DTIC ADA518756: Reforming Pentagon Decisionmaking

0.59 MB·English

by  Defense Technical Information Center

REFoRmING PENTAGoN DECISIoNmAkING
Gen. Peter Pace, usMc,
addresses members of
the Joint staff
Cullen) Reforming
Myles
E D.
R Air Force ( Pentagon Decisionmaking
S.
U.
U
T
A By CHRIS TOPHER J . L AMB and IRvIN g LACHOW This article is a synopsis of INSS
Strategic Forum 221, which can be
E
downloaded at ndupress.ndu.edu
F
A
major surprise in the Depart- concluded that people using mental shortcuts delays to “reason through” a formal decision-
ment of Defense 2006 Quadren- can produce good decisions in difficult cir- making process can result in disaster
nial Defense Review (QDR) cumstances. One of the most popular nonra- n conditions are dynamic or goals are
Report to Congress is the tional theories of decisionmaking, which can ambiguous; it makes sense in such circum-
prominence it gives to decisionmaking reform. be dubbed the intuitive model, proposes that stances to focus on a quick “good enough”
Pentagon leaders thought that improvements people make decisions by recognizing situa- solution that can be reevaluated later
to the decisionmaking process prior to the tions, matching them to previous situations n they have a great deal of relevant experi-
QDR would facilitate hard choices on new they have experienced, simulating various ence; the more relevant experience a person
military capabilities.1 Yet by the end of the solutions in their heads, and then picking the has, the more likely he is to use intuition and
QDR, they concluded that additional decision- first solution that is good enough to satisfy the use it well
making reforms were one of two fundamental problem at hand.3 In this model, popularized n the problem can be modeled in
strategic imperatives for the Department.2 In in Malcolm Gladwell’s Blink, biases are not mental simulations to determine what
this article, we recommend a Decision Support deviations from an ideal approach but rather would happen if a given option were chosen
Cell to improve Pentagon decisionmaking. helpful mental adaptations that enable quick, (for example, one study found that Navy
Before explaining how the cell would work, we accurate decisions. commanders serving on Aegis cruisers use
identify prerequisites for good decisionmaking Blink stimulated a defense of rationality intuitive decisionmaking for 95 percent of
and the problems and conditions that cur- and a resultant “blink vs. think” controversy their decisions).6
rently diminish the quality of that undertak- that captured popular imagination.4 However,
ing at the Pentagon. the clear consensus among experts is that In contrast, people generally use a ratio-
people use both intuitive and rational tech- nal process when:7
Reason and Intuition niques to make good decisions. People gener-
It is commonly assumed that people ally rely on their intuition when:5 n they are not under heavy time pressure
should make decisions as rationally as possible that requires mental shortcuts; with more
and that deviations from the rational ideal are n they face time-urgent situations such as time, people are more likely to follow the
undesirable. Recently, however, scientists have firefights or battlefield triage, where even short rational approach, if only to verify an initial
gut feeling
christopher J. Lamb is a senior research Fellow in the Institute for National strategic studies at the National n conditions are relatively stable and goals
Defense university. Irving Lachow is a senior research Professor in the Information resources Management are clear, permitting a rational approach to
college at the National Defense university. find an optimal solution
 JfQ / issue 43, 4th quarter 2006 ndupress.ndu.edu
Report Documentation Page Form Approved
OMB No. 0704-0188
Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and
maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information,
including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington
VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if it

does not display a currently valid OMB control number.
1. REPORT DATE 3. DATES COVERED
2006 2. REPORT TYPE 00-00-2006 to 00-00-2006
4. TITLE AND SUBTITLE 5a. CONTRACT NUMBER
Reforming Pentagon Decisionmaking
5b. GRANT NUMBER
5c. PROGRAM ELEMENT NUMBER
6. AUTHOR(S) 5d. PROJECT NUMBER
5e. TASK NUMBER
5f. WORK UNIT NUMBER
7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION
National Defense University,Institute for National Strategic Studies,260 REPORT NUMBER
Fifth Avenue SW Bg 64 Fort Lesley J. McNair,Washington,DC,20319
9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR’S ACRONYM(S)
11. SPONSOR/MONITOR’S REPORT
NUMBER(S)
12. DISTRIBUTION/AVAILABILITY STATEMENT
Approved for public release; distribution unlimited
13. SUPPLEMENTARY NOTES
14. ABSTRACT

15. SUBJECT TERMS
16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF 18. NUMBER 19a. NAME OF
ABSTRACT OF PAGES RESPONSIBLE PERSON
a. REPORT b. ABSTRACT c. THIS PAGE Same as 4
unclassified unclassified unclassified Report (SAR)
Standard Form 298 (Rev. 8-98)
Prescribed by ANSI Std Z39-18
LLaammbb aanndd LLaacchhooww
n they do not have the relevant experi- must accommodate rational, nonrational, tify and weigh all possible risks. Second, the
ence to provide a basis for pattern matching; and intuitive decisionmaking, depending on infrequency of war means there is not a large
then they should (and usually do) resort to a which is likely to generate better outcomes.9 experiential basis for making intuitive deci-
more rational model to guide them through To determine when these different approaches sions about what investments will produce
problem formulation, option identification, are appropriate, we must understand the cir- the best mix of capabilities for warfighting;
analysis, and selection of a solution cumstances in the Pentagon that affect senior therefore, a rational as opposed to an intuitive
n the problem is so computationally leader decisionmaking. system makes sense for investment decisions.
complex that it overwhelms Third, many Pentagon
the ability to grasp a given planning problems (for
situation, at which point the consensus among experts is that people use both intuitive example, logistics or
quality of decisions erodes strategic lift) are com-
and rational techniques to make good decisions
along with the ability to putationally so complex
recognize situations or run that they defy intuitive
mental simulations. judgment alone. Fourth,
Challenges there usually is sufficient time to allow a
Despite different models, human Pentagon decisionmaking reforms since rational process to unfold.
Reforming
decisionmaking actually represents a con- World War II are largely a history of efforts to The resource planning and allocation
tinuum. In most cases, people decide through curtail the power of the Services to veto joint systems designed to support senior Pentagon
a combination of reason and nonrational solutions that serve the entire military better. leaders, therefore, are ostensibly methodical F
mental shortcuts. For example, people can use Service cultures are beneficial for warfighting, and engineered to minimize risk: objec- E
Pentagon Decisionmaking
intuitive rules of thumb to bound the range but they can be counterproductive at higher tives are defined, conditions that inform the
A
of possible solutions for a problem analyzed decisionmaking levels where integrated effort objectives are identified, alternative ways and
rationally. Similarly, they can use a rational is required. means to achieve the objectives are explored, T
thought process to augment or verify initial Secretary of Defense Robert McNama- expected and unintended consequences are
U
intuitive judgments.8 ra’s planning, programming, and budgeting considered, and decisions are made, generally
How do these observations about the system was installed in 1961 to rationalize to eliminate as much risk in as many categories R
way people decide relate to decisionmaking in Pentagon decisionmaking with broader, as possible. The same holds true for contin-
the Pentagon? First, reforming strategic deci- more transparent, and more objective deci- gency planning systems that were designed to E
sionmaking must account for the way senior sion criteria. It survives to this day, albeit in rationalize campaign planning and war plans
leaders actually make strategic-level decisions modified form, because its rational design is review.
(for example, by acknowledging the power beneficial for several reasons. First, the high Bureaucratic Contributions and Limita-
of personal and bureaucratic biases). Second, stakes involved in deterrence and war argue tions to Rational Decision Support. To execute
prescriptions for improving decisionmaking for adoption of rational processes that iden- its ostensibly rational planning processes,

Decision Support Cell Structure
Decision Support Cell
(Director, Deputy Director, and Administrative Assistant)
Secretary of
Strategic Agenda Group
Analytic Decision Support Group Exercises and Simulations Group Defense
(Leader Core Functions)
(18–25 personnel) (7–10 personnel)
(7–10 personnel)
Standing Teams Led by Standing Teams Led by Ad Hoc Teams Led by
Decision Support Cell Staff Decision Support Cell Staff Decision Support Cell Staff
Roles and Joint
Strategic Direction Joint Operating
Responsibilities Contingency Joint Data Exercises 1 Exercises 2 Exercises 3
Concepts
Scenarios
Internal and External
Macro Investments
Relations
Joint
Joint Methods Institutional
Operational Exercises 4 Exercises 5 Exercises N
Contingency Oversight Performance Review of Analysis Knowledge
Metrics
Multi- Defense
Global Force Defense Lessons
Service Force Joint Data Technical
Management Modeling and Learned
Deployments Support Information
Board Simulation Organizations
Conferences Centers
Examples of Institutions Receiving Guidance From Decision Support Cell–Led Teams
ndupress.ndu.edu  issue 43, 4th quarter 2006 / JfQ 6
REFoRmING PENTAGoN DECISIoNmAkING
the Pentagon is divided into hierarchical much information away from me as possible, the net effect of their decisions across multiple
organizational structures that represent rela- [thinking] let’s just give him what we want objectives, but they ought to take advantage of
tively narrow bodies of expertise: policy, intel- him to have, not what he needs.”11 decision support that can better inform their
ligence, program analysis, acquisition, or bud- Secretary Donald Rumsfeld made it clear intuition. In practice, this means two elements
geting. Recently, Pentagon wits have taken to early on that he would challenge these bureau- are required for strategic decisionmaking in
calling these stovepipe organizations “cylinders cratic tendencies. In a speech on September the Pentagon: clear, transparent, and well-
of excellence,” which they in fact are. Their 10, 2001, he called the Pentagon bureaucracy coordinated rational analyses of alternatives
purpose is to build and nurture deep expertise the enemy, arguing that it “disrupts the defense from the decision support system, and well-
in narrow bodies of knowledge. These experts of the United States and places the lives of men honed personal intuition and judgment. These
identify issues, devise options and recom- and women in uniform at risk.”12 The next day, elements can best be harmonized through the
mendations, and forward them up the chain to however, he had to go to war with the bureau- creation of a Decision Support Cell.
senior officials. In this regard, Pentagon deci- cracy he had, not the one he wanted. There-
sion support is essentially “bottom-up” as well fore, like his predecessors, Secretary Rumsfeld Reform’s Critical Element
as “stovepiped.” uses workarounds to tame the bureaucracy. The Decision Support Cell would be a
These bottom-up rational decision pro- For example, he relies heavily on a few trusted dedicated staff located within the Secretary’s
cesses are limited by multiple bureaucratic and aides who are able to offer alternatives to office with a mission to enforce a degree of
human factors. Senior leaders need integrated the bland or contradictory decision support discipline and collaboration in strategic deci-
problem assessments and solution options, provided by the bureaucracy. Unfortunately, sion support for the Secretary and his closest
E but there are few incentives for their subordi- that practice helps convince subordinates that advisors. It should do three things. First, it
R nates to collaborate to provide them. Instead, having access to senior leaders and control- should help the Secretary focus the decision
subordinates are rewarded for developing and ling information to them is the key to success, support process on his own strategic agenda,
U protecting their own organizational equities. which further discourages information-sharing making sure that he receives integrated
Absent any incentive to sacrifice organiza- and collaboration. The lack of senior leader products in support of this agenda and that
T
tional equities for the common good, the feedback to subordinates compounds the the process provides necessary feedback and
A natural outcome of formal direction. Second, it should
E coordination in the Penta- the Decision Support Cell would be a dedicated staff improve the quality of the
gon is consensus products decision support routinely pro-
located within the Secretary’s office with a mission to
F that avoid and obscure the vided by the contingency plan-
enforce discipline and collaboration
need for tough tradeoffs. As ning and resource allocation
a result, many talented and systems, making sure underly-
motivated officials get their positions directly problem. If subordinates do not understand ing assumptions are clear and that all viable
to senior decisionmakers by circumventing the senior leaders’ decisions, they may conclude alternatives are rigorously examined. Third, it
formal coordination process. Proposals pre- that the wrong choice was made for the wrong should help senior leaders refine their intuitive
sented this way often are clear and creative but reasons, further deepening cynicism. decisionmaking with exercises that enlarge
reflect a perspective that does not benefit from Decision Support: Balancing Rationality their experience base.
access to all relevant information. and Intuition. Even if the rational planning Strategic Decisionmaking Focus. The
What Senior Leaders Need. Secretar- and resource allocation processes of the QDR Report underscored the importance
ies and Deputy Secretaries of Defense need Pentagon worked better, they would be insuf- of senior leader focus on a set of core func-
integrated decision support from the 30 or ficient for producing good strategic decisions. tions that only they can perform effectively,13
more subordinate bureaucracies that report Senior leaders must account for a broader but the lack of integrated, quality decision
directly to them, but they do not receive range of factors than those found in analyses support for strategic issues makes that dif-
this support, and they do not have time to conducted by lower level officials. Sometimes ficult. The Decision Support Cell should
produce it themselves. These leaders are the the significance of these factors is so great be charged with ensuring the collaboration
first real point of integration in the Depart- that they dwarf the marginal utility of rational among Pentagon bureaucracies necessary
ment of Defense, and they do not like it. analyses. Even when the results of the rational to put core senior leader issues in a strategic
Former Secretaries and Deputy Secretaries analyses offer valuable insights, senior leaders choice framework. In doing so, the cell
are virtually unanimous in their belief that ultimately must contrast choices across diverse would not usurp the functions of other staff
the Pentagon bureaucracy could be cut from value sets (operational, political, economic, elements but rather undertake integrating
25 to 75 percent without any degradation and so forth). It is difficult to compare ratio- activities that currently are either left to the
in the quality of decision support.10 The nally the value of better relations with a key Secretary or are not done at all.
Office of the Secretary of Defense is often ally, less friction with a powerful Senator, With a Decision Support Cell to
singled out as bloated and ineffective, but and more economical shipbuilding. Doing so coordinate decisionmaking in senior leader
the Joint Staff also hoards information and requires reliance on intuition, judgment, and core functions, the Secretary’s personal staff
defaults toward least common denomina- other nonrational factors. would be free to support his daily schedule
tor products. Colin Powell remarked that This is not to say that there is no role for and personal needs. Similarly, subordinates
while he was Chairman, the “sole purpose” rational decision support. Senior leaders must could concentrate on their areas of exper-
of his 1,500-member staff “was to keep as rely in part on their intuitive understanding of tise, knowing that the cell would ensure
70 JfQ / issue 43, 4th quarter 2006 ndupress.ndu.edu
Lamb and Lachow
collaboration when the Secretary needed it. experience in each these areas before taking level with a Decision Support Cell that bal-
Since the cell would have a holistic view of office, it is not realistic. ances objective analysis and intuitive wisdom.
the senior leader core functions, it would be Senior leaders can gain experience Those who fight the Nation’s battles deserve
in a position to advise the Secretary on the on the job, but that is time-consuming and nothing less. JfQ
importance of keeping abreast of these areas. inefficient and sometimes means learning by
It would also be in a position to identify spe- making mistakes. In fact, mistakes offer one
cific problems that require him to set priori- of the best ways to learn, but given the stakes notES
ties among competing interests. associated with strategic decisionmaking
Improving Rational Decisionmaking in the Pentagon, it is too costly a method to 1 Ryan Henry, “Defense Transformation and
the 2005 Quadrennial Defense Review,” Parameters
Support. Comparing and evaluating alterna- accept readily. A better approach would be to
(Winter 2005–2006), 5–15.
tives is impossible without a transparent set develop the senior leaders’ experience base
2 Department of Defense, 2006 Quadrennial
of baseline assumptions, operating concepts, with a tailored program that helps them to:15
Defense Review Report (Washington, DC: Depart-
methods, metrics, and data. Without these
ment of Defense, February 6, 2006), 65–66.
common, essential precursors to good analy- n identify and understand the decision
3 The term intuitive model refers to a class of
sis, with results that are comparable and repli- requirements of their job “naturalistic” models exemplified by Gary Klein’s
cable, senior leaders cannot usefully evaluate n practice difficult decisions in context Recognition-Primed Model. See Klein, Sources of
alternatives and their consequences. Currently, n review decisionmaking experiences to Power: How People Make Decisions (Cambridge:
no single organization has the interest, author- learn what works and what does not. MIT Press, 1998).
ity, and resources to produce such timely, 4 Malcolm Gladwell, Blink: The Power of Think- F
quality products. As a result, the foundational The best way to accomplish this goal ing without Thinking (New York: Little, Brown, E
2005); and Michel Legault, Think! Why Crucial
products for good decision support are pro- is with decisionmaking exercises or thought
Decisions Can’t Be Made in the Blink of an Eye (New A
vided too slowly and with insufficient quality experiments that are built on well-defined
York: Threshold Editions, 2006).
and quantity to support a common analytical scenarios and capture the essence of specific T
5 Klein, 95–96.
framework across the Department.14 decisions. These exercises could be conducted
6 Ibid., 97. U
The Secretary would need to empower as tabletop or virtual games or both. Decision-
7 Ibid.
the Decision Support Cell to set standards making exercises should not be confused with 8 David C. Gompert, Irving Lachow, and Justin R
and timelines for these analytic precursors large-scale games or field simulations; each Perkins, Battle-Wise: Seeking Time-Information
and to enforce a degree of transparency, col- experiment would be a focused event targeted Superiority in Networked Warfare (Washington, DC: E
laboration, and information-sharing among at the characteristics of a unique decision.16 National Defense University Press, 2006).
all the Pentagon headquarters elements that The Decision Support Cell should also 9 The Services clearly understand this point.
conduct analysis in support of senior deci- help record the results of real-world intui- Rational and intuitive decisionmaking processes,
and the situations in which each should be used, are
sionmakers. Exercising this kind of authority tive decisionmaking. Even though intuitive
discussed explicitly in Navy and Marine Corps doc-
underscores why the cell must be indepen- decisionmaking is somewhat idiosyncratic and
trine. For example, see Naval Doctrine Publication
dent of any Pentagon component and report often politically sensitive, the cell must capture
6: Command and Control.
directly to the Secretary. If it reported to senior leader concerns and desires solidly
10 Exit interviews with departing Secretaries of
someone lower, it might be unable to enforce enough to help middle management under-
Defense by the Pentagon Office of the Historian and
the necessary collaboration and competition stand the factors that informed a particular informal comments from participants in defense
of ideas needed to support senior leader deci- decision, which will increase trust in the system reform studies who interviewed former senior
sionmaking. If the cell belonged to an orga- and improve the quality of decision support. leaders.
nization charged with conducting analysis Creating the Decision Support Cell is 11 “Follow the Leader,” Context Magazine,
on specific problems or conducted analysis consistent with the 2006 QDR recommenda- interview with Colin S. Powell, available at <www.
itself, it would be predisposed to defend those tions for institutional reform. If that seems contextmag.com/setFrameRedirect.asp?src=/
archives/200002/Feature0FollowtheLeader.asp>.
analyses, immediately ruining its reputation like a tall order, we should remember that
12 “DOD Acquisition and Logistics Excellence
as an honest broker. the tactical military already achieved a com-
Week Kickoff—Bureaucracy to Battlefield,” remarks
Improving Intuitive Decisionmaking parable transformation in decision support.
as delivered by Secretary of Defense Donald H.
Support. The Decision Support Cell must also Following the Vietnam War, the Services
Rumsfeld, the Pentagon, September 10, 2001.
be able to support senior intuitive decision- introduced objective, empirical feedback into
13 2006 QDR Report, 65–66.
making by providing leaders with the breadth training exercises with the aid of new simula- 14 Christopher J. Lamb, Transforming Defense
and depth of experience needed in their jobs. tion technologies and after-action reports (Washington, DC: National Defense University
The list of diverse areas where meaning- to improve learning and decisionmaking. Press, September 2005), 4–17.
ful experience would be desirable includes The training revolution of the 1970s was not 15 Gary Klein, The Power of Intuition: How to
military operations, executive management, an easy transformation, but it was highly Use Your Gut Feelings to Make Better Decisions at
bureaucratic processes, political savvy, govern- effective because it combined the value of Work (New York: Currency Books, 2004), 36–63.
16 Ibid., 45.
ment budgeting, media relations, intelligence objective analysis of courses of action with
products and operations, and emerging the ultimate need for commanders to make
technologies. While it would be ideal for all intuitive assessments and decisions. The Pen-
senior leaders to possess a depth of real-life tagon could do the same thing at the strategic
ndupress.ndu.edu  issue 43, 4th quarter 2006 / JfQ 1