Download PDF Thank you for the opportunity to address acquisition reform lessons learned from the Department of Defense and the private sector and how they may apply to the acquisition practices of the Department of Homeland Security. The Committee’s interest in acquisition reform is timely, as I believe that a greater focus on these issues not only has the potential to enhance national security, but also save billions of taxpayer dollars.
The last set of meaningful, comprehensive acquisition reforms coincided with the onset of budget austerity at the end of the Cold War. These bipartisan reforms were led by the Department of Defense and the House and Senate Defense Authorization Committees and resulted in the incorporation of commercial advances in information technology and the adoption of some of the best commercial business practices of the time. Ironically, the budgetary increases of the last decade have not been kind to that reform effort. The rapid inflow of dollars to agencies often led to acquisition practices that were not as frugal or commercially oriented as they should have been.
In the last five years, the Department of Defense and the rest of the federal government have been on an accelerated path to a return to the acquisition practices of the 1980s, which were a morass of unique government regulations and rules. This approach, which was later the subject of the 1990s acquisition reforms, was not attractive to the most creative and innovative companies at the time and did not return value to the taxpayer.
During the 1980s, information technology bought by the government was generations behind what was available in the commercial marketplace. Government-unique contractors bid on rigid government requirements and specifications that were drawn-up by federal acquisition officials whose main preoccupation seemed to be to avoid a bid protest. The result was an adversarial system where low price shootouts for mythical programs that could not be executed were the norm. Meanwhile, a parallel commercial market existed that refused to do business with the government but could solve many of these “gold plated” requirements at a fraction of the cost.
The 1990s acquisition reform initiatives and legislation focused on best value and commercial item contracting and tried to change this situation. This approach was enshrined in a memo from then Secretary of Defense William Perry in 1994, as well as the 1994 Federal Acquisition Streamlining Act and the Clinger-Cohen Act of 1996, which all made significant progress in the immediate decade after implementation. However, the federal government’s recent shift to LPTA (Low Price Technically Acceptable) contracting and the return of a rules-based compliance culture that continues to add costs is rolling back the advances made in the past 20 years of acquisition reforms.
Sequestration and the Budget Control Act offer the opportunity to refocus the government and oversight agencies on the bottom line and to implement the acquisition reform goals of the mid-1990s tailored to any new circumstances from the last two decades. Nothing focuses minds faster than having to live within a constrained budget – be it your household’s, company’s or agency’s. These kinds of acquisition reforms are absolutely necessary at DOD and DHS, as without them and a corresponding change in business practices, budget reductions could result in a significant decrement to national security. The old adage to do more with less has to become a reality and the only way to do that is to take advantage of advances in technology and change underlying ways of doing business at the national security agencies.
There are many lessons learned that can be gathered from studying DOD’s acquisition practices since the end of World War II. The key is to identify which best practices could be replicated at DHS and which so-called “best practices” are really dead ends that have added costs but no value to the taxpayer or to our national security. Based on this history, I have developed for the Committee’s consideration what I are think are the most significant two guiding principles and five recommendations.
But before I delve into those, one lesson learned from past successful acquisition reform efforts is that they need significant Congressional involvement from members and staff steeped in common business sense to gain any traction. Without some kind of Congressional interest and sanction, executive branch bureaucracies will tend to ossify the acquisition system into a one-size fits all, cookie cutter, rules-based approach that is not nimble enough to execute deals that are in the best interest of the taxpayer.
Acquisition policy is not rocket science but it is complicated. As you delve into this issue, if something doesn’t meet the common sense test it is probably an area that needs reform or at a minimum a clear justification of its existence.
With that being said, I believe there are two guiding principles Congress should use when addressing acquisition reform proposals.
Principle #1: Government unique is expensive. Exquisite solutions oftentimes have exquisite price tags. The more you have a dedicated industrial base that just serves the government – the more expensive it will be. Any rule or requirement that only impacts the government market and not the commercial market will add cost. Requirements for information in formats not used in the commercial marketplace or for data that is not normally collected by companies all have a cost and the taxpayer pays for it – whether it is an increase in the costs of goods and services provided or from the reduction in competition and innovative ideas from those firms who chose not to comply with unique government or agency requirements and exit the business.
Let me be clear, some level of unique government procurement rules and oversight are necessary, but they have to be carefully assessed to ensure that they do not drive perverse incentives in the industrial base and in the agency. Congress needs to ensure that the current acquisition laws, regulations, policies, and rules are adding value and not destroying it and meet a clear cost-benefit test.
Principle #2: Be wary of one-size fits all solutions. Whenever a problem is found with acquisition, it first has to be determined if it is a systemic problem or a localized one. The acquisition system is currently plagued with a lot of legislation, regulations, rules, and policies enacted to address a singular scandal or perceived problem that are not appropriate to all types of acquisitions. Many solutions will likely have to be tailored to specific types of acquisitions - but always keep in mind Principle #1.
There are three types of acquisition I would focus on: large governmental systems similar to DOD weapon systems, services, and information technology. Within each of these types there are several categories of potential acquisitions depending on the application, the industrial base, and whether or not there is a compelling need for speed or innovation.
For example, shipbuilding is different than buying ground vehicles. Buying desktop computers will be different than a data analysis system that incorporates information from multiple sensors and sources. Construction services are different than medical services. Buying a vehicle for immediate deployment to protect troops from daily attacks from roadside bombs is different than upgrading a truck used on domestic military bases. Each category potentially has its own best practice that might require a tailored legislative or regulatory policy approach. Legislating to address a problem with systems acquisition brought on by a shipbuilding issue in the Coast Guard with a sole-source government unique contractor may be counterproductive if applied to information technology and services acquisition with plenty of competition and commercial alternatives.
One also has to be deliberate about what you want the acquisition system to do. Right now the acquisition system is asked to be efficient, effective, transparent, competitive, fair, innovative, and accountable – all noble principles but unfortunately all meaning different things to different people. Disagreements in what these principles mean and how they should apply can lead to oftentimes-disastrous consequences for the government. The laws, regulations, and processes involved in many of these principles can lead to a trade-off between these principles, as they are not necessarily complementary.
For example, an across the board initiative to improve accountability triggered by an agency contracting scandal (which might seem like a positive thing to do) could see a drop in efficiency, effectiveness, and innovativeness in the acquisition system due to new administrative burdens placed on the system. Too many of these burdens might lead commercial contractors to leave the marketplace or establish costly separate government unique entities within their firms to comply with these new accountability measures. The first question to ask in any procurement scandal is whether existing law is working. If law enforcement has the tools to deter, identify and prosecute cases of procurement fraud as it did in the recent Army Corp of Engineers kickback case, there is probably no reason to act.
I would propose the following five recommendations to improve acquisition at DHS:
1) Professionalize the acquisition workforce. Without a professional workforce that can exercise sound business judgment, successfully executed programs in the government will be a rarity. DOD’s acquisition workforce will face significant challenges ahead; particularly as older, more experienced acquisition professionals retire. Still, from most observers I have talked to about DHS, it appears that DHS’ acquisition workforce is far behind the professionalism and experience level of the DOD acquisition corps.
So as a first step the Committee may want to consider adopting DAWIA (Defense Acquisition Workforce Improvement Act) standards for DHS. It also may want to consider a funding mechanism such as found in the DAWDF (Defense Acquisition Workforce Development Fund) to pay for training and workforce development. Still, questions have been raised about the quality and implementation of the DAWIA required education and training provided by the National Defense University. The Committee may want to look at other outside the agency training options for DHS, be they from public universities or the private sector.
As this workforce development will be a long-term project, the Committee may want to consider in the short-term centralizing the best acquisition talent in the Department to specialize in certain types of procurements and buy for the entire enterprise. Centralization, however, runs the risk of the creation of bureaucratic barriers that make it more difficult to award contracts -- so only the most difficult acquisitions or categories of acquisitions should be considered for this option.
DHS could also consider contracting out for acquisition assistance from non-conflicted firms in areas of systemic workforce weakness. The National Reconnaissance Office (NRO) since the 1960s has done something similar to access non-conflicted (Organizational Conflict of Interest or OCI-free) systems integration experience that it lacked to help it deal more effectively with its contractors.
If, after a few years, the Committee is still not satisfied with the progress in developing an adequate DHS acquisition workforce it should follow the progress of a potential British experiment. In the UK, the British Ministry of Defense (MOD) is proposing to contract out the entire acquisition function to a private non-conflicted firm. The British MOD, if it actually does this, would be embarking on a grand experiment in government acquisition and if it were successful would have significant lessons learned for the US government.
2) Don’t replicate R&D for solutions that already exist. One of the most wasteful things the government can do is use its limited research and development (R&D) funds on things that already exist. Yet, there is a tendency to do exactly that in the government where a “not-invented-here” syndrome tends to prevail. Replicating has two costs – duplication but also in opportunity costs from lost R&D that cannot be applied to other solutions.
DHS’ contract spend of $12.4 billion in 2012 is unlikely to drive many commercial markets. If DOD, which contracted for over $361 billion in goods and services in 2012 (or another agency), has already developed something that works, DHS should buy it off the shelf. The same applies to the commercial marketplace. That this doesn’t happen (even at DOD) is a problem with the requirements portion of the acquisition system. It is a difficult sell within DOD to only accept 80-90 percent of what they think they might need. Instead DOD tends to embark on 15 year development programs and invest billions of dollars that never quite meet those requirements when they could have had something deployed immediately that meets most of the users needs for a fraction of the eventual cost of the “required” system.
As a tool to guard against this, an agency needs to have conducted significant pre-program market research before it embarks on a procurement to really know what is already out there in the commercial market or has been uniquely developed and successfully deployed in the other government agencies. The agency will likely need some kind of robust requirements review process charged with disciplining unique requirements both prior to and after program initiation. Even when an agency does buy “off the shelf” there is a real danger that subsequent needs for “minor modifications” will equate to large dollars in development costs. The requirements system needs to be effectively disciplined to prevent this from happening. DOD’s experience with configuration steering boards, while still in its infancy, may be one way of trying to enforce this kind of discipline.
I will digress for a moment to discuss why leveraging other people’s money is so important. At one time the federal government and DOD dominated R&D spending. For example, according to figures compiled by the National Science Foundation, the federal government provided 67% of R&D funding in 1964 and served as the driver of innovation in the economy. Today, the private sector now provides over 60% of R&D funding and accounts for over 70% of its performance and is where innovation is concentrated.
But that is only here in the U.S. Global R&D now stands at almost $1.5 trillion a year. There has been a significant trend in the globalization of R&D in the last several decades so that now U.S. R&D is only 28% of global R&D and the U.S. government’s share is now at around 11%. And unlike in the past there are now many more avenues for solutions out there than just U.S. government-unique research and development. DHS with its limited R&D funds could try to go it alone, but a more prudent use of funds would be to only spend its R&D on something that no-one else is doing and leverage off of everyone else, first by looking at the portion in the U.S. government’s portfolio, then U.S. commercial, followed by the R&D conducted by allied governments and finally in the global commercial market.
3. Make it really hard to start a new “too big to fail” program and then enact a strong Nunn-McCurdy like system to cancel programs if they do not meet goals. By first buying systems as much as possible off the shelf, it would be hoped that there would be very little that DHS would be doing that is DHS unique in systems acquisition. It would be expected after comparing contract spending at DOD and DHS, that DHS would only have a handful of programs that are equivalent to DOD’s MDAPs or Major Defense Acquisition Programs. However, the Committee may want to focus its oversight on programs with a smaller dollar threshold than the legislated DOD MDAP threshold.
Major systems acquisition is one area where there are significant lessons to be learned from DOD. DOD has a history of great technological innovation and periods of technological evolution that involve incremental improvements to existing systems. We are currently in one of the latter periods with innovation being primarily driven by the commercial market and rapid acquisition programs outside of the traditional DOD acquisition process.
Throughout both types of periods there has been one constant – cost overruns, schedule slippages and performance issues. Ron Fox from the Harvard Business School in his aptly titled book “Defense Acquisition Reforms 1960-2009: An Elusive Goal” sets the stage on the history of acquisition reform with regards to large systems development:
“Defense acquisition reform initiatives have been Department of Defense perennials over the past fifty years… Many notable studies of defense acquisition with recommendations for changes have been published, and each has reached the same general findings with similar recommendations. However, despite the defense community’s intent to reform the acquisition process, the difficulty of the problem and the associated politics, combined with organizational dynamics that are resistant to change, have led to only minor improvements. The problems of schedule slippages, cost growth, and shortfalls in technical performance on defense acquisition programs have remained much the same throughout this period.”
Fox begins his history by referencing the first large scale acquisition reform study of the 1960’s – The Weapons Acquisition Process: An Economic Analysis, by Merton J. Peck and Frederic M. Scherer published in 1962. This study reviewed the results of weapons acquisitions of the 1950s and identified six major problems with these acquisitions: “1) schedule slippage; 2) cost growth; 3) lack of qualified government personnel; 4) high frequency of personnel turnover; 5) inadequate methods of cost estimation; and 6) insufficient training in the measurement and control of contractor performance.” Fox comments on Peck and Scherer’s conclusions: “Fifty years later, acquisition reforms continue to seek remedies to the same problems.”
GAO and others have tended to coalesce around the following solutions to cost, schedule and performance problems at DOD: the need for stable requirements; stable budgets; proven and mature technologies; and stable personnel. Many of these ideas were incorporated for DOD into law in the last couple of years to ensure that early on in a program these objectives are met in what is called a Milestone B certification (Section 2366b, Title 10 U.S. Code), which is at the end of DOD’s technology development phase of acquisition. This should be a difficult hoop to get through and large programs should not be initiated and significant funding brought to bear until there is equivalent type of certification at DHS. The Committee may also want to consider only approving programs at Milestone B that will be completed and deployed in less than three to five years. There is no reason that DHS should emulate DOD’s overly lengthy 15-20 systems acquisition time cycle.
If a program at the equivalent of Milestone B meets these criteria, there should be a better chance of the program to successfully meet its cost, schedule and performance goals. Once this certification by the senior acquisition official in the Department is made, you may want to consider adopting some type of Nunn-McCurdy reporting and oversight requirements for DHS. Because costs estimates for the program should be more realistic at Milestone B, this is when I would recommend beginning the Nunn-McCurdy baseline. Past practice often had DOD setting this baseline earlier without meeting the objectives in what is now contained in the Milestone B criteria. These premature baselines resulted in unrealistic expectations for the program and subsequent Nunn-McCurdy cost breaches. With a more realistic cost estimate established at Milestone B, if programs do exceed the “critical” cost overrun thresholds set in Nunn-McCurdy, the program should be cancelled in all but exceptional circumstances.
4. Establish an Innovation Fund that allows for the rapid deployment of operational prototypes and the maturation of technology to support systems acquisition. To get to the level of technological maturity necessary for large programs to meet their Milestone B certifications and to continue pushing the technological envelope in areas necessary to meet changing national security requirements, the Committee should look at DOD’s informal rapid acquisition system that developed to meet wartime needs over the last decade. These programs should be relatively small and focused on deploying operational capability in parts of the agency in a six-month to two-year time frame. Because of the inflexibilities usually inherent in agency budget systems, I would recommend establishing some type of flexible R&D fund that can quickly fund rapid prototyping initiatives similar to the rapid equipping initiatives in the military services. These rapid operational prototypes could be initiated by a similar requirements process developed in DOD known as the JUONS (Joint Urgent Operational Needs Statements) process.
These types of initiatives serve several purposes. The first is to get technology out into the field faster and meet user requirements in a compelling need situation. The second is to prove technology at a smaller unit level that could be potentially scalable and transferred into a major systems program. A third outcome is that short timeframe to deployment forces the agency to incorporate off the shelf technologies quickly into new types of capabilities. To get some commercial companies who might not otherwise participate in DHS acquisitions, these rapid prototypes may require the use of DHS’ other transaction authority.
5) Services and IT acquisition: Identify and adopt commercial best buying practices. DOD is a large buyer of services, on which it spends more than half of its contract dollars. It also is a large buyer of information technology (IT). Needless to say these are different kinds of procurements than weapon systems. It would be expected that most of DHS’ future contract dollars will be spent in these two areas as well.
Back in the 1990s and early 2000s, Congress asked GAO to go out and determine the best commercial practices for these types of acquisitions. This was different work for an agency more used to compliance auditing, but GAO rose to occasion and created an exceptional body of work. Much of it was then incorporated into the information management provisions of the Clinger Cohen Act of 1996 and the services acquisition management provisions in various National Defense Authorization Acts of the early 2000s. These reforms and GAO reports currently serve as the basis for “best practices” for buying IT and services in the government.
Since that time, there has not been a lot of “best practices” oversight to be found in the government. Since this work is now 10-20 years old it is probably time to task GAO (or another entity if GAO no longer has the right expertise to perform such an evaluation) to re-look at some of these best practices. It could be assumed that the private sector, some U.S. government agencies and other governments (either state or foreign) have developed new ways to better manage the purchases of IT and services. These new best practices could be used to update the Clinger-Cohen Act, Title 10 and any government-wide services acquisition legislation and regulations.
In conclusion, these are but a few ideas for the Committee to consider as it looks to reform the acquisition practices of the Department of Homeland Security. I think there are many more “best practices” out there and I commend the Committee as it first looks for lessons learned (both good and bad) to help guide its oversight efforts in this area. I look forward to any questions the Committee might have. Thank you.