Monday, August 3, 2009

Defense Acquisition Reform Needs Common Sense

An interesting piece on defense acquisition reform came out by Scott Reynolds, a professor of program management for the Defense Systems Management College at Defense Acquisition University on how we can come to see poor performance in the defense acquisition process as a matter of fact rather than something that is fundamentality flawed. Although stating that the defense acquisition process is in need of reform is subject to debate, there can be no doubt that the current way of doing business is unsustainable, and that serious consideration has to be given to the way DoD procures goods and services.

According to Mr. Reynolds, the first bit of common-sense approach to fixing defense acquisitions is to avoid the temptation to reform.

…In his book “Knowledge for Action,” Harvard University’s Chris Argyris explains that large organizations like DoD use policy changes as a defense mechanism to avoid embarrassment. As a result, organizations often create policies that hinder improved performance rather than creating workable solutions. A November 2005 Government Accountability Office report highlights that it is not the absence of DoD policy, but Defense’s failure to follow its own policy, that causes most program problems...

Execution is the critical part of any successful operation. That is true in business as it is in defense. The problem, in my opinion, is certainly the inability to execute current policy. This is exacerbated with the lack of a trained and skilled acquisition workforce in sufficient numbers to execute current policy. Further adding insult to injury is Congress, who continue to create misguided, albeit well-meaning legislation and policy to fix current issues that only require proper execution and leadership to carry out. It is a continuation of strategy that focuses on symptoms and not the disease. The result is a patchwork of confusing and contradicting policy, rules, and regulations (the Federal Acquisition Regulation sums up that sentiment) that inhibit risk-taking and form a culture of waste, fraud, and abuse as the acquisition workforce stumbles to complete its procurement mission with no clear roadmap to execute best-in-class procurement.

Another issue is the problem of cost overruns, which seem ever-present on any major acquisition program and further trickling down to non-major procurement actions as well. The issue here is that dishonesty and unrealistic budgetary estimates are created to allow for a higher chance of actually funding a program, instead of creating realistic estimates that normally would come from an outside entity through an independent cost estimate. Further creating a chain reaction of cost “unrealism” is that industry will further lower their bids from inaccurate funding baselines in an effort to win business. As Mr. Reynolds points out, the government needs creative presentation of finances to get the contract awarded and the program under way, and thus creates a shared lie.

…As one would expect when programs are underfunded from the start, negative progress reports are generated early on. Pressure immediately rises, and the partnership of those complicit in the lie is quickly tested. In some cases, the government conveniently forgets the shared lie and, to save the program, criticizes the industry partner for failing to deliver. The industry partner points to poorly defined or creeping requirements.

The industry-government partnerships quickly dissolve into contract language discussions instead of product delivery efforts…

One only needs to look at the FBI's failed Virtual Case File (VCF) system for a textbook example of how these issues cascade into a non-reality based procurement process. To fix these problems, the next iteration of the VCF was the Sentinel program, with the hope that the FBI would have learned from its mistakes and implemented the needed budgetary changes. However, the FBI's policies and procedures that formed the basis for Sentinel's schedule and cost estimates were not fully consistent with reliable estimating practices and the FBI also did not effectively perform key tracking and oversight practices for its many support contractors that were performing program management functions. This according to a report by the GAO, describing the failed leadership at FBI and the inability to face realism when it came to actual costs of a politically visible system. New GAO reports show progress, and Sentinel is on the right path, but much more work needs to be done to correct the mistakes of the past.

The other area of concern described is the revolving door of leadership for major weapon systems. Military officers often rotate out every 2 to 3 years, and thus create an environment of ever-increasing change, increased complexity, and loss of institutional knowledge that often leads to schedule delays and subsequent cost overruns. The rotational assignments for program managers is an issue on many fronts, and will need to be addressed by either finding a way to keep leadership on without destroying careers, or having constant civilian leadership on weapon programs with the military acting in an advisory and oversight capacity.

The bottom line is that the billions we are pouring into weapons systems that under deliver and overpromise without accountability cannot continue under the enormous fiscal pressures we face as a nation. A focus on executing goals through realistic cost estimates, stable and effective leadership, and streamlined ways of procurement with less regulation and not more would signify moving acquisition reform in the right direction.


  1. Please consider the more current findings of GAO as relates the Sentinel program and FBI practices

  2. Harlan - I chose to link and mention the first set of findings to illustrate that a year after the program was initiated, the lack of implementing lessons learned from previous failures is truly astonishing. Yearly reports will hopefully continue to show improvements as best practices area applied to this program, as the 2008 report indicates.