is that the mock RIF was flawed since it concluded that only 10 government employees would accept comparable employment with the contractor. Intelcom claims that its historical data shows that, at a minimum, 85 percent of incumbent government employees have remained on the project when Intelcom assumed a former government operation.

The Handbook, Chapter 3, para. E.3.a and AFR 26-1, vol. 1, para. 9 15.c.(1), specifically recognize that severance pay, relocation costs, and retraining expenses are all appropriate one-time labor related expenses, but that "care must be taken that only those expenses which can reasonably be expected to be paid out" should be included in the cost comparison. have consistently held that the mock RIF procedure is a proper method of calculating associated costs, specifically, severance pay and relocation costs, and recognized that such estimates involve complex and somewhat subjective judgments. Dyncorp, B-233727.2, supra; Raytheon Support Servs. Co., B-228032.2, Dec. 20, 1987, 87-2 CPD Para. 641.

The mock RIF indicated that of the 86 persons at the Minneapolis-St. Paul Air Reserve Center affected by the mock RIF, 31 persons would receive severance pay, 5 employees would be relocated, and 6 employees would be retrained. The mock RIF was conducted by unaffected personnel, who reviewed the affected positions and interviewed supervisory personnel.

Intelcom argues that the mock RIF and government estimate improperly allowed for severance pay, and retraining and relocation costs for certain employees, not included in the MEO, who would be affected by a RIF action caused either (1) by the personnel actions necessary to accomplish the MEO and continue in-house performance or (2) by contracting out. Intelcom notes that costs that would be incurred both in achieving the MEO and converting to contract performance cannot be included in the government estimate. However, our review of the record does not substantiate Intelcom's contentions; only the personnel actions directly attributable to a RIF action on the MEO personnel were considered.

Intelcom also notes that the Air Force included costs for retraining, relocating or severance pay for employees not in the MEO but who were "bumped" through application of the RIF procedures. Intelcom asserts that this was improper inasmuch as applicable regulations only allow the costs of MEO employees to be accounted for in the cost evaluation. However, we find that since "bumping" and "retreat" rights are integrated into the RIF procedures, the costs associated with non-MEO employees who were "bumped" by MEO employees as part of a RIF of MEO employees occasioned by a determination to contract out for the function would be directly attributable to that decision and thus are appropriate one-time conversion costs. See AFR 26-1, vol. 1, paras. 9-15.c.(1) and e.

Based on our review of the mock RIF, we have found no indication that it was conducted in an unreasonable or unfair manner. Moreover, since the government estimate was prepared before receipt of proposals without any particular contractor in mind, Intelcom's purported historical data indicating a high rate of attracting incumbent government employees does not show that the government estimate indicating that only 10 of 86 affected employees will accept employment with the contractor is erroneous. /1/ Raytheon Support Servs. Co., B-228032.2, supra, at 3.

Specifically, with regard to severance pay, Intelcom correctly notes that AFR 26-1, vol. 1, para. 9-15.c.(2)(b), which implements para. E.3.c of Chapter 3 of the Handbook, requires a waiver from the Air Force major command before a figure of more than 2 percent of labor related costs can be used for severance pay in the cost comparison. Those regulations provide that this waiver is necessary because past experience indicates that only a small fraction of employees are actually separated from government service when a contractor takes over a function. A waiver can be granted if the evidence presented outlines "exceptionally unique" circumstances that exist making the two percent factor inappropriate and the waiver should not be granted "in the absence of compelling evidence." AFR 26-1, vol. 1, para. 9 15.c.(2)(b).

We have recognized that a waiver of the 2 percent limitation on severance pay can be based upon the results of a mock RIF, which here we have found was not conducted in an unfair or unreasonable manner. See Raytheon Support Servs. Co., B-228032.2, supra. Both the base appeal team and second tier appeal board reviewed the documentation and found that the waiver had been requested and granted by the appropriate authority in the Air Force in accordance with OMB Circular A-76 and Air Force regulations.

Intelcom complains that this waiver is invariably granted for base support services for Air Force Reserve bases, such that the exceptions to the 2 percent limitation on severance pay have swallowed the rule. The Air Force does not deny that such waivers have been frequently granted, but states that each waiver was separately justified. The Air Force further indicates that it is not surprising that the Air Force Reserve bases have similarities which create the exceptional circumstances justifying the waiver. For example, each base has a relatively small work force, a large part of which are Air Reserve Technicians-- a job which requires membership in the Air Force Reserve as a condition of civilian employment-- such that very few ordinary civilian employees, who were affected by a RIF, could be placed in these positions. Further, the Air Force has provided data that shows there is little opportunity for placement of civilian employees into positions at these bases, since few such positions are vacant and the bases are geographically located where other federal employment is not readily available, particularly in the trades and crafts in question here. Under the circumstances, we do not find the oft-granted waivers of the 2 percent limitation for Air Force Reserve bases is indicative that this waiver does not meet applicable requirements or is not in good faith.

With regard to relocation costs, we have recognized that, contrary to Intelcom's contentions, a mock RIF can be used to determine the number of employees who will be relocated if a contractor is selected. See Dyncorp, B-233727.2, supra. Although Intelcom claims that the approximately $20,000 per relocated employee cost is grossly excessive given that Minneapolis- St. Paul is in the middle of the country, the Air Force estimate is based upon fiscal year 1986 and 1987 data of the actual costs of relocating civilian employees from Minneapolis-St. Paul, adjusted to present (1989) value. Notwithstanding that Intelcom complains that the Air Force has selected a questionable and statistically insufficient sample on which to base its cost estimate, we believe the figure is reasonable. See Dyncorp, B-233727.2, supra, at 7 ($22,000 per employee in relocation costs on an A- 76 cost comparison conducted about the same time as the cost comparison conducted here).

With regard to retraining costs, since the Handbook and AFR 26-1 indicate retraining costs can be accounted for as a one-time conversion cost, it is within the sound exercise of government's discretion to include these costs in a cost comparison where they have been properly justified and documented. Inasmuch as we have upheld the use of mock RIF procedures to identify employees entitled to severance pay and relocation costs, we see no reason why mock RIF procedures cannot be utilized to identify what displaced employees should be retrained. The mock RIF identified six employees who would be retrained; Intelcom has not shown that this number is excessive.

Intelcom also asserts the retraining costs are excessive. However, the government estimate documented these costs and they were reviewed in the base and second tier appeal, where they were found justified and reasonable. Under the circumstances, we do not find these costs were improperly added to the Intelcom price as a one-time conversion cost.

Intelcom also questions adding $37,378 in one-time conversion costs for the equivalent of two GS-11 employees (one in the Affirmative Employee Branch and one in the Labor and Employee Relations Branch) for 6 months to supplement the existing personnel office and to process the RIF for the 86 affected employees. The Air Force explains that these personnel were necessary because the RIF entails: "Working RIF procedures to determine best offers, issuing required notices and processing personnel actions, registering employees in the Department 0f Defense (DOD) Priority Placement Program and assisting them in applying under the Office of Personnel Management (OPM) Displaced Career Program, establishing a Reemployment Priority List, counseling employees, computing and processing retirements and handling the inevitable grievances and appeals." The base and second tier appeals reviewed and approved of the inclusion of this cost in the cost comparison. AFR 26-1, vol. 1, paras. 9-15.c. and e. provide that the conversion from in-house performance to contract performance requires "an agency to take certain actions that are not necessary if the activity continues in-house," and that such costs can be added to the contractor price as one-time conversion costs. Therefore, even though AFR 26-1, vol. 1, para. 9-15 does not mention the possibility of including the costs of supplementary personnel to process a RIF as a one-time conversion cost, we find this cost can be included in the cost comparison, where, as here, the agency's need for such personnel is established. Although Intelcom disputes the need, grades and duration of these supplementary personnel specialists, it is apparent that the Air Force is in the best position to determine its personnel needs to accomplish a RIF, and we will not question such a determination unless Intelcom shows it was done fraudulently or in bad faith or is inconsistent with applicable procedures. See Dyncorp, B-233727.2, supra; Trend Western Technical Corp., B-221352, May 6, 1986, 86-1 CPD Para. 437. On this record, we cannot say these costs are not legitimate.

Intelcom next contends that the contract administration costs included in the cost comparison are overstated and that waiver of the OMB Circular A- 76 limitations on such costs was improperly requested and granted. this regard, para. C.3 of Chapter 3 of the Handbook and AFR 26-1, vol. 1, para. 9-13.c. provide that the contract administration staffing requirements should be limited to those specified on tables included in those regulations unless a waiver is obtained. Here, since the 78-person MEO, on which the government estimate is based, was between 66 and 91 persons, the tables in the regulations indicate that 4 persons should be assumed for contract administration for cost comparison purposes. Waiver of this staffing limitation may be requested from the Air Force major command "in the event the function undergoing comparison is so technically specific or geographically dispersed that the contract administration ... personnel requirements would exceed the limits established in" the tables. AFR 26-1, vol. 1, para. 9-13.c.

In this case, based on an operational audit of the function, a waiver of the four person limitation was requested of the major command to allow eight persons for contract administration. The major command activity granted the waiver but approved only six persons rather than eight.

Intelcom asserts that the waiver was improper since it was based upon erroneous data and because the function in question was neither "technically specific" nor "geographically dispersed." Intelcom contends that this resulted in an overstatement of $217,533 in contract administration costs.

While Intelcom makes much of the fact that neither the waiver request nor the waiver mentions the applicable waiver standards and argues there is not a "scintilla of evidence" that the waiver met applicable standards, the Air Force states that in view of the variety of technically specialized disciplines, the function's base-wide coverage, and the operational audit data, the waiver was reasonably requested and granted. The operational audit listed the various contract administration and quality assurance evaluation tasks needed to monitor the contract, together with their estimated frequency and accomplishment times. The audit showed sufficient tasks and time to justify eight contract administration positions, and the record, including the RFP and operational audit, indicates that this requirement involved a wide variety of disciplines.

Intelcom notes that the waiver request indicated that the MEO was 113 persons rather than the actual 78-person MEO, and that this erroneous higher number permits 5 contract administration personnel. However, the base appeal states this was a preliminary figure, that is, the number of positions under study at that time. Moreover, cognizant Air Force officials, as well as the base and second tier appeals, confirmed the propriety of the waiver to allow six contract administration personnel instead of the eight requested. Under the circumstances, the erroneous MEO number on the waiver request is at best a mere procedural defect not affecting the propriety of the cost comparison. Raytheon Support Servs. Co., B-228352, Jan. 19, 1988, 88-1 CPD Para. 44.

Intelcom has challenged, on a task by task basis, the methodology, assumptions and results of the operational audit. Intelcom argues that the operational audit reviewed an inappropriate Air Force Reserve base to gather duration and frequency data for the operational audit and argues that the operational audit did not use the random sampling methods of monitoring contract performance contemplated by Air Force policy.

Intelcom's critique was given in-depth consideration in the base appeal which found that the operational audit underestimated contract administration requirements by 21.59 hours per month. In response to the protest, the Air Force performed another in-depth review of the operational audit based upon more current data and made various adjustments, lowering the estimated times for contract administration tasks by 22.09 hours per month. However, since the operational audit justified eight contract administration personnel, rather than the six persons actually authorized in the waiver, the relatively minor adjustments/errors in the operational audit do not belie the waiver request. Moreover, in the base and second tier appeal decisions, the Air Force found that the operational audit data was obtained from appropriate sources and verified by an actual on-site operational audit not conducted by affected personnel. Although Intelcom contends that the contract should not be so actively monitored as indicated in the operational audit, the protester's disagreement over the number of administrators to assure contract performance provides no basis for our Office to overturn the base and second tier appeals decisions that the waiver was valid. See Dyneteria, Inc., B-225581.3, Jan. 9, 1987, 87-1 CPD Para. 30.

Intelcom next contends that the Air Force in-house estimate did not include properly computed costs for the IQW in line items 0007 and 0010 and that the government estimate is therefore understated by $90,125. The record shows that the government estimated MEO included a 32-person civil engineering group, which was responsible for the regular contract engineering and maintenance work tasks as well as for the IQW in question. Intelcom's contention that the government has miscalculated the IQW is based upon a chart contained in the government estimate that computes the relative percentages of time that each MEO civil engineering person is estimated to spend on IQW, as opposed to regular contract work. This chart shows the Air Force calculated these figures by dividing the number of hours each employee would devote to IQW by 2,087 hours. While Intelcom does not dispute the numerator of this calculation-- the hours devoted to IQW-- Intelcom asserts that the denominator should have been 1,744 hours. In this regard, Intelcom contends that the Air Force calculation violated AFR 26-1, vol. 1, para. 9.5.b., which requires that 1,744 hours be used to calculate the productive work hours of a full-time equivalent employee. Intelcom asserts that the Air Force use of 2,087 hours as the denominator in the equation produced an underestimate of the number of productive manhours by 4,297 hours over the life of the contract with an associated cost of $90,125.

Intelcom has misconstrued the nature of the Air Force chart that gave rise to this basis of protest. For A-76 cost comparison purposes, the number of hours to be used in calculating the costs of government employees is 2,087 hours. AFR 26-1, vol. 1, para. 9-5.d.(4). The number of hours used to calculate productive hours of government employees (considering leave, training and other unproductive times) to determine the number of full time equivalent employees needed to accomplish the work statement is 1,744 hours. AFR 26-1, vol. 1, para. 9.5.b. The chart in question was used solely as a method to break out the respective labor costs of each civil engineering employee's IQW and regular contract work and had nothing to do with calculating the number of personnel needed to perform the contract work. The record shows that elsewhere in the government estimate all hours needed to accomplish IQW were specifically accounted for in calculating the 32 person MEO for the civil engineering branch. AFR 26-1, vol. 1, para. 9 5.d.(4) required 2,087 hours to be used in the cost calculation to determine the yearly cost of employees-- which is what the chart in question was related to. Consequently, we do not agree that the Air Force has not accounted for the hours needed to accomplish the IQW work or the amount of the IQW costs.

Intelcom also protests that the government failed to include 2,585 hours of overtime totaling $98,760 in the cost estimate. This overtime was necessary to accomplish certain tasks covered by the work statement that must be accomplished outside regular duty hours. However, the base appeal notes that the local union agreement provides for "uncommon tours of duty" to accomplish work that ordinarily would require that overtime be paid. Intelcom did not specifically contest this determination in the second tier appeal, but asserted that the overtime in the government estimate was generally understated. The government overtime estimate was reviewed and upheld in the base and second tier appeals and Intelcom has not demonstrated that those decisions were erroneous. We have upheld an agency's estimate for overtime hours where, as here, the estimate is reasonable on its face, recognizing that the government may have inherent advantages in organizing its manpower that a contractor cannot achieve in an A-76 exercise. Dyncorp, B-233727.2, supra, at 10. Consequently, we find this item of the cost estimate was justified.

Finally, Intelcom protests that the government estimate did not provide for wage escalation for certain personnel who Intelcom contends are Service Contract Act type employees, which resulted in an understatement in the government estimate of $45,574 or, alternatively, that Intelcom was misled into including $14,369,79 in such wage escalation costs in its proposal. Intelcom also protests that the government estimate did not provide for 142 hours of training required in line item 0008 in the amount of $3,704.34 and that the government improperly added $5,336 in one-time conversion costs to Intelcom's price for a physical inventory of stock if the function is contracted for. The $64,918.38 difference between the government estimate and Intelcom's total performance cost figure is greater than the total of these remaining contested items. Therefore, we will not consider these matters, since even if we were to resolve all of these points in Intelcom's favor, it would not affect the cost comparison result.

The protest is denied.

/1/ The Air Force states that it checked one of Intelcom's examples of success of attracting incumbent government employees and found that only one person did not receive severance pay by virtue of accepting comparable employment with Intelcom. The Air Force suggests that Intelcom's figures in this regard are grossly exaggerated.