If you have read the prior posts in this series, you are aware that agencies conduct realism analyses as part of proposal evaluation for cost-reimbursement contracts to ensure that “the estimated proposed cost elements are realistic for the work to be performed; reflect a clear understanding of the requirements; and are consistent with the unique methods of performance and materials described in the offeror’s technical proposal.” FAR 15.404-1(d). The last element?consistency with an offeror’s proposed technical approach?provides a basis for a protest when a disappointed offeror learns or suspects that an agency failed to consider its technical approach when assessing its proposal for realism. This post describes the facts and rulings in three cases that illustrate the protest issues that can develop from such analyses.
Offerors Can’t Have It Both Ways
In Magellan Health Services, the solicitation instructed offerors to base their cost proposals on cost tables included in the solicitation, which provided labor categories and an estimated quantity of hours for each category. Offerors were instructed to propose costs for the levels of effort described in each table and to submit a cost proposal that was fully supported. The awardee based its proposal on the agency’s estimate for all but two labor categories; and it proposed levels of effort that were lower than the estimate for those two categories.
During discussions, the agency expressed concern that the awardee’s proposed labor rates were too low and would make it difficult to hire and retain qualified personnel. In its final proposal revision, the awardee stated that it would (i) match existing salaries and (ii) increase them for a standard merit raise. But those increases were not reflected in its final cost proposal. When the agency made its award decision, it considered only the offerors’ proposed costs and not the concerns expressed about the awardee’s labor rates or the agency-determined probable costs.
GAO explained that there were several problems with the agency’s cost realism analysis. First, GAO faulted the agency for making its award decision based on the offerors’ proposed prices, rather than the probable costs. In accepting the awardee’s proposed costs, the agency failed to consider the $1 million increase that resulted from the cost evaluation and discussions. The agency’s conduct contradicted the purpose of a realism analysis—determining how well the proposed costs and profit represent the cost of the contract—because the agency omitted included costs (i.e. the awardee’s proposed fixed fee). Second, GAO criticized the agency for applying a downward adjustment to the awardee’s proposed costs because an agency should apply a downward adjustment only when the agency concludes that the actual costs will be lower than proposed—not because the proposed costs are insufficiently documented. In Magellan Health, the agency omitted the costs of the awardee’s proposed signing bonuses and educational materials in its cost proposal because the awardee did not provide supporting documentation. Third, GAO criticized the agency’s failure to address the inconsistency between the awardee’s technical approach and its proposed costs, explaining: “a proper cost realism evaluation prevents an offeror from improperly ‘having it both ways’?that is, from receiving a technical evaluation rating based on its proposed performance but failing to propose costs that reasonably reflect that performance.” GAO found that the cost analysis defects totaled more than $10 million and sustained the protest.
Assumptions in Price Analysis Must Be Supported by Technical Approach
When an agency relies on aspects of an offeror’s technical approach in its price realism analysis, the agency’s reasoning must be consistent with the offeror’s technical approach as proposed. In Solers, Inc., the protesters challenged the award of a task order, arguing that the agency did not reasonably evaluate the realism of the awardee’s proposed labor mix. In its analysis, the agency considered the offerors’ direct labor rates, direct labor hours, indirect rates, travel and other direct costs, subcontractor costs, and fee. However, the Agency did not consider the offerors’ proposed labor mix or level of effort, and it failed to distinguish between the fixed-price and cost-reimbursement elements.
At GAO, the agency argued that its price analysis was reasonable and that it considered the unique aspects of each offeror’s technical approach because the evaluators had asked the offerors questions about their proposed labor mixes and levels of effort during discussions. GAO rejected this argument, explaining that although the discussion questions revealed concerns about the offerors’ labor mixes and levels of effort, the agency failed to determine that the elements were realistic. Instead, the record contained generalized conclusions that the awardee’s prices were realistic without any supporting analysis.
GAO also determined that the agency’s cost-related conclusions were not supported by the awardee’s technical proposal. For example, when the agency determined that the awardee proposed too few personnel for one labor category and too many for another, the agency assumed that the awardee would be able to shift personnel from one category to another. There was nothing in the awardee’s technical proposal to support this assumption. GAO concluded that the record did not demonstrate how the agency determined that the awardee’s proposed price/cost was realistic for the work to be performed and sustained the protest.
The Agency Must Consider Labor Rates
An agency’s price realism analysis will likely not withstand GAO scrutiny when it fails to consider a significant aspect of an offeror’s costs based on its technical approach. In Metro Machine Corp., the protester challenged the award of a contract for maintenance and modernization work on two classes of ships, arguing that the agency mechanically applied the Government estimates for labor hours and material costs and failed to account for each offeror’s technical approach. Offerors were advised to submit proposed estimated cost data based on two notational work item packages, and the cost realism analysis would focus on those proposed costs. Offerors were instructed to use Government estimated labor hours and material costs but were allowed to deviate from the estimates if the deviations were supported by clear and compelling evidence. When the agency evaluated the offerors’ proposals, it denied almost every proposed deviation. As a result, the offerors’ probable costs were all based on the same labor hours and material costs.
The protester argued that if the agency had accounted for the awardee’s technical approach?instead of mechanically applying the Government estimate?it would have increased the awardee’s proposed costs. GAO rejected this aspect of the protest, finding that although GAO had sustained protests based on similar mechanical adjustments in the past, in this case, the solicitation advised offerors that deviations from the estimate would be rejected if they were not supported by clear and compelling evidence. Because the protester had not challenged the solicitation, this protest allegation was untimely.
However, GAO sustained the protest on another cost realism ground. The awardee had proposed to team with other contractors and provided the percentage of work the team members would perform. In its evaluation, the agency did not account for the fact that the awardee’s teammates had proposed higher direct labor rates. GAO reasoned that this failure was particularly problematic because by using the Government estimate for labor hours and material costs, labor rates became the sole means to compare the offerors. By failing to take into account a major aspect of the awardee’s technical approach (and the effect on its proposed costs), the agency failed to account for the awardee’s true costs of performance. Accordingly, its cost realism analysis was improper.