Jump to content
The Wifcon Forums and Blogs
  • entries
  • comments
  • views

When a source selection takes longer than World War II

Vern Edwards


This blog post is adapted from an article I wrote for the June 2017 issue of The Nash & Cibinic Report.

Copyright 2017 is by Thomson Reuters.

         After the attack on Pearl Harbor on December 7, 1941, it took the United States 1,366 days to defeat Japan and receive its formal surrender on the U.S.S. Missouri on September 2, 1945. Thus far, it has taken the United States Army 1,416 days and counting to award a contract for support services at its Yuma Proving Ground. In the course of that procurement the Army has so far made three source selection decisions and suffered four protests to the Government Accountability Office and four protests to the U.S. Court of Federal Claims. The Army has taken corrective action twice and has proposed to do so a third time. It has had two different source selection authorities, two different chairs of the source selection evaluation board, and three contracting officers. The source selection was still underway as of the end of March 2017. The Yuma Proving Ground service procurement is evidence of two problems that plague Government contracting: first, a failure of professional competence and second, the continuation in the 21st Century of a 19th Century procurement system under the Competition in Contracting Act of 1984 (CICA) as amended.

         As described in the latest protest decision in this saga, Jacobs Technology, Inc. and TRAX International Corp., --- Fed. Cl. ---, 2017 WL 1291795, March 31, 2017, the request for proposals, issued on May 16, 2013 and amended six times, called for a cost-plus-award-fee contract with a one-year performance period and three one-year extension options. The contractor’s job would be to support military testing. Historically, the incumbent contractor’s staff had consisted of approximately 1,234 full-time equivalents.

        The source selection evaluation factors were mission capability, past performance, small business participation, and cost. The three non-cost factors were approximately equal in importance to each other and, in the aggregate, approximately equal to cost. The RFP required offerors to submit “limited” written proposals and make oral presentations. The Army received four proposals, including one from the incumbent contractor, TRAX International Corp. It took the Army until September 10, 2014—483 days after issuance of the RFP—to make its first award, to Jacobs Technology, Inc., which it determined to have “the lowest priced offer of the most highly rated proposals.” Jacob’s evaluated probable cost was $411,678,388. The contracting officer found the incumbent’s higher evaluated cost to be unreasonable.

A Nineteenth Century Procurement Process

         It is in the CICA requirement to evaluate cost that we find a 19th Century procurement system at work to the Government’s detriment. In order to ensure that offerors had a common basis for estimating cost, the RFP required offerors to base their estimates on an historical labor “baseline” established by the Army and included with the RFP. Unfortunately, the baseline did not include crucial information, the result being that the offerors did not, in fact, estimate costs on the same basis. The Army discovered the problem during the proposal evaluation.


In June 2014, before the first contract award, two procurement officials on the “Cost Team” issued a clarification to the Source Selection Advisory Council to explain the “significant variance in cost” between the two highest-rated offerors, Jacobs and TRAX. The Army attributed the variance to a “[p]roblem” it found with the historical baseline staffing data provided in Technical Exhibit 10, item 6, specifically noting that the data did not include productive hours, average wage rates, or full-time equivalents. Thus, when Jacobs and TRAX submitted baseline staffing proposals, they took different approaches. Jacobs and TRAX each provided a baseline number of total labor hours that would be expended in the base period of the contract. Both Jacobs and TRAX also calculated the baseline total labor hours for the first year by multiplying (1) average productive labor hours per employee and (2) 1,234 employees, the figure provided in Technical Exhibit 10, item 6. See id. However, for the first prong of the equation, Jacobs’ productive labor hours per employee was [***], whereas TRAX’s productive labor hours per employee was [***]. See id. As a result, before either offeror applied any labor efficiencies that could reduce the offeror’s proposed hours, Jacobs proposed [***] total baseline hours and TRAX proposed [***] total baseline hours for the base period. This baseline number affected the cost of the proposals.

That’s when the contracting officer should have called a halt, amended the RFP to provide the missing information to offerors, instructed offerors to modify their proposals if they wished to do so, and then allowed the evaluation to proceed. Instead, the Army pressed on.


Despite the different approaches, the Army determined that even if TRAX and Jacobs had proposed the same number of hours, whether through Jacobs’ or TRAX’s method, Jacobs would still have proposed a lower cost. The Contracting Officer thus found that Jacobs presented “the lowest priced offer of the most highly rated proposals,” and the Source Selection Authority (“SSA”) determined that Jacobs’ proposal “provide[d] the best overall value,” TRAX’s proposed cost was deemed “unreasonable.” Accordingly, on September 10, 2014, the Army awarded the contract to Jacobs. [References to the administrative record omitted.]

         When TRAX, the incumbent, discovered the baseline problem after contract award, it protested to the GAO on that and other grounds. The Army announced that it would take corrective action and the GAO dismissed the protest. And then, astoundingly, the Army did not fix the problem.


As with the first award evaluation, a procurement official on the Cost Team issued a clarification to the Source Selection Advisory Council to explain the variance in Jacobs’ and TRAX’s costs. That clarification identified the same baseline staffing issue related to Technical Exhibit 10, item 6, and indicated that Jacobs and TRAX had again applied different methodologies to determine baseline staffing for the base year, resulting in a lower baseline of labor hours for Jacobs. Nonetheless, the SSA found that Jacobs’ proposal “provide[d] the best overall value” based upon an evaluation of the four factors described in the RFP, and the Army awarded the contract to Jacobs for a second time on December 10, 2015. [References to the administrative record omitted.]

TRAX again protested the Army’s failure to evaluate provide a common cost baseline. The Army again announced that it would take corrective action and issued Amendments 7 and 8 to the RFP. The GAO again dismissed the protest. TRAX and Jacobs then protested the amendments and the Army issued Amendment 9, cancelling them. TRAX and Jacobs submitted revised cost proposals, and guess what?


Significantly, Jacobs again submitted a lower cost than TRAX for total evaluated probable cost/price. As before, Jacobs proposed a lower number of baseline productive hours for the first year of the contract… The [Source Selection Authority] acknowledged Jacobs’ and TRAX’s different approaches regarding overtime, but found that both Jacobs and TRAX had complied with the RFP by beginning with 1,234 full-time equivalents and then applying management approaches and efficiencies to determine baseline staffing. The SSA further determined that Jacobs offered [***], and ultimately found that Jacobs’ proposal presented the best overall value. On October 28, 2016, the Army awarded the contract to Jacobs for a third time.  [References to the administrative record omitted.]

TRAX protested yet again. This time it complained not only about the failure to establish a common baseline, but also that the Army had allowed Jacobs to change its oral presentation slides without requiring it to make another oral presentation, as required by the RFP. The Army implemented a CICA stay in accordance with FAR 33.104(c). Jacobs filed suit at the Court of Federal Claims complaining that the Army should have overridden the stay. The Army announced that it would take a third corrective action. The GAO again dismissed TRAX’s protest, but Jacobs filed suit to block the Army from taking its proposed corrective action. TRAX then sued complaining about the Army’s continuing failure to solve the common baseline problem and alleging bias on the part of Army officials. The court denied the protests, freeing the Army to proceed with a third corrective action. That is where things stood as of last March 31.

         The Army got into trouble because it tried to find a way to provide offerors with a basis to estimate what could not be estimated—the annual cost of a complex, long term, frequently changing service requirement, a problem that the court described beautifully at the end of its decision.


TRAX argues that it has made a threshold showing of conduct that is hard to explain absent bad faith, stating that the Army has “knowingly and deliberately ignored” problems related to baseline staffing during the three awards in this procurement, “each time to the benefit of Jacobs”… TRAX, however, has failed to make a threshold showing of conduct by the Army that is hard to explain absent bad faith… Rather, the errors appear to simply reflect the difficulties the Army has experienced in attempting to set a template for baseline staffing. Such difficulties stem, at least in part, from the nature of the Yuma testing facility, which has an uneven workflow and staffing needs that are hard to predict. See, e.g., AR 3-97 (stating that the Yuma facility requires an “ever-changing work effort” where the demands “change[] constantly” and contractors must be “flexible and adaptable”); AR 5b-457 (explaining that the Army could not provide “forecasted” information regarding full-time equivalents needed beyond the base period of the contract).

There was no way for any of the offerors to estimate what it would cost to provide support services at Yuma Proving Ground for a year. As the Army had explained to the court:


Yuma’s primary mission is to provide “the most flexible, responsive, innovative and diverse set of capabilities and services across the spectrum of natural environments.” It tests military equipment of varying types, with the materials tested and the “associated infrastructure necessary to conduct each test chang[ing] constantly.” The Army describes Yuma as a “premier Army asset” with capabilities that are continually expanding, resulting in an increased need for personnel to support the testing conducted at the facility.

The only way that the Army could comply with CICA and evaluate cost was develop a common cost baseline, tell the contractors to plug in some numbers, and compare results as if they were meaningful indicators of what the offerors’ performance costs would be. But the resulting cost estimates probably bore little if any meaningful relationship to what would happen after contract award. They probably had little if any meaningful predictive value. The apparent cost difference between Jacobs and TRAX may have been illusory. The Army used the numbers so it could “include cost or price to the Federal Government as an evaluation factor that must be considered in the evaluation of proposals,” as required by10 U.S.C.A. § 2305(a)(3)(A)(ii) and 41 U.S.C.A. § 3306(c)(1)(B).

      Orders are orders. Onward the Light Brigade. A lot of wasted effort, time, money, and litigation, the waste being attributable to a 19th Century obsession with cost competition, abetted by incompetent people. In today’s world of procurements of long-term, complex support services, CICA must be amended to permit source selection without evaluation of price or cost in appropriate circumstances, followed by one-on-one negotiations with the selectee before contract award.

A Better Approach To Support Services Contracting

         A better way to have contracted for the Yuma Proving Grounds support services would have been to award a contract with two line items. One line item would be for a firm-fixed-price contract with award-fee for day-to-day general management and supervision services and for administrative services such as safety, quality control and assurance, purchasing and materials management, property management, human resources management, and accounting. The other line item would be for a cost contract (no fee) for materials, labor, and other direct costs to be incurred for performance of support services. Under such a contract, with a day-to-day contractor management team in place and an advance agreement on any indirect cost rates applicable to the cost (no fee) line item, the parties would jointly plan and manage performance. The contractor would make its best effort to execute the plan by performing the support services under the cost (no fee) line item.

         Instead of pretending that the contractor would independently plan, manage, and execute performance in accordance with an unavoidably incomplete "performance work statement," the parties would jointly plan the work to be done and its scheduling, budget, required materials, labor staffing, and labor compensation. The Government would have the final say on planning decisions. The Government would obligate funds for the firm-fixed-price with award-fee line item upon contract award, but it would fund the cost (no fee) line item in the amount required on a quarterly basis. The contractor would then execute through day-to-day performance.Such an approach to contracting would be relational, rather than transactional. See Edwards and Nash, “A Proposal for a New Approach to Performance-Based Services Acquisition,” Defense Acquisition Review Journal (September 2007), and Vitasek and Frydlinger, “Are You Suffering From the Contracting Paradox? Introducing the Relational Contract,” Contract Management (May 2017).

          An agency would choose the contractor for such a contract on the basis of the best-qualified offeror with a fair and reasonable price, as described in “Postscript: Highest Technically Rated Offerors With Fair and Reasonable Pricing—The GAO’s Sevatec Decision,” The Nash & Cibinic Report (March 2017). The agency would evaluate only the offerors’ qualifications--based primarily on experience, past performance, and a brief oral presentation about their company and proposed contract management organization--and the fairness and reasonableness of their proposed firm-fixed-price and award fee for the day-to-day management services. Price evaluation would not involve nonprice/price tradeoffs or direct comparisons of offerors' prices. Since offerors would not propose estimated costs for performance of the support services under the cost (no fee) line item there would be no need to establish a common cost estimating baseline, and no need to evaluate the realism and reasonableness of offeror estimated costs. The agency would get a contract that is better adapted to the realities of long-term complex service contracting, and they would get it through a simpler source selection process.

          A little innovation can go a long way.



Recommended Comments

All this is absolutely correct. When I choose a consultant for a relatively undefined project, I am looking for knowledge, experience, and approach. Cost is a vanishingly small concern. So long as the cost is within the range of "reasonable" it's good. The government almost never takes a similar approach--and it should, in many cases.

We have gotten away from the notion, which was in vogue 20 years ago, that the government and contractor were "partners" in their contracts. (Remember "Alpha Contracting"?) Until we walk away from the adversarial positions, enforced by policy guidance from current "leadership," we are not going to see fundamental changes in acquisition strategy, and we are going to see suboptimal outcomes.

Share this comment

Link to comment

Why no fee on the Cost portion?   Yuma Proving Grounds isn't really in the same realm as the Jet Propulsion Lab, etc.  I would make the whole thing subject to the Award Fee process (to paraphrase Adam Smith: "the best way to incentivize a for-profit entity is...wait for it... profit...").

Share this comment

Link to comment

The no fee on the cost portion made sense to me.  To agree on a fee, the parties have to agree on an estimated cost -- but here, the parties cannot agree on an estimated cost, so they cannot agree on a fee.  The fixed price portion can include whatever fee the hand of competition will allow.

Share this comment

Link to comment

I like it! Contracting approach that is relational, rather than transactional. That would be a good approach to conducting business.

Share this comment

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now