This piece is pretty remarkable. It’s penned by two true Washington heavyweights — and it’s bipartisan. One author, Peter Levine, is a Democrat, long the top staffer on the Senate Armed Services Committee and then Defense undersecretary for personnel and readiness at the Pentagon. The other, Bill Greenwalt, is a Republican, long the top acquisition policy expert on the SASC and then deputy Defense undersecretary for industrial policy. And it’s a readable piece about the all-important topic of how the Pentagon should buy from the commercial sector.
Every word is worth reading to find out what the 809 Panel did right and what needs to be done to fix some of its missteps. Read on! The Editor.
The Section 809 panel offers an important new idea on how the defense acquisition system could be improved to provide the Pentagon greater access to innovative new technologies that are rapidly developed in the private sector.
But the panel fumbled its recommendations with a contradictory and incoherent legislative approach that would hurt more than it helps. That should not undermine the significance of the panel’s underlying idea: a new focus on the source of funding for commercial items instead of the marketplace in which they are sold. This could encourage companies to develop defense-unique products at their own expense. Today, many are driven away when they are forced to accept regulated pricing and burdensome, government-unique contract terms and conditions.
Successes and Failures
We had high hopes when we helped to write the commercial buying acquisition reforms in the Federal Acquisition Streamlining Act of 1994 and the Clinger-Cohen Act of 1996. These reforms broadened the definition of commercial items, established a commercial item preference, and authorized waivers to government-unique requirements and contract clauses for the purchase of commercial solutions.
As a result, the government went from 476 contracts under simplified commercial item procedures in 1996 to nearly 13 million contracts worth almost $75 billion in 2011. As our colleague Jon Etherton has explained, this enabled the U.S. government to catch up with the information technology revolution that had swept the private sector in the 1990s and likely saved the Department of Defense (DOD) billions of dollars by avoiding unnecessary research and development and the extended acquisition lead times associated with government-unique products.
Despite initial successes, the full promise of the commercial item reforms has not been achieved. Our first disappointment has been the failure by the U.S. government and DoD to waive burdensome government-unique contract clauses and legal requirements for commercial items and commercial-off-the shelf (COTS) products. Strong legislative authorities designed to eliminate these clauses went largely unused, leading their total number to almost triple, from 57 in 1995 to 165 today.
Our second disappointment has been the stagnation and even reversal of Pentagon efforts to use the commercial item preference as a means to bring commercial innovation into the department. The Inspector General’s insistence on the need for cost data on commercial contracts has been emblematic of this problem. One company was even pushed to the extreme of using the judicial process to ensure that the DoD would consider its commercial solutions.
Our greatest disappointment, however, has been the department’s inability to incentivize commercial firms to develop and tailor commercial processes and solutions to meet the military’s needs. We had hoped that commercial items legislation would empower it to seek out and encourage innovative commercial approaches to government-unique problems. Instead, it has too often thrown bureaucracy at the problem by insisting on traditional processes and pricing approaches that drive commercial companies away from the defense market.
As a result, the government often trails the private sector by a decade or more in access to commercial innovation. Some innovative companies are not willing to do business with the Department at all, while others insist on selling through separate subsidiaries and middlemen – an approach that burdens the acquisition process, undermines agility, and stifles innovation. Some of these obstacles have been overcome through the use of flexible Other Transaction Agreements (OTAs), but the overuse of OTAs could lead to abuses and legislative restrictions that limit the future availability of this critical acquisition tool.
We expected the Section 809 panel to address these problems by reviewing acquisition regulations in the same way the Section 800 panel of the 1990s reviewed acquisition legislation, making specific recommendations on requirements and contract clauses that should be waived for COTS and commercial items, and explaining on a case-by-case basis why they shouldn’t apply. The panel missed an opportunity to improve the federal acquisition system when it failed to take on this challenge.
A Failed Attempt
The Section 809 panel completed its work earlier this year, producing a three-volume report of more than 2,400 pages. Instead of reviewing existing statutes and regulations, the panel sought to re-envision broad acquisition issues, such as the structure of the defense acquisition organization, the composition of the acquisition workforce, and the workings of the department’s funding allocation system and the congressional appropriations and reprogramming processes.
One of the areas on which the panel placed the greatest emphasis—revisiting it in all three volumes of the report—was the idea of a “dynamic marketplace,” in which different sets of acquisition rules would apply to different categories of sellers. Unfortunately, the panel’s views on the dynamic marketplace fluctuated wildly over time. The panel called for repealing the special treatment of COTS items in Volume 1 of its report, only to recommend the establishment of a new category of “readily available products”—essentially the same thing — in Volume 3. In Volume 1, the panel recommended separate definitions of commercial products and commercial services (a recommendation that Congress has now followed), only to recommend eliminating these definitions entirely in Volume 3.
The panel initially proposed a marketplace centered on four separate acquisition “lanes,” as suggested by Ben Fitzgerald in his 2016 paper on the “Future Foundry.” These four lanes roughly parallel the existing system of COTS (or readily available) items, commercial items, customized commercial items, and defense-unique items, but with an openness to modifications and updates. Volumes 2 and 3 of the report changed course, proposing to collapse the marketplace into three lanes, with just a single middle lane between the extremes of readily available items and defense-unique items.
In more than 100 pages of discussion, however, the panel failed to come up with a workable definition for this middle lane. The panel defined the middle lane to include any readily-available product that could be modified to meet DoD needs “using commercial processes and equipment.” But it never explained what commercial processes and equipment are. If commercial processes and equipment mean the processes and equipment used by private sector companies, this definition could include virtually everything that DoD buys. For example, an F-16, a product that already exists and can be ordered with no customization, appears to be a “readily-available” product. A Joint Strike Fighter, which is being developed and modified with the processes and equipment of a private sector corporation, might be a readily available product “with customization.”
The panel tried to reverse ground by saying that “defense-unique” development (the third lane) includes any product, the development of which is wholly or partially financed by DoD to serve a defense-unique need. If this definition covers any DOD-financed modification of a commercial product, it would reverse existing provisions providing commercial item treatment for the purchase of products with modifications of a type customarily available in the commercial marketplace, or minor modifications made to meet government requirements.
Under this approach, the first government dollar spent to convert a Boeing 767 into a tanker aircraft would make the entire aircraft a defense-unique item. Streamlined commercial purchasing procedures would no longer be possible for “green aircraft” or military-hardened electronics. Worse still, commercial software, which almost always requires modification to meet military requirements, would be subject to the full range of DoD-specific contract terms and conditions. As a consequence, some companies would no longer sell cutting-edge technology to DOD at all, even as this technology remained available to other customers, including foreign militaries.
The panel appears to have believed that its definitions of “readily available” and “defense-unique” products were mutually exclusive, but this is not the case. If, in fact, a particular product is covered by both definitions, which set of rules is supposed to apply? Are the streamlined procedures supposed to be applied to everything or to nothing—or is there some invisible line between the two that we cannot see? The panel’s grand design falls apart because it cannot answer these basic questions.
A Major Improvement
It does not have to be this way. The Section 809 panel had an important idea: that the source of funding should be considered in determining whether or not a product is commercial. If the panel’s idea could be applied to improve the existing system, rather than wasted in a fruitless effort to reinvent acquisition from the ground up, the change would improve the government’s access to technology, rather than confusing it.
The existing definitions of COTS and commercial items have the advantage of 25 years of practice and precedent to assist in their application. They expressly address the issues of minor modifications and commercial modifications on which the panel stumbled so badly, enabling the department to purchase a broad array of commercial items under streamlined procedures. These definitions, however, are based on a narrow, buyer’s perspective of commercial items. The underlying idea is that if a product has been tested in the commercial marketplace, the government should be able to rely on the product and the price and can safely use simplified procedures and commercial pricing.
The panel’s source-of-funding approach, by contrast, views commerciality from a seller’s perspective. The underlying idea is that if a product is developed exclusively at private expense, the seller has earned the right to sell it at commercial prices without having to accept burdensome government-unique terms and conditions, regardless of whether the product is tested in the commercial marketplace or is available to other buyers. If adopted, this seller’s perspective would provide its own advantages to the government, encouraging commercial entities to invest their own funds in innovative solutions to DoD problems.
If a private company is willing to design a capable new defense-unique software product or space system for DoD at its own expense, the military is likely to benefit, even if it has to pay commercial prices. The U.S. military needs private sector innovation today more than ever—especially in advanced technology areas like artificial intelligence, communications, sensors, and advanced software, where commercial technology is evolving at a rate far faster than the DoD development cycle. At a time when the Pentagon is desperately seeking new tools to gain faster access to private sector innovation, the source-of-funding approach, with its seller’s perspective, could open a promising new acquisition pathway.
Fortunately, this is a case where DoD can have its cake and eat it too. Rather than throwing out the existing commercial item provisions and trying to address every difficult definitional issue from scratch, Congress and the department have the option of retaining the existing definitions and adding on new categories of commercial items based on source-of-funding. This combined approach would enable the Pentagon to benefit from both the certainty provided by the existing buyer-based commercial item categories and the innovation that would be encouraged by the new seller-based commercial item categories.
The new category of commercial items could, as the Section 809 panel suggested, cover any product that is developed exclusively at private expense, without the investment of a single federal dollar (including independent research and development). Such products would be purchased under streamlined commercial procedures, even if they have a defense-unique purpose and will never be tested in the commercial marketplace. If the development has already been completed, the product would be treated as a COTS item; if the product is not yet available but development will be completed exclusively at private expense, it would be treated as a commercial item. In either case, commercial item preferences would apply.
In fact, the Department might choose to go a step farther and enable even traditional government contractors to compete in this new innovative arena by offering their own products that are developed exclusively at private expense, without any government-financed development. To achieve this, DOD would have to establish a new category of funding—not independent research and development (IR&D), which is highly subsidized by the Department, but self-funded research and development (SFR&D), which would not be reimbursed.
DOD is already behind the commercial market in several of the technologies identified in the National Defense Strategy as keys to future defense applications. Due to the globalization of the commercial market, these dual-use technologies—which will be even more important as future force multipliers and differentiators—are now, and will continue to be, available to U.S. adversaries. However, without reform, new incentives, and new acquisition tools, the fruits of this commercial innovation may not be available.
Currently, DoD’s processes prevent most of the commercial sector from partnering with the government. Government-unique requirements serve as a barrier to firms that might be willing to develop products that we need at their own expense, assuming developmental risk as they do for their commercial customers. Commercial companies may be willing to sell their off-the-shelf wares to the Pentagon, but they won’t modify that technology to meet our needs, because that could trigger government-unique requirements. The Section 809 panel correctly recognized that DoD needs these companies to not only modify their items, but also use their commercial developmental processes and commercially-trained scientists and engineers to solve some of our unique defense problems.
While OTAs are lowering barriers to fill some of DoD’s innovation needs, the use of an exceptional tool like them is not a sustainable approach. Treating self-funded research and development as a commercial item for government contracting purposes has the advantage of a defensible rationale that is built into its structure. The result would be to open up commercial markets, making more companies willing to do business with DOD. This would be a win-win for the military and the U.S. industrial base: companies would be able to sell defense-unique products that are tailored and modified to meet DoD needs without burdensome regulatory requirements and the department would gain access to private sector innovation. It is an idea too good to lose sight of in a 2,400-page report.