Jump to content

The End of Incentives?

Posted by Vern Edwards, 01 December 2009 · 4,086 views

On October 14, the FAR councils published an interim rule about incentive contracts. (FAC 2005-37, 74 FR 52858.) FAR 16.401(d) now requires a determination and findings signed by the head of the contracting activity (HCA) before using any incentive contract. Given the bureaucratic hassle of processing such a formal document, I think we might reasonably expect to see a decline in the use of incentives. Given the emphasis on incentives up until a few years ago, this is an interesting policy development.

Incentives have been around for a long time. In a 1980 article in National Contract Management Quarterly Journal entitled, "Incentive Contracting in the Aerospace Industy, Part I," Arthur J. Nolan wrote:

They called it the "bonus for savings" provision in World War I, and the "target price" contract in World War II; but contrary to popular belief, there is every evidence that what we now know as the incentive contract has had not isolated--but abundant--use in government procurement throughout the last half century. Neither is the multiple incentive contract, the more complex incentive arrangement, to be regarded as an innovation; its use dated back at least to World War II.

There are two main categories of incentives. First, there are the predetermined formula type incentives, which include (1) the fixed-price incentive (firm targets), (2) the fixed-price incentive (successive targets), and (3) the cost-plus-incentive-fee. (FAR 16.402.) Second, there are the subjective, after-the-fact incentives, which include the cost-plus-award-fee and the fixed-price with award fee. (FAR 16.401(e), 16.404, and 16.405-2.) The fixed-price incentive (firm targets), which is the most complex incentive, is said to have been invented by the Navy during World War II, but abandoned because of a shortage of auditors. The cost-plus-incentive-fee arrangement, which is the simplest incentive, has been in use since at least the early 1950s. The cost-plus-award-fee incentive was invented in the early 1960s, either by the Navy or NASA or both, and has since become the most popular of the incentives.

The modern era of incentives began in the 1960s when President Kennedy's Secretary of Defense, Robert S. McNamara, urged their use in lieu of cost-plus-fixed-fee contracts, claiming that they would save money. Doubters emerged right away. Influential members of Congress complained that contractors inflated their cost estimates in order to negotiate artificially high targets and then underran so as to increase their profit or fee. That concern lead to the passage of the Truth in Negotiations Act, Public Law 87-652, in 1962. (Now implemented in FAR 15.403.) The Act originally would have applied only to incentive contracts, but DOD and NASA complained that such application would make it hard to persuade contractors to accept incentives, so Congress applied it to all negotiated contracts expected to be valued in excess of $100,000. (The threshold has since been increased to $650,000.) Economists, most notably Harvard professor Frederick Scherer and Rand analyst Irving Fisher questioned the underlying assumptions of the advocates of incentives. Most serious studies concluded that there was little verifiable evidence that monetary incentives had motivated contractors to do what's necessary to reduce contract costs or that costs had in fact been reduced. In a typical report, "Incentive Versus Cost-Plus Contracts in Defense Procurement," Journal of Industrial Economics (March 1978), researchers John Hiller and Robert Tollison wrote:

In this paper, a theoretical analysis, based upon wide-ranging empirical studies (see footnote 2), is made of the following question: Given the out-comes of thousands of contracts over the I950s, I960s, and I970s, has there indeed been a 10% cost saving associated with the introduction of incentive contracts? Indeed, has there been any saving?

* * *

Two conclusions emerge. First, in even the most conservative cases, the McNamera claim of a minimum 10% cost saving is much exaggerated. Although in some cases it may be true, given the high variance of outcomes, it hardly serves as a reasonable evaluation of incentive contracting experience. Second, in the cases where one does not assume a reduction in wastes, but one does assume some inflation of target values, the costs under the incentive contract actually rise above the costs of the cost-plus contracts. Whether the costs are, in fact, higher is difficult to say, being subject to high variance in the parameters. However, it is quite clear that the savings due to incentive contracting are not as obvious as has been typically thought.

By the end of the 1960s there appears to have been widespread agreement that multiple incentives⎯i.e., predetermined formula-type incentives applied to cost and to delivery or technical performance⎯were so complex as to be generally impractical, and they all but disappeared. Most formula-type incentives today are cost-only. Multiple incentives were replaced by the subjective, after-the-fact incentive⎯award-fee. Unfortunately, agencies did not think things through and never developed a clear set of principles for the use and management of award-fee incentives, resulting in misapplication and misuse.

Nevertheless, incentives are intuitively appealing. They appeal to common sense, and the use of incentive arrangements can convey the impression that an agency is being demanding and tough with its contractors, holding their feet to the fire by linking profit to results instead of awarding cost-reimbursement contracts. The intuitive appeal and agency "success stories" have been enough to persuade policymakers and Congress to push agencies to use incentives, even going so far at one point as to require their use in association with performance-based service contracting. FAC 97-1, 62 FR 44813 added coverage about service contracts that effectively mandated the use of incentives. Until 2005, FAR 37.602-4 said:

Contract types most likely to motivate contractors to perform at optimal levels shall be chosen (see subpart 16.1 and, for research and development contracts, see 35.006). To the maximum extent practicable, performance incentives, either positive or negative or both, shall be incorporated into the contract to encourage contractors to increase efficiency and maximize performance (see subpart 16.4). These incentives shall correspond to the specific performance standards in the quality assurance surveillance plan and shall be capable of being measured objectively. Fixed-price contracts are generally appropriate for services that can be defined objectively and for which the risk of performance is manageable (see subpart 16.1).

That rule, pressure from on high, and touting by various "experts" pushed a lot of agencies into using incentives despite a lack of know-how. Inspectors general began to report poor practices. The comments of the Department of Energy Inspector General in its report, Use Of Performance-Based Incentives At Selected Departmental Sites, DOE/IG-0510, July 2001, are typical:

The Department did not utilize performance-based incentives in a manner that would consistently result in improved contractor performance. Twelve of nineteen performance-based incentives selected for review at the Savannah River Site and the Kansas City and Oak Ridge Y-12 Plants were not clearly designed to facilitate such improvement. Some performance incentive fees were increased without a corresponding increase in performance expectations. In other cases, the "challenge" to the contractor in the form of the performance standard was lowered while the monetary incentive remained unchanged. In all cases, these actions were taken without satisfactory explanation. Further, some incentives were established after the expected outcome had been achieved.

See also NASA IG report IG-00-043, Consolidated Space Operations Contract -- Cost-Benefit Analysis and Award Fee Structure, September 2000:

[W]e found that NASA did not properly structure the award fee for the CSOC [Consolidated Space Operations Contract] to evaluate performance of the Integrated Operations Architecture (IOA). The CSOC Award Fee Plan lacks defined criteria for measuring performance, appropriate evaluation periods, and proper emphasis on cost performance. Without these provisions, NASA cannot measure contractor performance to assess the appropriate amount of award fee and provide an effective incentive for the contractor.

We can trace the origins of the new policy in FAC 2005-37 directly to the GAO's December 2005 report, DOD Has Paid Billions in Award and Incentive Fees Regardless of Acquisition Outcomes, GAO-06-66, which severely criticized DOD's use of award-fee and formula-type incentive contracts. Most readers focused on GAO's criticism of DOD's use of award-fee incentives, but the report addressed all incentives and included this rather startling assessment:

DOD's use of monetary incentives is based on the assumption that such incentives can improve contractor performance and acquisition outcomes; however, past studies have challenged the validity of this assumption. Research on incentive fees going back to the 1960s has concluded these incentive fees are not effective in controlling cost. Studies conducted by GAO, Harvard University, and the RAND Corporation, among others, have concluded that these incentives do not motivate cost efficiency, in part because profit is not the contractor's only motivation. Other considerations, such as securing future contracts with the government, can be stronger motivators than earning additional profit. More recently, research on award fees revealed that while these fees are an intuitively appealing way to improve contractor performance, they do not always operate that way in practice. Contractor respondents in one study stated that award fees motivate performance to some extent; however, the consensus was that they do not in and of themselves increase performance significantly.

That assessment didn't come as a surprise to anyone familiar with the serious literature on incentive contracts, but most contracting practitioners do not know that literature. What they know is what I call the popular literature, especially articles that have appeared in Contract Management magazine by any of the amateur experts that have sprung up from time to time to tout incentives on the basis of who-knows-what evidence. They passed on "success stories" published by various people who had awarded an incentive contract and who were determined to declare their effort a success even before the data were in.

(The classic "success story," told dozens of times, is about the Army's use of performance incentives in the 1908 contract with the Wright Brothers for the first military aircraft. But that contract was an incentive in form only, not in fact. As is typical of the tellers of "success stories," the people who told this one didn't check their facts. The Wrights were not motivated by the money and the incentive had no effect on their design. According to one of the Army officers who participated in the flight acceptance tests, Lt. (later Brigadier General) Frank Lahm, USA, the Wrights declined the opportunity to maximize their profit under the incentive. See Chandler and Lahm, How Our Army Grew Wings (ARNO Press 1979), p 159, and my 2002 article, The True Story of the Wright Brothers Contract: It's Not What You Think, which is available on the Wifcon Reading page.

Regardless of whether they are motivationally effective, formula incentives on cost can have practical utility if designed and managed properly. They provide alternatives to the extremes of firm-fixed-price and cost-plus-fixed-fee by establishing a sliding payment scale based on cost. Instead of fixing a price or a single fee for performance, the parties agree in advance on what the government will pay the contractor at any given cost outcome within a range. Such a scheme can be equitable if the parties share a common understanding of the costs of performance and negotiate a fair and reasonable pricing formula. But in order for a cost incentive to be effective in that way the parties must engage in detailed discussions, share information, and then establish a pricing formula that reflects the nature and sources of cost uncertainty. This simply is not possible in competitive procurements under FAR Part 15 given the way that agencies usually conduct them. Moreover, such discussions require a lot of knowledge and skill. Whether such an arrangement will motivate the contractor to reduce costs is another matter. Contractors are motivated by many things.

I think that award-fee incentives, if properly designed and managed, can improve communications between the government and its contractors, which might improve the chances of success when requirements cannot be fully defined at the time of contract award or when they are defined but it is not clear whether they can be produced. But the use of award-fee incentive is as much an art as anything, despite the current calls to make them less subjective. The key to success is a sophisticated fee determining official with intimate knowledge of the work under contract, the facts of performance, and the contractor, and with excellent judgment and first rate communication skills. I don't think those qualities are easy to come by.

So what now? Do we abandon incentives? Before we do that, we should think things through.

First, incentive arrangements are much more complex than most practitioners realize. The easiest way to confirm this is to read the 1969 DOD/NASA Incentive Contracting Guide. Although out of print, it has never been surpassed as a textbook on the use of the predetermined, formula-type incentive arrangements. The current coverage in the DOD Contract Pricing Reference Guides, Volume 4, Chapter 1, is superficial by comparison. Anyone who reads the DOD/NASA guide will be impressed by the complexity of incentives. Incentives are so complex that they should be used only by the most knowledgeable and skilled practitioners.

Second, there has never been a fully worked out theory or set of principles for award-fee incentives, and agency-published guidance focuses on organizational and procedural questions. No one has ever bothered to think things through. Indeed, there has been very little up-to-date thinking about award-fee incentives. The DOD/NASA guide discusses award-fee only briefly and the discussion is half-baked. Consider, too, that the sample award-fee criteria in the DOD Procedures, Guidance, and Instruction (PGI) at 216.405-2 (DFARS Table 16-1) have not been substantively changed for more than 30 years. That table reflects the ongoing confusion over the subjective versus objective and process versus results theories of award-fee.

Third, use of incentives makes contract management much more complex, especially when a contract is subject to changes during performance that will require equitable or other adjustments to price or cost and fee. A contracting officer must have considerable knowledge and skill in order to preserve incentive effectiveness in the face of significant or numerous pricing adjustments. It is not work for amateurs or even ordinary journeymen.

Fourth, official training about incentives is superficial, at best.

Fifth, even if incentives can "work" to motivate contractors, poor design and unskillful management makes it unlikely that they will.

In an article I wrote for The Nash & Cibinic Report in 2005, "Award Fee Incentives: Do They Work? Do Agencies Know How To Use Them?," 20 N&CR ? 26, I recommended that the FAR councils require HCA approval for the use of an award-fee incentive. The councils have gone me one better⎯they require HCA approval for the use of any incentive. It remains to be seen whether that limitation will be retained in the final rule. I hope so. We have been entirely too enamored of incentives, and reduced use will greatly simplify contract formation and management. HCAs should approve the use of incentives only after they have satisfied themselves that the people who want to use them know what they are doing. If they do, we'll see fewer incentives.

The FAR Council (the Administrator of Federal Procurement Policy, Secretary of Defense, Administrator of GSA, and Administrator of NASA) should now ask themselves a couple of questions. Do incentives work? Are they worth the trouble? If the answer to either of those questions is no, then perhaps the coverage in FAR 16.4 should be removed or drastically abridged. If both answers are yes or maybe, and the FAR Council doesn't want to eliminate the coverage on incentives, then the FAR Council should take the opportunity to (1) think through the concept of the award-fee incentive and (2) provide the incoming generation of contracting personnel first-rate training in the use of all types of incentives. The latter can be done by developing a rigorous DAU course devoted entirely to the topic. In developing such a course, DAU should look to the DOD/NASA Incentive Contracting Guide as a knowledge baseline and a foundation for a new textbook. On-the-job training in incentive contracting is no good, because the generation that is now headed for retirement does not understand incentives well-enough to train anyone in their design or use.

Have we seen the end of incentives? No. A better question is whether we have seen the beginning of a new era in the use of incentives.

Excellent blog Vern. If I have to ask my HCA for authority to use an award fee contract, I'd better be able to provide a draft of the Award Fee Determination Plan and be able to explain why I want to incentivize the factors I have included. From both sides of the house I've seen too many AFDPs that include items that, while easy to measure, don't necessarily tell the CO if the vendor is doing a good job in accomplishing the objectives of the contract. Asking the vendor to propose metrics as part of the RFP is fine, but too often we don't include the goals we're trying to achieve so we get measurement for measurement's sake.
  • Report
Do today's contracting officers even know how to use this contract type? I doubt it.

  • Report