Link: GAO Opinion
Agency: Department of the Army
Disposition: Protest sustained.
Keywords: Evaluated Costs
General Counsel P.C. Highlight: An Agency must have a rational basis to make evaluation adjustments to an offerors proposed costs.
MPRI, Division of L-3 Services, Inc. and LINC Government Services (LGS) protest the award of a contract by the Department of the Army, U.S. Army Materiel Command, to DynCorp International, LLC under a request for proposals (RFP), for Combined Security Transition Command-Afghanistan (CSTC-A), Afghanistan Ministry of Defense (MoD), and Afghan National Army (ANA) mentoring and program support.
The solicitation contemplated the award of a cost-plus-fixed-fee contract, with a two month phase-in, a base period of two years, and one option year, for the services of qualified personnel to provide dedicated in-depth mentoring, training, subject matter expertise, and programmatic support to CSTC-A staff and the Afghan MoD for the purpose of assisting the MoD and associated Afghan National Army (ANA) forces in assuming full responsibility for their own security needs. In total, the SOW provided for the contractor to provide 275 staff in five skill levels or categories, including senior mentor (eight staff), mentor (128), subject matter expert (32), senior trainer (30), and trainer (77), each with specified minimum educational and experience qualifications. Award was to be made on a “best value basis” considering four evaluation factors: (1) capability (with subfactors for key personnel, management plan, technical approach, quality control and transition plan); (2) performance risk, under which the offeror’s past performance was to be evaluated; (3) small business participation; and (4) cost.
While MPRI’s proposal initially received an overall good rating under the capability factor, as well as good ratings under the key personnel and management plan subfactors, these ratings were ultimately downgraded to acceptable on the basis of agency concerns arising from MPRI’s proposal of a significant reduction in direct labor rates relative to those under its incumbent contract. Although MPRI in its proposal generally attributed its proposed reduction in compensation to “updating salaries based on the current market conditions,” the Army determined that MPRI had “grossly underestimated” its labor costs such that a “direct labor cost growth of approximately [REDACTED]% would occur” as MPRI was forced to increase its labor compensation to the current levels under its incumbent contract. The downgrading of MPRI’s proposal under the key personnel and management plan subfactors resulted in the overall reduction to acceptable under the overall capability factor. MPRI’s proposal otherwise received an excellent rating under the technical approach subfactor on account of its exhibiting a “clear understanding of the CSTC-A mission” and including a clear description of how MPRI would perform the SOW; a good rating under the quality control subfactor (notwithstanding the lack of a dedicated quality control manager) on account of such strengths as its tailored quality control process measuring individual and program performance, weekly activity reports to CSTC-A allowing rapid resolution of problem areas, and an annual work plan for each employee and section; and a good rating under the transition subfactor.
While LGS’s proposal received a good rating under the management plan subfactor on account of such strengths as the program manager being empowered to make decisions and commit corporate resources, it received only an acceptable rating under the technical approach subfactor as a result of an evaluated failure to adequately describe its process for mentoring Afghan personnel, substituting instead an over-reliance on prior military experience.
Although DynCorp’s evaluated cost ($249.1 million) was significantly higher than MPRI’s ($212.7 million), the source selection authority (SSA) determined that DynCorp’s proposal was “vastly superior” under the capability factor, noting in particular the “inherent risk” in MPRI’s proposed reduction in salaries and the fact that five of its eight senior mentors will serve only as team leaders and not also as mentors to Afghan officials. The SSA determined that these considerations outweighed MPRI’s lower cost. Likewise, the SSA determined that DynCorp’s superiority under the capability factor, including a strong corporate knowledge of and experience in similar mentoring, in contrast to LGS’s over-reliance on its leaders’ military background, and DynCorp’s existing relationship with its subcontractors, offset its higher cost. The agency thus made award to DynCorp, following which MPRI and LGS filed these protests challenging the evaluation.
MPRI challenges the agency’s cost realism analysis, which resulted in an upward adjustment in its proposed labor rates. GAO states that when an agency evaluates proposals for the award of a cost-reimbursement contract, an offeror’s proposed costs are not considered controlling because, regardless of the costs proposed the government is bound to pay all actual, allowable costs. Consequently, an agency must perform a cost realism analysis to determine the extent to which an offeror’s proposed costs represent what the contract should cost, assuming reasonable economy and efficiency. An agency’s cost realism analysis requires the exercise of informed judgment, and GAO will review this judgment only to see that it was reasonable. While a realism analysis need not achieve scientific certainty, the methodology employed must provide some measure of confidence that the agency’s conclusions about the most probable costs under an offeror’s proposal are reasonable and realistic.
The record shows that the agency initially reviewed offerors’ proposed labor rates for discrepant rates by comparing them to a range of rates for each position calculated based on one standard deviation (OSD) from the average of the five offerors’ proposed rates for the position. The agency then further reviewed the rates based on the circumstances of each offeror, adjusting some, but not all, of the rates outside the range, as well as some, but not all, of the rates within the range. As shown by the chart, MPRI’s proposed labor rates for the labor categories were lower than the OSD range the agency considered realistic. Based on its realism analysis, the agency determined that the reduction was not justified and adjusted MPRI’s rates for all five labor categories upward to the rates under MPRI’s current Afghanistan mentoring contract. This actually left three of the five resulting rates higher than the OSD range.
As an initial matter, GAO finds the agency’s rejection of MPRI’s proposed labor rates as unsupported to be reasonable. GAO states that an offeror has the burden of submitting an adequately written proposal, and it runs the risk that its proposal will be evaluated unfavorably when it fails to do so. MPRI’s proposal generally attributed the percentage reduction in its incumbent labor rates to “current market conditions,” but included no information regarding current market conditions. Further, MPRI’s proposed rates not only were significantly lower than its current rates for the same work, but also were significantly lower than the rates under DynCorp’s MSNTC-I contract. Finally, MPRI’s proposed rates were lower than the average of all offerors’ proposed rates for certain labor categories; lower than the OSD range for other labor categories; lower than all of the other proposed rates for the certain labor categories; and lower than some of the other proposed rates. While GAO finds that the agency reasonably rejected MPRI’s proposed labor rates as unrealistic, GAO does agree with MPRI that the extent of the resulting upward adjustment in the rates was unreasonable. In this regard, GAO reviews an agency’s conclusions about the most probable costs under an offeror’s proposal in view of the cost information reasonably available to the agency at the time of its evaluation. In increasing MPRI’s labor rates to the level under its current contract, thereby rejecting any reduction, the agency’s realism evaluation assumed rates for MPRI that were higher than the average proposed rate for each of the labor categories; higher than the OSD range for three of the five labor categories; higher than the rates proposed by any offeror for three of the labor categories; and higher than the rates proposed by three of the other offerors for the remaining two categories. The adjusted rates for MPRI also were higher than the rates for three of the five labor categories under DynCorp’s similar MSNTC-I contract, which rates DynCorp itself proposed to reduce for this procurement. In some instances, the adjustment left MPRI’s rates significantly higher than these other reference points.
The significance of these reference points in determining the realism of MPRI’s evaluated rates is highlighted by testimony at the hearing conducted by GAO in this matter. In this regard, when asked what the most probable labor rates would be for foreign nationals in Afghanistan, the cost analyst responded that “competition generally dictates what a reasonable price is,” that the “market rates” were determined by competition, and that the average of the rates proposed by the five offerors thus represented “a reasonable starting point.” The cost analyst then went on to state that MPRI’s current contract rates did not represent “the market rates.” GAO concludes that the record does not support the magnitude of the upward adjustments to MPRI’s proposed labor rates, and that the cost evaluation therefore was unreasonable.
MPRI also challenges the downgrading of its technical proposal under the capability factor based on the cost evaluation conclusions. Again, the Army determined that MPRI had “grossly underestimated” its labor costs as MPRI was forced to increase its labor compensation to the levels under its current contract, with the result that MPRI would experience “high turnover, a lack of qualified personnel, and/or be forced to work with personnel of lesser quality than those proposed.” GAO agrees that the technical evaluation was flawed. While it may be that any reduction in compensation would lead to some additional turnover, it is reasonable to assume that the degree to which MPRI’s rates were deemed inadequate determined the extent to which its proposal was downgraded under the capability factor. Thus, since GAO has found that the inadequacy of MPRI’s rates was unreasonably exaggerated in the evaluation–as reflected in the excessive increase in MPRI’s proposed rates–GAO also finds that the downgrading of MPRI’s technical proposal based on the same flawed cost evaluation results likewise was unreasonable. Accordingly, GAO concludes that MPRI was prejudiced by the agency’s actions and sustain the protest on this basis.
LGS and MPRI maintain that DynCorp’s proposal should have been evaluated less favorably under the performance risk factor. GAO states that it will review a past performance evaluation to ensure that it was reasonable and consistent with the solicitation’s stated evaluation criteria and applicable statutes and regulations. The evaluation of past performance, by its very nature, is subjective; a protester’s disagreement with the agency’s judgment does not demonstrate that those judgments are unreasonable.
The evaluation of DynCorp’s proposal as offering low performance risk was reasonable. In this regard, the solicitation provided for the agency to “conduct a Performance Risk assessment based on the quality of the offeror’s recent (within the past three years) and relevant past performance as well as proposed major subcontractors, as it relates to the probability of successful accomplishment of the required effort.” Offerors were required to identify a minimum of three and a maximum of five contracts; describe for each how the effort is “similar to the requirements of [the] solicitation,” the objectives achieved, the reasons for any shortcomings, and any corrective action taken to avoid reoccurrence; and furnish a past performance questionnaire (PPQ) “to the Government/commercial contracting activity and Government/commercial technical representative responsible for the past/current contract. DynCorp’s proposal cited five prior contracts and DynCorp’s MNSTC-I contract was viewed by the agency as most relevant to the requirement here; indeed, the contracting agency for the MNSTC-I contract described it as identical to the requirement here. DynCorp’s proposal cited its accomplishments under that contract, and also cited two letters of concern regarding a delay in the contract award, which had resulted DynCorp’s losing 90% of its job candidates and had caused initial problems in staffing the contract effort. Despite the staffing issues, the contracting agency’s PPQ response for this contract characterized DynCorp’s overall performance as excellent or very good under nearly all PPQ evaluation categories; stated that the contract had “low administrative issues” and that the agency was “[v]ery pleased with the management of this contract and its responsiveness to [the government’s] changing needs”; and indicated that the agency would recommend DynCorp for other contracts. The Army assigned DynCorp a low performance risk rating based on its finding that DynCorp had performed contracts similar in size, scope and complexity to the requirement here; in particular, that it had trained senior leaders from 11 different ministries and agencies under its MNSTC-I contract; that it has had a “permanent presence in Afghanistan since 2003,” including five years mentoring and training the Afghan police; and that its performance had been rated as either exceptional or very good.
The protesters primarily assert that the agency failed to consider the problems under DynCorp’s contracts, and focus in particular on the agency’s failure to receive a PPQ for the CIVPOL-A contract. The evaluation was unobjectionable. As discussed, the agency was aware that DynCorp had encountered problems under its prior contracts. While DynCorp acknowledged shortcomings in its performance under certain contracts, it also explained the measures taken to address the problems. The PPQs received by the agency confirmed that DynCorp’s acknowledged shortcomings were not viewed as significant by the contracting activities in question, and that DynCorp instead was considered to be a very good or excellent contractor. GAO concludes that there is no basis for questioning DynCorp’s low performance risk rating.
GAO sustains MPRI’s protest on the grounds that the cost and technical evaluations were unreasonable and recommends that the agency reevaluate the proposals using a methodology that reasonably accounts for the likely cost for each offeror to staff its proposed capability approach. In addition, GAO recommends that MPRI be reimbursed the costs of filing and pursuing its protest, including reasonable attorneys’ fees. MPRI’s protest is sustained. LGS’s protest is denied.