Competition and Innovation Under Complexity

7/25/2019 Competition and Innovation Under Complexity

1/21

This product is part of the RAND Corporation reprint series.

RAND reprints present previously published journal articles,

book chapters, and reports with the permission of the publisher.

RAND reprints have been formally reviewed in accordance with

the publishers editorial policy, and are compliant with RANDs

rigorous quality assurance standards for quality and objectivity.

Visit RAND atwww.rand.org

Explore RAND National Security Research Division

View document details

For More Information

This PDF document was made available

fromwww.rand.orgas a public service of

the RAND Corporation.

6Jump down to document

THE ARTS

CHILD POLICY

CIVIL JUSTICE

EDUCATION

ENERGY AND ENVIRONMENT

HEALTH AND HEALTH CARE

INTERNATIONAL AFFAIRS

NATIONAL SECURITY

POPULATION AND AGING

PUBLIC SAFETY

SCIENCE AND TECHNOLOGY

SUBSTANCE ABUSE

TERRORISM ANDHOMELAND SECURITY

TRANSPORTATION ANDINFRASTRUCTURE

WORKFORCE AND WORKPLACE

The RAND Corporation is a nonprofit

research organization providing

objective analysis and effective

solutions that address the challenges

facing the public and private sectors

around the world.

Browse Books & Publications

Make a charitable contribution

Support RAND
http://www.rand.org/pdfrd/http://www.rand.org/pdfrd/nsrd/http://www.rand.org/pdfrd/nsrd/http://www.rand.org/pdfrd/pubs/reprints/RP1386/http://www.rand.org/pdfrd/http://0.0.0.0/http://0.0.0.0/http://www.rand.org/pdfrd/research_areas/arts/http://www.rand.org/pdfrd/research_areas/children/http://www.rand.org/pdfrd/research_areas/civil_justice/http://www.rand.org/pdfrd/research_areas/education/http://www.rand.org/pdfrd/research_areas/energy_environment/http://www.rand.org/pdfrd/research_areas/health/http://www.rand.org/pdfrd/research_areas/international_affairs/http://www.rand.org/pdfrd/research_areas/national_security/http://www.rand.org/pdfrd/research_areas/population/http://www.rand.org/pdfrd/research_areas/public_safety/http://www.rand.org/pdfrd/research_areas/science_technology/http://www.rand.org/pdfrd/research_areas/substance_abuse/http://www.rand.org/pdfrd/research_areas/terrorism/http://www.rand.org/pdfrd/research_areas/terrorism/http://www.rand.org/pdfrd/research_areas/infrastructure/http://www.rand.org/pdfrd/research_areas/infrastructure/http://www.rand.org/pdfrd/research_areas/workforce/http://www.rand.org/pdfrd/pubs/online/http://www.rand.org/pdfrd/giving/contribute.htmlhttp://www.rand.org/pdfrd/nsrd/http://www.rand.org/pdfrd/giving/contribute.htmlhttp://www.rand.org/pdfrd/pubs/online/http://www.rand.org/pdfrd/research_areas/workforce/http://www.rand.org/pdfrd/research_areas/infrastructure/http://www.rand.org/pdfrd/research_areas/infrastructure/http://www.rand.org/pdfrd/research_areas/terrorism/http://www.rand.org/pdfrd/research_areas/terrorism/http://www.rand.org/pdfrd/research_areas/substance_abuse/http://www.rand.org/pdfrd/research_areas/science_technology/http://www.rand.org/pdfrd/research_areas/public_safety/http://www.rand.org/pdfrd/research_areas/population/http://www.rand.org/pdfrd/research_areas/national_security/http://www.rand.org/pdfrd/research_areas/international_affairs/http://www.rand.org/pdfrd/research_areas/health/http://www.rand.org/pdfrd/research_areas/energy_environment/http://www.rand.org/pdfrd/research_areas/education/http://www.rand.org/pdfrd/research_areas/civil_justice/http://www.rand.org/pdfrd/research_areas/children/http://www.rand.org/pdfrd/research_areas/arts/http://0.0.0.0/http://www.rand.org/pdfrd/http://www.rand.org/pdfrd/pubs/reprints/RP1386/http://www.rand.org/pdfrd/nsrd/http://www.rand.org/pdfrd/


2/21

Report Documentation PageForm 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 t o comply with a collection of information if it

does not display a currently valid OMB control number.

1. REPORT DATE

2009

2. REPORT TYPE

final

3. DATES COVERED

00-00-2009 to 00-00-2009

4. TITLE AND SUBTITLE

Competition and innovation under complexity

5a. CONTRACT NUMBER

5b. GRANT NUMBER

5c. PROGRAM ELEMENT NUMBER

6. AUTHOR(S)

Jeffrey Drezner

5d. PROJECT NUMBER

5e. TASK NUMBER

5f. WORK UNIT NUMBER

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES)

RAND Corporation,1776 Main Street,Santa Monica,CA,90407

8. PERFORMING ORGANIZATION

REPORT NUMBER

RP-13869. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITORS ACRONYM(S)

11. SPONSOR/MONITORS REPORT

NUMBER(S)

12. DISTRIBUTION/AVAILABILITY STATEMENT

Approved for public release; distribution unlimited

13. SUPPLEMENTARY NOTES

Online access http://www.rand.org/pubs/reprints/RP1386/

14. ABSTRACT

Major defense acquisition programs have become more complex across a number of dimensions, including

technology, organization, and environment. This paper explores how that increased complexity affectscompetition and innovation in the context of defense acquisition. Complexity is one of many factors that

affect the use of competition and innovation. It has contributed to changes in the nature of systems DoD

buys, changes in defense industry structure, how competition is applied at the program level, and the

drivers of innovation. Acquisition officials should consider these impacts when applying competition in an

increasingly complex acquisition environment, and their implications for innovation.

15. SUBJECT TERMS

16. SECURITY CLASSIFICATION OF: 17. LIMITATION OFABSTRACT

Same as

Report (SAR)

18. NUMBER

OF PAGES

19

19a. NAME OF

RESPONSIBLE PERSONa. REPORT

unclassified

b. ABSTRACT

unclassified

c. THIS PAGE

unclassified

Standard Form 298 (Rev. 8-98)Prescribed by ANSI Std Z39-18


3/21

3 1

C O M P E T I T I O N A N D I N N O VAT I O N U N D E R

C O M P L E X I T Y

J E F F R E Y A . D R E Z N E R

Te products o the Department o Deense (DOD) acquisition pro-

cess are perceived as becoming increasingly complex, emphasizing

multiunction and multimission system congurations. Such weapon

systems utilize network capabilities and systems-o-systems engineer-

ing and integration methodologies throughout their lie cycles. Temanagement and oversight o these complex programs have similarly

become more complex. Changes may be needed in the organizations

and procedures used to manage the development, production, and

sustainment o these complex weapon systems.

Tis chapter discusses how complexity may affect the conditions

under which competition and innovation yield the desired benets.1

Competition and innovation are not ends in themselves, but rather

are a means to attain certain benets in the context o weapon systemdesign, development, production, and support. What are those bene-

ts? What are the conditions under which competition and innovation

yield the desired benets? Have those conditions changed in ways that

affect either the role o competition and innovation in deense pro-

grams or the benets derived rom that application?

Te ollowing discussion denes what is meant by complexity in

the context o weapon system acquisition. It next describes the tradi-

tional view o competition and innovation in the acquisition environ-ment prior to and through the 1990s. Given the changes commonly

associated with complexity as dened here, the discussion then ex-

amines the implications or competition and innovation and ends by

4


4/21

3 2 Organizing or a Complex World

identiying implications or acquisition policy. Tis chapter draws sub-

stantially on past published work by RAND and others as well as un-

published work at RAND; useul reerences are listed at the chapters

end.1

D E F I N I N G C O M P L E X I T Y I N

D E F E N S E A C Q U I S I T I O N P R O G R A M S

Beore we can useully discuss the implications o complexity or the

use o competition and innovation in weapon system design, develop-

ment, production, and support, we must rst establish a working deni-

tion o complexity. In the context o DOD weapon systems, complexity

can be thought o in three overarching dimensionstechnical, orga-nizational, and environmental. echnicalcomplexity includes weaponsystem unctionality and capability, including that related to the use

o embedded inormation technology. Organizationalcomplexity ad-dresses the structures and interactions o the government and industry

organizations responsible or system design, development, produc-

tion, and support. Environmentalcomplexity includes the political andeconomic context o the acquisition process, the threat environment,

and the operational environment (how the systems are intended tobe used). We expand on these three dimensions o complexity in the

paragraphs that ollow.

Weapon systems have become more complex over time. Tis is

something o a truism and applies to the historical evolution o pro-

grams, not just to the more recent programs that have caught our at-

tention. In general, new programs appear to be more complex than

their immediate predecessors in terms o technology, unctionality,

and, perhaps to a lesser extent, their operational concept. Historically,this is the result o a natural evolution in which weapon designers and

military users continually strive to improve and enhance warghting

capabilities. Under certain conditions, the use o competition stimu-

lates innovation in weapon systems. Such an evolutionary pattern o

improvement, whether derived rom demand-pull or technology-

push, applies equally to the commercial sector as well. It is the relative

increase in complexity rom one generation to the next that is o spe-

cial interest. I each evolutionary step is relatively small, then manage-ment and oversight processes and practices will have time to adapt in

parallel, and the required degree o adoption will be small. However, i

the evolutionary step is large, there may be a signicant mismatch be-


5/21

Competition and Innovation under Complexity 3 3

tween the complexity o the acquisition program and the institutional

capacity to manage that program effectively.

aken together, the three dimensions o complexitytechnical, or-

ganizational, and environmentalsuggest that we have entered an erain which the relative increase in complexity rom the previous genera-

tion is airly large.

Te relative complexity o the weapon system itsel is captured in

technical complexity. Elements o technical complexity include the use

o electronics, inormation technology, and sofware to provide criti-

cal unctionality and capability beyond more traditional means. Tat

these are increasing can be measured by the percent o acquisition pro-

gram unds devoted to these technologies. Tese technologies residein sensors, data processing, automation, communication, and data ex-

change. Many recent weapon systems are multiaceted, multiunction,

multimission systems that include many more specic unctions and

perormance capabilities than predecessor programs. Some programs,

such as the rst generation o semiautonomous unmanned air vehicles

(UAVs) have no strong precedent and introduce entire new sets o ca-

pabilities.2Many recent programs also include the notion o systems

o systems (SOS) in which many distinct systems are linked togetherthrough a common data network. In an SOS, each weapon system pro-

vides unctionality by itsel, but when linked together, the entire SOS

provides capability that no single component system, nor all o those

systems operating independently, could. Te technical challenges in

such complex systems emphasize systems engineering, sofware engi-

neering, and system integration to a much higher degree than in the

past. Te Joint Strike Fighter (JSF, F-35), the Future Combat System

(FCS), and DDG-1000 Zumwalt Class destroyers are ofen cited as ex-amples o complex systems. Such programs also tend to be airly large

(as measured by total program cost), which also makes them politi-

cally visible, adding an organizational dimension to complexity.

In a recent analysis, Robert A. Dietrick concluded that the com-

plexity o weapon systems has been increasing over time.3He denes

complexity in terms o the number o interactions among subsystems

and the degree o integration o those subsystems, as well as the degree

o integration at the component and part levelall aspects o tech-nical complexity. Dietrick provides examples in aircraf avionics, air-

borne sensors, and computer processors; his denition o complexity

is similar to what we mean by technical complexity. Further, Dietrick


6/21


suggests that increased complexityreally increased unctionality and

capabilityadversely affects program cost, schedule, and perormance

outcomes, though it is only one such actor.4

It is not just the weapon system itsel that is complex, however.Te second dimension o complexity concerns the organizations re-

sponsible or program management and program execution. Complex

weapon system programs are managed by increasingly complex orga-nizations. Te relative increase in capabilities designed into modernsystems requires increased breadth and depth o the government and

industry workorce. Te relatively large size (cost) o these programs

adds an increased political dimension to program management. Large

government program offices are staffed by a mix o military, civilian,and support contractors perorming the ull range o unctions across

a programs liecycle. Tere are generally high levels o teaming among

the industry components (at the prime contractor level and at lower

tiers) because no single rm possesses the resources, capabilities, and

political diversity required to ully execute the program itsel. Govern-

ment has increasingly relied on industry or both programmatic and

technical capabilities, including program management, industrial base

management, requirements ormulation, systems engineering, andsystem integration. Officials o at least three programsDD(X) (now

DDG-1000), Deepwater, and FCShave publicly stated that one rea-

son they relied on industry or such important program management

unctions was due to a concern that the capabilities required to man-

age these complex systems did not exist in-house.

One consequence o complexity is the very large cost o complex

systems. JSF, i it ollows current plan, will be the largest deense acqui-

sition program ever executed, and FCS and the DDG-1000 ZumwaltClass Destroyer are in the same league. Expensive programs are politi-

cally visible and thereore vulnerable, which causes them to be man-

aged with this in mind.

Te lower industrial base tiers have become increasingly important

as a source o innovation required to achieve program technical and

system perormance objectives. DOD policy, and economic policy

more broadly, has ofen asserted that smaller rms are ofen more in-

novative. Mark Lorell has observed that it was ofen (though not exclu-sively) a smaller or second-tier rm that developed a key technological

innovation leading to the next stage in the evolution o the U.S. combat

aircraf industry.5Continued support o the Small Business Innovative


7/21


Research (SBIR) grant program also seems to support the notion that

smaller rms located in the lower deense industry tiers are an impor-

tant source o innovation. In most major deense acquisition programs,

however, government-managed competition only occurs at the primecontractor level. Although the prime contractors might hold competi-

tions among lower-tier rms or specic capability, the government

may have little insight into these lower-level competitions, and little di-

rect knowledge o the industrial base beyond the key second-tier rms

involved in a program. Tus, the DOD has little inormation, and little

ability to inuence, competition in a portion o the market that may be

an important source o innovation. As the top-tier rms ocus more on

system engineering and system integration unctions, the lower tiersbecome an important source o technological innovation that is not

being actively managed by DOD.

Finally, the complexity o the acquisition environmenthas increased.Te threat environment is both broader and less predictable than in

the past, resulting in increased complexity in terms o orce and ca-

pability planning. Te operational concepts o some complex systems

are themselves complex in order to ully take advantage o new net-

centric capabilities (e.g., FCS). Nontraditional or asymmetric warare(e.g., counterinsurgency) introduces additional operational complex-

ity. Te complexity o the government and industry organizations and

the rules governing themstatute, regulation, policy, processeshas

also increased markedly.

Tese three dimensions o complexitytechnical, organizational,

and environmentalcan be expected to affect the use o, and benets

rom, competition in weapon system programs, including the resulting

innovation attained though competition.However, other actors affect competition and innovation that are

not necessarily related to complexity, such as:

Signicant consolidation throughout the deense industry at allthe tiers, but especially at the prime contractor level;

Fewer and less requent new program starts; and

Large programs (e.g., JSF, FCS) that in the past would each have been multiple independent programs.

Tese trends and their implications need to be considered in any

assessment o the affect o complexity on competition and innovation.


8/21


As we argue below, these noncomplexity trends may in act dominate

any effects on competition, while increased complexity has opened

new areas to competition and innovation.

T R A D I T I O N A L V I E W S O F C O M P E T I T I O N

A N D I N N O V AT I O N

Competition and innovation are not ends in themselves, but rather are

means to achieve certain goals.

Competition has long been a oundation o acquisition policy and

contract awards or research and development (R&D), production,

and services. In act, there is a very strong bias in acquisition policy

and ederal regulation toward the use o competition, most recentlyillustrated by a policy directive rom the under secretary o deense

or acquisition, technology, and logistics.6In the deense acquisition

context, we expect competition to provide lower prices, higher-quality

products, cost control, improved efficiency, and innovation. In this

sense, competition is sometimes thought o as a primary driver o in-

novation, though innovation may have other sources as well.

Te conditions under which competition yields these benets in-

clude the ollowing:

A large viable industry base, such that more than two rms orteams (with different rms) bid on a project. Viability includes

both nancial strength and a healthy and capable workorce.

Some degree o industry or product sector maturity. I only theinitial innovator plays, there is no competition.

Product substitutability, which means that products are unc- tionally similar across different rms.

Many programs (i.e., requent new starts) and a stable or grow-ing budget. Tis condition is equivalent to a stable or growing

demand unction.

Minimal barriers to entry. Such barriers might include capitalequipment requirements or investment levels, workorce knowl-

edge and skills, and even amiliarity with government and DODcontracting and budgeting statutes and regulations, as these will

affect program execution.


9/21


Tese conditions are part o the microeconomic model generally

taught in undergraduate introductory economics classes. In particular,

the plausibility o the invisible hand o a competitive market produc-

ing desirable outcomes depends on these and other conditions (e.g.,ree and ull inormation). Te lack o these conditions in particular

deense sectors may prevent the expected benets o competition rom

being realized.

It is important to note that an industry sector with only two rms

and a government policy (implicit or explicit) to maintain the viabil-

ity o both rms does not provide competition at the top tier (prime

contractors). Although competitions can be held between teams led by

these two different rms, each team knows at the outset that even i itloses, it will still receive a large enough portion o the program, or oth-

er programs, to remain viable. Te industry base or large Navy surace

combatants (Bath Iron Works and Northrop Grumman Ship Systems)

and Navy submarines (Northrop Grumman Newport News and Elec-

tric Boat) are good examples o this challenge.7Both the DDG-1000

and Virginia Class submarine programs have made preservation o the

supporting industry base explicit goals o their acquisition strategies.

As a result, the use o competition in these programs, and the benetsexpected rom competition, differ somewhat rom the traditional.

Competition is thought o as a primary driver o innovation. How-

ever, competition is not sufficient in itsel to generate innovation. In-

novation depends on other actors as well, including unding levels;

the existence o a core or critical mass o talent, capabilities, and

resources in the same place at the same time (to include virtual co-

location and other advanced collaborative tools in some cases); and a

regulatory and institutional environment that encourages intelligentrisk taking and out-o-the-box thinking.

Innovation is expected to result in new warghting capabilities

based on new concepts or technologies. Innovation is valued to the

extent that it creates a warghting competitive advantage between the

United States and its adversaries. Innovation is also expected to be a

primary source o a rms competitiveness (thus coming ull circle in

this discussion). Beyond innovation o weapon systems or their use,

innovation is also expected to result in improved business, design, de-velopment, production, and support processes (generally, increased

efficiency).


10/21


Innovation arises rom R&D investment, creativity, expertise, and

sensing market trends. echnology-push and demand-pull both play

roles in deense innovation. Tere are several rameworks that allow

one to organize and think about the relationships between the actorsaffecting innovation. One such ramework includes personnel capa-

bilities and management, program management more generally (ex-

ible vs. rules based), organization (institutional structure), technology,

and workorce education and experience. Tese are not trivial actors:

Deense Advanced Research Projects Agency (DARPA) was set up

specically to enhance innovation in deense-related technologies and

concepts. With its highly educated workorce, exible management,

and relatively loose organizational structure, DARPA encourages out-o-the-box thinking. Its rules are set up to enable testing new concepts

and technologies as quickly and inexpensively as possible. And DARPA

has had many notable successes.8

Paul Bracken extends the work o two prior studies o innovation

to develop a ramework or model o innovation specic to the deense

industry.9Six sets o actors are identied:

National actors, which include education level, strength in science and technology, and supporting inrastructure (e.g., com-munication, transportation).

R&D investment in a wide variety o projects, technologies, andsectors.

Status and attractiveness o the sector (i.e., excitement and dyna-mism) as indicated by the degree to which industry in that sector

is admired by consumers and students, the degree to which it ispushing the state o the art, and its ability to attract and retain

top people.

Competition in the sector, as determined by company strategies,industry structure, and rivalry.

Demand conditionsin other words, the customer demandingcapabilities requiring innovative new technologies.

Related supporting industries including lower tiers and scienceand technology (S&) base.

Note that competition is present in this model as a actor directly

affecting innovation. Te characteristics o the competition are impor-


11/21


tant under this rameworkthat is, competition or ideas rather than

cost or market share as the key driver o innovation in technology and

product capability.

Additional actors affecting innovation or the conditions that acili-tate innovation not explicitly identied in the models above include

the ollowing:

An institutional and regulatory environment that encouragesnew concepts;

Early adopters who are willing to buy and use initial versions othe innovation;

A potential or signicant demand or the product;

High potential payoff; and

Minimal barriers to entry.

A supportive institutional and regulatory environment is a critical

oundation or innovation. An institutional structure that continually

reinorces the status quo will hinder the ability o new concepts to be

developed and tested. Feedback rom early adopters is needed to helprene the product, demonstrate utility, and transition the innovation

rom the lab to a user community. In the past, the government has

ofen been that earlier adopter. A large demand unction establishes

a potential market able to sustain enough sales to make the initial in-

vestment worthwhile. Since that investment entails risk, there must be

a perception o a payoff commensurate with perceived risk, whether

in terms o system perormance, prot, or market share. Barriers to

entry must be low enough to avoid seriously hindering the investmentrequired or rms to establish a new market niche.

Industry sectors that are highly regulated tend to be relatively poor

innovators. Increased ormal rules and processes or large rms and

bureaucracies may stie innovation; there is less inherent exibility,

different expectations, and less openness to change. A tight regulatory

structure and ormal rules o behavior are thought to limit innovation

(e.g., DARPA vs. DOD).

Tere is a set o assertions commonly made with respect to compe-tition and innovation or which evidence is problematic. Tat does not

mean that these assertions are incorrect, only that they are difficult to

demonstrate with high condence.


12/21


Smaller, more exible rms are more innovative. Some evidencesupports this, though small rms ofen have difficulty nding

the resources required to ully develop, test, market, and gain

acceptance or a new concept or technology.

Commercial rms are believed to be more innovative than thedeense industry. Tis assertion underlies DOD policies con-

cerning the use o commercial processes and products as well

as alternative contracting strategies such as the other transac-

tion authority (OA) established to attract nontraditional rms

to deense work.

Innovation ofen comes rom second tier or niche rms, not justthe industry leaders. An interesting example o this phenom-

enon is in the military aircraf sector over the past 100 years:

each new technology era in military aircraf (biplane, propeller

monoplane, subsonic jet, supersonic jet, and stealth) was initiat-

ed by a second-tier aircraf rm or a niche rm (e.g., aircraf en-

gines) that would then become a dominant player or that era.10

o some degree, this assertion offers some support or the notion

that smaller rms tend to be more innovative.11

H O W C O M P L E X I T Y M I G H T A F F E C T C O M P E T I T I O N A N D

I N N O V AT I O N I N D E F E N S E A C Q U I S I T I O N P R O G R A M S

Complexity itsel has affected the nature o competition and innova-

tion in the deense industry.

Many o the more recent programs are larger and more technically

complex in terms o the use o inormation technology, system inter-dependence, and interoperability. Larger complex programs may re-

quire larger rms with substantial resources, breadth o capability, and

the inrastructure to manage them effectively. Firms remaining in the

deense market are relatively larger than they used to be and are them-

selves more complex (vertically and horizontally). Te lead rm may

ocus more effort on system engineering/integration roles, including

sofware development, rather than component and subsystem devel-

opment and abrication. In this sense, industry consolidation might beseen as an enabler or managing complexity.

Te top-tier deense rms have restructured to better address tech-

nical, organizational, and environmental complexity. echnical com-


13/21


plexity emphasizes systems and systems-o-systems engineering and

integration, which in turn require an emphasis on this capability at the

prime contractor level. Most o the top-tier deense rms have restruc-

tured in a way that reects this ocus, combining their military workunder a new integrated deense business unit and hiring or training

systems engineers. Boeing, Northrop, and Lockheed Martin all ol-

lowed this pattern. Tese integrated deense business units also posi-

tion the rms to better address interdependency and interoperability

across system types, a challenge driven at least in part by technical, or-

ganizational, and environmental complexity. Te emphasis on systems

integration and system engineering capabilities offers a new niche or

competition and innovation; the prime contractor competition in sev-eral recent programsmissile deense, FCS, DD(X)emphasized sys-

tems engineering and integration explicitly.

Tis also elevates the role o the lower tiers; DOD-managed or

inuenced competition may now be more applicable and more im-

portant below the level o prime contractor. I DOD decides compe-

titions at the prime contractor level because the government itsel is

unable to address the organizational complexity o a program, then

competition at the lower tiers will be lef to these large rms, who maydecide such competitions based on different criteria than the govern-

ment might preer. At a minimum, increasing DOD awareness o the

complete business base supporting a program may provide valuable

inormation to policymakers on how competition can be applied in a

particular case.

Complexity has also inuenced the actors affecting innovation

in many o the same ways. High barriers to entry remain, including

capital investment and a workorce with the requisite characteristics.Complexity introduces yet another set o required workorce and orga-

nizational capabilities. Tere are many ewer rms at top industry tiers

in mature industry sectors (e.g., xed wing and rotary aircraf, large

surace combatants, submarines, heavy armored vehicles). Te gov-

ernment or deense-specic barriers to entry also remain, including

knowledge and business processes that satisy statutes and regulations

as well as limited prot and limited growth in the deense sector.

An abundance o technical innovations (and associated concepts)has driven some o the complexity seen in todays acquisition pro-

grams. Complex systems have both advantages and disadvantages;

they tend to be more costly, less reliable (more parts), harder to x,


14/21


and less predictable in behavior due to emergent properties. However,

they also offer new capabilities useul to the warghter.

But technical, organizational, and environmental complexity have

also created new opportunities requiring substantial innovation inconcepts and technology, leading to new capabilities and new niches

within the deense industry. In more established sectors, innovation

can be in the systems integration unction, or in people, organiza-

tions, or management structures that bring diverse skill sets together.

Te potential o inormation technology to provide new capabilities

or replace manned unction with unmanned systems (e.g., automated

re control, shipboard reghting, autonomous vehicles) has only just

begun.Although technical complexity dominates many discussions, it is

not just technology that can be complex. Organizational and environ-

mental complexities also offer opportunity or innovation. Te chang-

ing nature o the threat has opened new sectors where less maturity

gives innovation a relatively higher expected payoff. Tese capability

areas include unmanned vehicles (air, ground, sea surace, and under-

water), counterinsurgency (improvised explosive device, or IED, de-

eat, detection, communication/translation), space, and cyber warare.Such new capabilities have implications or organizational structure o

both the acquiring and user communities within DOD.

Te technical, organizational, and environmental complexity dis-

cussed above may affect the conditions under which competition and

innovation yield their expected benets within the context o deense

acquisition. However, there are other actors that also affect competi-

tion and innovation in deense programs, independent o complexity.

Tere are relatively ewer new programs as compared to prior peri-ods, at least in established deense sectors, reducing opportunities or

competition (and innovation), but this was driven largely by budget

pressure in the 1990s. Tere are ewer rms in the deense industry at

all tiers, and, in some cases, very ew rms are capable o designing,

developing, and producing critical materials or components. Barriers

to entry in the deense industry have always been high and are perhaps

even higher now, at least at the top tiers. Workorce capability in the

deense industry has also been identied as an issue; the older, experi-enced workorce is nearing retirement, and ewer younger workers are

entering the deense industry. A scarcity o certain skills in the work-

orce can lessen a rms ability to compete.


15/21


I M P L I C A T I O N S F O R P O L I C Y M A K E R S

Complexity has contributed to changes in the nature o the weapon

systems that DOD buys as well as changes in deense industry struc-

ture, how competition may be applied at the program level, the val-ue o that competition, and the drivers o innovation. Policymakers

should be aware o such changes when considering allocation o unds

across possible weapon system investment portolios, new program

starts, acquisition strategies or programs, and management structure

and processes.

Acquisition officials should consider the ollowing observations

when thinking about the application o competition to programs with-

in an increasingly complex acquisition environment and implicationsor innovation:

Little real competition currently exists in mature deense in-dustry sectors. Complexity o programs or systems is only one

cause. Other causes include ewer new programs providing op-

portunities or competition, an industry base that continues to

consolidate in terms o the number o rms with specic capa-

bilities, and increased teaming on large programs (i.e., spreadingthe business base).

Te globalization o the deense industryan issue that has notbeen addressed in this chapteroffers some competitive oppor-

tunities by expanding both the number o programs and number

o rms in the broader deense market. U.S. rms have competed

in programs or other nations by offering versions o products

sold to the U.S. military. Non-U.S. rms have competed in DOD

programs either directly or by teaming or acquiring U.S. rms(e.g., BAE and EADS). Tere are both near- and long-term im-

pacts to globalization that warrant urther study.

Relatively new deense industry sectors such as unmanned ve-hicles offer opportunity or competition that can lead to innova-

tion as well as provide other benets expected rom competition.

Tese new sectors are expanding markets with lower barriers to

entry and ew truly dominant players.Te organizations that manage complexity in weapon systemprograms are themselves complex. Tis applies to government

and industry program offices as well as oversight organizations


16/21


in the military services and the Office o the Secretary o De-

ense (OSD). In complex organizations, the interactions o many

stakeholders can occasionally produce counterintuitive results.

Government has traditionally ocused competition at the primecontractor level. With competition among these large rms in-

creasingly ocused on system engineering and system integration

unctions, the competition that might produce technological in-

novations may more ofen happen at lower tiers. Te government

currently has ew mechanisms to inuence or manage competi-

tion among lower-tier rms.

Bureaucracies tend not to innovate well, by their very nature.Tey are generally set up to ensure standardized processes

rather than to develop new ideas. Tis characteristic applies to

both government and the increasingly large deense industry

rms in the top tier. In contrast, innovation seems to be acili-

tated by removing programs or projects rom the mainstream.

Examples include DARPAs accomplishments as well as the ac-

complishments o the several rapid reaction organizations set

up to support warghters in Iraq and Aghanistan. Historically,the relative success o classied (or black) programs has been

attributed in part to the nonstandard acquisition environment

accorded them. Similarly, some large deense rms have set up

advanced program operations to insulate them rom the main-

stream and oster innovation, such as Boeings Phantom Works

and Lockheeds Skunk Works.

One o the more important observations is that the actors affect-ing competition the mostewer programs, budget pressure, indus-

try consolidationhave little to do with complexity per se. Although

complexity may change the nature o a competition by emphasizing

large-scale systems engineering and integration rather than strict cost

and perormance variables, these other actors will still limit how com-

petition can be applied in mature deense industry sectors. In contrast,

complexity appears to have provided more opportunity or competi-

tion and innovation in relatively newer deense industry segments.How can complexity in weapon system development be man-

aged? Tere are two interrelated approaches; a mix o both is prob-

ably needed. One approach is to limit technical complexity in weapon


17/21


system design by developing metrics or such complexity and using

those metrics as part o the decision process when ormulating a pro-

grams acquisition strategy. Such metrics might include the number o

independent systems or large subsystems that need to be integrated,the number o interactions o systems within a weapon system, the

number o external (or complementary) systems interactions required,

and the number o organizations involved in design, development, and

management.

A second approach is to adapt management techniques and insti-

tutional structures to better manage complexity. Hypotheses could be

developed and tested at a smaller scale (e.g., program level) beore ap-

plying more widely. For instance, i technical complexity in a weaponsystem makes cost, schedule, and perormance more difficult to pre-

dict, then an organization structured to respond to such uncertain-

ties can be designed. Being responsive to uncertainty requires a good

monitoring approach as well as considerable exibility in making cost-

perormance tradeoffs and allocating unds across a program. Pilot

programs o the past have used this basic approach and have ound

some successe.g., the initial JDAM (joint direct attack munition)

pilot program or DARPAs Predator and HAEUAV (high-altitudeendurance unmanned aerial vehicle) programs. Simpliying decision

processes may help minimize organizational complexity.

Policymakers should also acknowledge that the technical, organi-

zational, and environmental complexity actors affecting acquisition

suggest that we may not want to preserve the current government and

industry structure; rather, we may want to consider how government

can effect changes that respond to the evolving nature o acquisition

and that create an environment that encourages innovation. Similarly,it is not clear that the current acquisition process needs to be main-

tained. Changes in the characteristics o what we buy and in the nature

o the threat suggest a need or changes in the processes and institutional

structures associated with acquisition. Te policy levers that DOD has

used in the past to shape industry, generate competition, and stimulate

innovation are still relevant today and include the ollowing:

DOD is the only buyer or many new technologies. It can act as the early adopter or innovative concepts and technologies.DOD can use this status to shape R&D in the directions it wants

to go.


18/21


RD&E (research, development, testing, and evaluation) und-ingboth the amount and distributionis a major lever or

DOD. DOD can diffuse private sector risk and ensure that a

broad set o concepts and technologies are being pursued.

Te requency and type o new programs, clearly related to und-ing amounts and distributions, are also critical. More programs

provide more opportunity or competition and innovation. Te

increased use o smaller, ocused concept and technology dem-

onstration projects is an important policy lever. Advanced ech-

nology Demonstrations (ADs) and Advanced Concept and

echnology Demonstrations (ACDs) are examples o programstructures whose use acilitates both competition and innovation.

Careul attention must be paid to transitioning the results o such

technology demonstration activities to major deense acquisition

programs, particularly in terms o the doctrinal and sustainment

issues ofen overlooked in technology demonstrations.

Improved use o evolutionary acquisition strategies may alsooffer opportunities or competition and innovation. Such pro-

grams could be planned as a series o incrementally developedcapabilities in which some portion o that incremental capability

can be competed in an effort to encourage innovation.

Use o less constrained contracting mechanisms, such as OA,can attract nontraditional rms and allow the exibility to both

generate and pursue new ideas.

At the same time however, it is important to recognize that cur-

rent acquisition policy and practice, which have remain relativelyunchanged or several decades, embody lessons in how to acquire

complex systems and thus should not be discarded under the pretext

o change without careul review.

N O T E S

1. Tis work, which was sponsored by the Office o the Secretary o Deenseunder contract W74V8H-06-C-0002, reects the views o the author. It does

not reect the views o RAND or any o its sponsors.2. Robert S. Leonard and Jeffrey A. Drezner, Innovative Development: Global

Hawk and DarkStarHAE UAV ACD Program Description and ComparativeAnalysis,MR-1474-AF (Santa Monica, Cali.: RAND Corporation, 2002).


19/21


3. Robert A. Dietrick, Impact o Weapon System Complexity on SystemsAcquisition, in James R. Rothenue and Marsha J. Kwolek, Streamlining DOD

Acquisition: Balancing Schedule with Complexity (Montgomery, Ala.: Center

or Strategy and echnology, Air War College, Air University, Maxwell AirForce Base, September 2006).4. Because technical complexity is very difficult to measure empirically, ew

analyses o program outcomes do more than simply raise the issue and asserta relationship.

5. Mark Lorell, Te U.S. Combat Aircraf Industry, 19092000: Structure,Competition, Innovation,MR-1696-OSD (Santa Monica, Cali.: RAND Cor-poration, 2003).

6. John J. Young, Jr., under secretary o deense or acquisition, technology,

and logistics, Memorandum, Subject: Prototyping and Competition,Septem-ber 19, 2007. Competition has a long history in the U.S. deense industry.Te very strong positive bias toward competition has its roots in the cultureo capitalism and entrepreneurship that has driven much U.S. economichistory.

7. Bath Iron Works and Electric Boat are both subsidiaries o General Dy-namics. Te various shipbuilding portions o Northrop Grumman have re-cently merged into a single entity. Tus, in some ways, there are really onlytwo rms covering all large Navy shipbuilding programssubs, surace com-

batants, carriers, and amphibious assault ships.8. Richard H. Van Atta and Michael J. Lippitz, ransormation and ransi-

tion: DARPAs Role in Fostering an Emerging Revolution in Military Affairs,Volume 1: Overall Assessment, IDA Paper P-3698 (Alexandria, Va.: Instituteor Deense Analyses, April 2003); Deense Advanced Projects ResearchAgency, echnology ransition, January 1997, http://www.darpa.mil/body/pd/transition.pd.

9. Paul Bracken, Innovation and the U.S. Deense Industry, June 6, 2002(unpublished input to RAND project). Te two prior studies are Michael Por-

ter, Te Competitive Advantage o Nations(New York: Free Press, 1990), andRichard Nelson, ed., National Innovation Systems(New York: Oxord Univer-sity Press, 1993). See also, John Birkler et al., Competition and Innovation inthe U.S. Fixed-Wing Military Aircraf Industry,MR-1656-OSD (Santa Monica,Cali.: RAND Corporation, 2003).10. Lorell, Te U.S. Combat Aircraf Industry.11. A similar analysis in other industry sectors has not been perormed, so itis uncertain how widespread this pattern is.

R E F E R E N C E SAdedeji, Adebayo, David Arthur, Eris Labs, Fran Lussier, and Robie Samanta-

Roy. NASAs Space Flight Operations Contract and Other echnologicallyComplex Government Activities Conducted by Contractors, July 29, 2003,Congressional Budget Office.
http://www.darpa.mil/body/pdf/transition.pdfhttp://www.darpa.mil/body/pdf/transition.pdfhttp://www.darpa.mil/body/pdf/transition.pdfhttp://www.darpa.mil/body/pdf/transition.pdfhttp://www.darpa.mil/body/pdf/transition.pdf


20/21


Arena, Mark V., Robert S. Leonard, Sheila E. Murray, and Obaid Younossi.Historical Cost Growth o Completed Weapon System Programs. R-343-AF. Santa Monica, Cali.: RAND Corporation, 2006.

Birkler, John, Giles Smith, Glenn A. Kent, and Robert V. Johnson.An Acquisi-tion Strategy, Process, and Organization or Innovative Systems.MR-1098-OSD. Santa Monica, Cali.: RAND Corporation, 2000.

Birkler, John, Anthony G. Bower, Jeffrey A. Drezner, Gordon Lee, Mark Lo-rell, Giles Smith, Fred imson, William P.G. rimble, and Obaid Younossi.Competition and Innovation in the U.S. Fixed-Wing Military Aircraf Indus-try. MR-1656-OSD. Santa Monica, Cali.: RAND Corporation, 2003.

Bracken, Paul. Innovation and the U.S. Deense Industry. June 6, 2002 (un-published; input to RAND project).

. Innovation Systems in National Deense. April 25, 2002 (unpub-lished; input to RAND project).

Chao, Pierre, A., Guy Ben-Ari, Greg Sanders, David Scruggs, and NicholasWilson. Structure and Dynamics o the U.S. Proessional Services IndustrialBase, 19912005. Washington, D.C.: Center or Strategic and InternationalStudies, May 2007.

Deense Advanced Projects Research Agency. echnology ransition. January1997, http://www.darpa.mil/body/pd/transition.pd.

Dietrick, Robert A. Impact o Weapon System Complexity on Systems Acqui-sition. Chapter 2 in James R. Rothenue and Marsha J. Kwolek, Streamlin-ing DOD Acquisition: Balancing Schedule with Complexity. Montgomery,Ala.: Center or Strategy and echnology, Air War College, Air University,Maxwell Air Force Base, September 2006.

Dombrowski, Peter J., Eugene Gholz, and Andrew L. Ross. Military rans-ormation and the Deense Industry afer Next: Te Deense Industrial Im-plications o Network-Centric Warare. Newport, R.I.: Naval War College,September 2002.

Leonard, Robert S., and Jeffrey A. Drezner. Innovative Development: GlobalHawk and DarkStarHAE UAV ACD Program Description and Com-

parative Analysis.MR-1474-AF. Santa Monica, Cali.: RAND Corporation,2002.

Lorell, Mark. Te U.S. Combat Aircraf Industry, 19092000: Structure, Com-petition, Innovation. MR-1696-OSD. Santa Monica, Cali.: RAND Corpo-ration, 2003.

Schank, John F., Giles K. Smith, John Birkler, Brien Alkire, Michael Boito,

Gordon Lee, Raj Raman, and John Ablard. Acquisition and CompetitionStrategy Options or the DD(X): Tus U.S. Navys 21st Century Destroyer.MG-259/1-Navy. Santa Monica, Cali.: RAND Corporation, 2006.

Van Atta, Richard H., and Michael J. Lippitz, ransormation and ransition:DARPAs Role in Fostering an Emerging Revolution in Military Affairs, Vol-
http://www.darpa.mil/body/pdf/transition.pdfhttp://www.darpa.mil/body/pdf/transition.pdf


21/21


ume 1: Overall Assessment.IDA Paper P-3698. Alexandria, Va.: Institute orDeense Analyses, April 2003.

Young, John J., Jr., under secretary o deense or acquisition, technology, and

logistics. Memorandum. Subject: Prototyping and Competition, September19, 2007.

Competition and Innovation Under Complexity

Documents

Transcript of Competition and Innovation Under Complexity