BNUMBER: B-270827; B-270827.2 DATE: April 30, 1996 TITLE: Crown Healthcare Laundry Services, Inc. ********************************************************************** Matter of:Crown Healthcare Laundry Services, Inc. File: B-270827; B-270827.2 Date:April 30, 1996 Jesse W. Rigby, Esq., Clark, Partington, Hart, Larry, Bond, Stackhouse & Stone, for the protester. Gregory H. Petkoff, Esq., and Kirkland D. Foster, Esq., Department of the Air Force, for the agency. Peter A. Iannicelli, Esq., and Michael R. Golden, Esq., Office of the General Counsel, GAO, participated in the preparation of the decision. DIGEST Protest that the Air Force improperly conducted a cost comparison between the Department of Veterans Affair's (VA) and the protester's offers to do laundry services to justify the Air Force's decision to convert laundry services contract to performance by the VA under interagency sharing agreement is denied where the protest fails to show that the Air Force's methodology was unreasonable or inconsistent with Office of Management and Budget Circular No. A-76. DECISION Crown Healthcare Laundry Services, Inc. (Crown) protests the Air Force's decision to have the Department of Veterans Affairs (VA) perform laundry services for Keesler Air Force Base (Keesler), as solicited under invitation for bids (IFB) No. F22600-95-B-0030, rather than to continue to contract with Crown for those services. The Air Force decision to have the VA do the work was based upon an Office of Management and Budget (OMB) Circular No. A-76 comparison of the estimated cost of VA performance with Crown's bid price. Crown contends that the cost comparison was flawed for various reasons discussed below.[1] We deny the protest. Issued on February 6, 1995, the IFB solicited bids for performing laundry services[2] for Keesler for a basic contract period of 1 year with options for 3 additional years. The IFB indicated that an A-76 cost comparison study would be conducted and included a performance work statement (PWS) to be used in formulating bids and the government cost estimate. The VA provided its cost information to the Air Force along with an interagency sharing agreement[3] indicating its willingness to provide laundry services to Keesler; the Air Force used the VA's cost data to formulate the in-house cost estimate. For purposes of this A-76 cost comparison, the VA was considered the in-house bid. Two bids were received from commercial bidders. The low-priced commercial bid was withdrawn based upon its containing a mistake, leaving only Crown's bid and the VA's estimate in contention. Based upon its cost comparison, the Air Force determined that it would be cheaper to have the work performed in-house by the VA than to continue to have Crown perform the work.[4] After the agency provided the cost comparison study to Crown, Crown appealed the decision to have the VA perform the contract, alleging several errors in the way the VA's costs were calculated. On August 14, 1995, Keesler's administrative appeal review team denied Crown's appeal, finding that the cost comparison was accurate and complete. Crown filed a second administrative appeal which was denied by the Air Force on December 12. Shortly thereafter, Crown filed the first of two protests in our Office.[5] The protester contends that the Air Force's cost comparison was flawed because the Air Force relied upon VA cost estimates that underrepresented the actual costs of having the VA do the work. The protester states that the VA's cost estimate was based upon doing less work than was described in the PWS upon which Crown based its bid. Crown also argues that the VA underestimated the weight of the laundry to be cleaned, thereby understating the actual costs of in-house performance. Crown further contends that the Air Force erred because it added the agency's costs of administering a contract with Crown to Crown's bid but did not add the cost of administering its contract with the VA to the VA's cost estimate. OMB Circular No. A-76 describes the executive branch's policy on the operation of commercial activities that are incidental to performance of government functions. It outlines procedures for determining whether commercial activities should be operated under contract by private enterprise or in-house using government facilities and personnel. Generally, such decisions are matters of executive branch policy that our Office declines to review. However, we will review A-76 decisions resulting from an agency's issuance of a competitive solicitation for the purpose of comparing the cost of private and governmental operation of the commercial activity to determine whether the comparison was faulty or misleading. See United Media Corp., B-259425.2, June 22, 1995, 95-1 CPD para. 289; Tecom, Inc., B-253740.3, July 7, 1994, 94-2 CPD para. 11. In an OMB Circular A-76 cost comparison, bidders and the government should compete on the basis of the same scope of work. See DynCorp, B-233727.2, June 9, 1989, 89-1 CPD para. 543. OMB Circular A-76 requires agencies to prepare in-house cost estimates on the basis of the most efficient and cost effective in-house operation needed to accomplish the requirements. See Tecom, Inc., supra. We review agency decisions to perform services in-house instead of contracting for them solely to ascertain whether the agency followed the announced "ground rules" for the cost comparison. Id. Here, we find that Crown's contentions that the agency's cost comparison was flawed are without merit. We first address Crown's contention that the VA's cost estimate was based upon doing less work than was called for in the PWS. Basically, Crown contends that the PWS requires whoever performs the laundry services to count each laundry article at the time of pickup from and delivery to Keesler. Crown states that its bid prices included the costs of two delivery trucks, two drivers, and four other employees who would assist in counting the laundry items, while the VA's estimated costs were based upon having only one delivery truck and driver provided by a private firm. Crown asserts that on some days the workload volume will make it impossible for the VA to count the items and make timely pickups and deliveries using only one truck and driver. Crown also asserts that the VA's inability to make timely pickups and deliveries will be exacerbated because the truck to be provided to the VA by the commercial firm is too small for the average daily workload. The Air Force reports that the same PWS was presented to both Crown and the VA for bidding purposes and that the PWS does not require counting of all laundry items at the time of pickup/delivery. The agency states that it never intended for the launderer to count Lot I medical center items (i.e., patient laundry), which comprise about 35 percent of the total workload, because of the possibility of contamination. The agency states that the rest of the laundry will to be counted by government clerks at each pickup/delivery point and, therefore, it will not be necessary for the launderer to count the items again as the agency's count is considered accurate and may be accepted by the launderer. The contracting officer also asserts that the weighing of patient linen being picked up and counting of other items (i.e., hospital uniforms) being delivered will not consume enough of the driver's time to interfere with the delivery schedule. Moreover, the contracting officer states that the VA is not committed to using a particular commercial firm for pickup/delivery and that the VA may use any contractor it chooses or may perform the task itself.[6] We think that the Air Force reasonably interpreted the PWS as not requiring the launderer to count each article of linen at pickup and delivery. The PWS requires the launderer to "receive, account for, launder and return" all items. While the protester argues that "account for" means that the laundering entity must actually count each individual item at both pickup and delivery, Webster's New Collegiate Dictionary defines "account for" as the furnishing of a justifying analysis or explanation; the definition does not state that to account for the items means literally counting items. The PWS does not require the contractor to count items, leaving to the discretion of each offeror to decide how to accomplish this task. For example, the PWS specifies that Lot I medical center items (i.e., patient laundry)-- the largest category of laundry by far--will be picked up in a commingled state in laundry bags and weighed rather than counted. The PWS also specifically states that an Air Force representative will count all laundry items before turning them over to the laundry truck driver. Even though the launderer might want to check the Air Force's count of items picked up, it appears that one reasonable approach would be for the launderer to accept the Air Force's count initially and then confirm its accuracy or inaccuracy during the cleaning process, at which time the launderer is required to mark-in and tag the items, examine and note their condition, and to count the items to ensure that all items turned in for cleaning are, in fact, returned to Keesler. Of course, if the Air Force count were regularly found to be inaccurate, the launderer could always decide to count the items itself or initiate other remedial action. Crown's contention that the VA's private contractor's truck is too small is not convincing. The PWS does not specify any number or size of trucks that must be used; rather, the PWS simply states that delivery vehicles shall be of sufficient size and quantity to make timely pickup and delivery. In this regard, we note that Crown's laundry facility is approximately 125 miles from the Keesler pickup points, while the VA's facility is only 12 miles distant. Thus, as a practical matter, the VA could use one, smaller truck and make several trips a day as the contracting officer asserts. Furthermore, the VA obtained a quotation from the commercial trucking firm for estimating purposes but is not obligated to use that particular firm or any particular size truck. The protester next contends that the cost comparison was flawed because it used VA cost estimates that were based in part on the VA's estimate of the total weight of the laundry to be cleaned. Crown contends that the VA's weight estimate was too low. Since the VA used its weight estimate to calculate variable costs (such as personnel, supplies, utility usage, and equipment maintenance), Crown believes that the VA's cost figures greatly understate the actual cost of doing the work. In calculating its costs, the VA converted the IFB's estimates[7] of the number of items of each type into pounds and then used the total estimated pounds to calculate its variable costs. The VA estimated the total weight of the workload to be 2,400,211 pounds per year.[8] However, by weighing several of each type of linen and computing the average weight, Crown calculated that the actual weight of laundry is 2,615,616 pounds per year. Thus, Crown contends that all of the VA's variable costs would have to be adjusted upward to account for the additional poundage. Crown's argument is not persuasive. The record shows that the VA weighed 100 linen items of each type and used the average weight of each type article to compute the overall weight of the laundry. During the course of the two administrative appeals, both Crown and the Air Force weighed various linens to see if the VA's methodology for calculating the total pounds was reasonable. In fact, the sampling of items and their weights contained in the record shows that the weight of any particular type of item can vary considerably. For example, the VA weighed three double sheets and found their weights to be 1.72, 1.79 and 1.75 pounds; the Air Force weighed 20 double sheets and found their average weight to be just 1.5 pounds; and Crown weighed two double sheets and found their weights to be 1.92 and 1.76 pounds. The unrefuted explanation for these significant variations is that the weight of linen items routinely varies due to variations in size, style, thread count, fabric blend, manufacturer, number of times washed, and amount of humidity in the air. From this and other examples in the record, it is clear that independently computed total weight estimates, representing hundreds of thousands of linen items of many different types, could easily vary by a significant amount. We have no reason to dispute the validity of the VA's weight estimate. The figures in the IFB, whether expressed in pounds or number of items, were merely estimates. Moreover, the IFB did not require offerors to convert the estimates from numbers of items to pounds, and, consequently, the IFB did not prescribe a conversion factor or particular methodology for making such a conversion. Thus, it was up to each offeror to best determine how to compute its bid price or costs. As shown above, we think the VA, which is experienced in providing large-scale laundry services, devised a rational method for calculating the weight of the laundry, which it then used to compute its cost estimates. In addition, the Air Force points out that the VA facility currently has seven machines operated by four employees, that each machine currently cleans only two-thirds of its potential capacity, and that the unused capacity is more than enough to clean all of Keesler's laundry. Therefore, the Air Force states, and it is unrefuted by Crown, that the VA has a large amount of unused capacity and will be able to clean significantly more laundry than it has estimated without purchasing new cleaning machines or adding operators. Moreover, the VA points out that not all of its utility costs would increase by its performing work for Keesler. For example, the VA's plant is fully air conditioned (air conditioning represents about 25 percent of the VA's total utility costs) and the level of air conditioning is not dependent on the amount of laundry cleaned for Keesler. (The VA nonetheless included a proportionate share of the cost of air conditioning and other utilities in its overall cost estimate.) As discussed above, we think the Air Force reasonably determined that the VA could do the laundry services as described in the PWS at the cost it estimated and properly used the VA's cost estimates in its A-76 comparison. While the PWS fully described what was required of the laundering facility, Crown and the VA have very different ideas of how many workers will be needed to perform pickup/delivery tasks. In keeping with the requirement for using the most efficient in-house operation, the VA decided that it could accomplish pickup and delivery tasks using a minimal crew, while Crown believed it needed a much larger crew. These two experienced offerors also used similar methodologies to estimate the weight of the total workload with very different results. In our opinion, Crown and the VA simply exercised their independent business judgments and skills and arrived at different logical conclusions regarding the cost of doing the work. Thus, Crown has not shown that the VA's cost estimates or the Air Force's use of them in the cost comparison were unreasonable. Crown also contends that the Air Force should not have added $201,934 in contract administration/quality assurance costs to its bid price, because it will cost the Air Force a comparable amount in contract administration/quality assurance costs if the VA performs the work and the Air Force did not add such costs to the in-house cost estimate. However, in view of our finding that the Air Force otherwise reasonably estimated the costs of VA performance, we do not believe that Crown suffered any competitive prejudice, because the estimated cost of having the VA perform the work was $424,740 less than contracting with Crown. Therefore, even if the extra amount representing contract administration/quality assurance costs were deducted from Crown's cost estimate, Crown would still represent a higher cost to the government. See Orbital Sciences Corp., B-254698, Jan. 5, 1994, 94-1 CPD para. 2. In any event, the record shows that the Air Force will be utilizing the VA's excess laundry capacity and intends generally to rely upon the VA's management for quality assurance. Moreover, the VA cost estimate included the salary of a VA linen coordinator and the Air Force included in the VA estimate the cost for an Air Force housing manager; these two employees will perform quality assurance, liaison, and some administrative tasks. Thus, the actual costs of quality assurance and contract administration are essentially included in the cost comparison. The protest is denied. Comptroller General of the United States 1. The protester raised numerous arguments and provided details in its two appeals to the Air Force and in its protests before our Office. We reviewed the entire record, considered all of the arguments, and found no basis for sustaining the protests. We will discuss only the dispositive arguments in this decision. 2. The IFB segregated the laundry into six lots, generally representing various organizations at Keesler, as follows: Lot I - medical center (further divided into patient laundry and medical staff uniforms); Lot II - billeting; Lot III - linen exchange; Lot IV - Keesler organizations; Lot V - commissary; Lot VI - child development center. 3. In its February 15, 1996, letter commenting on the agency's report on its initial protest, Crown filed a supplemental protest alleging that the interagency agreement which allows the VA to compete is illegal. This later-raised issue is untimely under our Bid Protest Regulations, 4 C.F.R. sec. 21.2(a)(2) (1996). The record shows that Crown raised this issue with the Air Force in letters dated May 9 and October 10, 1995, but did not raise them in its initial protest with our Office. Therefore, because these supplemental issues are untimely and were raised in a piecemeal fashion, they will not be considered. See GEC Avionics, Inc., B-250957; B-250957.2, Feb. 25, 1993, 93-2 CPD para. 24. 4. The agency calculated that the actual cost of contracting with Crown would be $3,911,490 for the base plus option years while the actual cost of having the VA perform the work in-house would be $3,282,451. After adding $204,299 (representing the costs of converting from contracting to in-house performance) to the VA cost estimate, the Air Force concluded that having the VA perform the work, instead of contracting out to Crown, would save the government $424,740. 5. Crown continues to perform the laundry services for Keesler pending our decision on its protest. 6. The VA did not submit to our Office any comments on the protests or on the Air Force's reports. However, in response to Crown's administrative appeals, the VA stated that Crown's manpower estimates for laundry distribution and related tasks are "quite extravagant." The VA stated that its cost estimate was based in part upon a price quote received from a commercial trucking company after a representative from that company was shown the PWS and given a tour of the pickup and delivery sites involved. 7. The IFB included estimates of the amount of laundry in each lot for use in preparing bids and cost calculations. The estimates were stated either in terms of the number of pounds or the number of articles to be cleaned. For example, medical center patient laundry was estimated to consist of 835,500 pounds, while medical staff uniforms were listed by type and number of items (e.g. 4,025 dental smocks). 8. In considering Crown's administrative appeals, the Air Force increased the VA's estimate to 2,450,558 pounds for the purpose of comparing Crown's bid price with the VA's costs.