[American Institute of Aeronautics and Astronautics Space 2005 - Long Beach, California ()] Space 2005 - Political Sustainability in Space Exploration: A Game Theoretic Approach

American Institute of Aeronautics and Astronautics1

Political Sustainability in Space Exploration:A Game Theoretic Approach

D. A. Broniatowski* and A. L. Weigel**

Massachusetts Institute of Technology, Cambridge, MA 02139

Abstract

The swiftly changing nature of the policy environment introduces significant uncertaintyinto the design of those technical systems that rely on public resources. Politics, the processthrough which stakeholders possessing different values and interests interact, necessarilyimpacts technical design as requirements change to suit different needs. Simple game-theoretic models may be used to provide insights into resource allocation dynamics betweenNASA and the U.S. Congress. This paper utilizes game theory, with a particular emphasis onBrams’ Theory of Moves, to model the process by which stakeholders within NASA and theU.S. Congress may cooperate or compete to arrive at an affordable and politicallysustainable funding level on a yearly basis. In doing so, this paper advances a game-theoreticdefinition of political sustainability.

NomenclatureCEV = Crew Exploration VehicleF = Decision to FundFY = Fiscal YearNF = Decision Not to FundNS = Decision Not to deliver a ServiceP = Politically Sustainable Funding LevelS = Decision to deliver a ServiceTOM = Theory of MovesX = Funding granted to NASA by Congress for human spaceflightX* = The President’s human spaceflight funding request for NASA

I. Introduction

Today’s aerospace systems are becoming increasingly complex, with longer design lifecycles. In particular,many of the National Aeronautics and Space Administration’s (NASA) systems, such as the International SpaceStation, the Space Shuttle, and now, the Crew Exploration Vehicle (CEV), are to be operated for decades at a time.These lifecycle times appear increasingly extended when contrasted with the American presidential term of fouryears, or a Congressional turnover rate of two years. Thus, if an engineering system is to maintain the supportrequired for continued operation, it must be able to deliver value under a constantly shifting political environment.These systems, and others, must therefore be designed with political sustainability in mind. The possibility of achange in stakeholders’ support must be made an explicit part of the design. Once the system has been designed,political sustainability becomes an issue of maintaining the support that currently exists and potentially building thebase for future support. In the case of NASA’s systems, this support often takes a budgetary form. Budgetarypolitical sustainability is aimed at maintaining a program’s support both among the public, within Congress, andmore generally, among all invested stakeholders.

* Graduate Research Assistant, Aeronautics and Astronautics/Technology and Policy Program, 77 MassachusettsAve., Rm. 17-110, and AIAA Student Member.** Assistant Professor of Aeronautics and Astronautics and Engineering Systems, 77 Massachusetts Ave., Rm. 33-404, and AIAA Member.

Space 200530 August - 1 September 2005, Long Beach, California

AIAA 2005-6799

Copyright © 2005 by David A. Broniatowski and Annalisa L. Weigel. Published by the American Institute of Aeronautics and Astronautics, Inc., with permission.


Understanding the mechanisms and processes by which a program may exhibit budgetary political sustainabilityis of paramount importance to the space exploration enterprise. This paper is therefore motivated by the desire tounderstand the dynamics surrounding changes in a program’s budgetary environment, and how best to accommodateor adapt in the face of those changes. In particular, budgetary political sustainability has proven to be a driver for theVision for Space Exploration, which instructs NASA to “Implement a sustained and affordable” space explorationprogram [1]. Design for budgetary political sustainability is at the core of this charge.

One may gain an understanding into the dynamics affecting program budgets by examining the interplay amongstakeholders. As is common in the presence of finite resources, there will be competition among individual NASAprograms, in addition to the existing rivalries between aerospace companies that receive NASA contracts.Additionally, the Executive Branch must submit the United States budget, of which the NASA budget is but a smallpart, to Congress, itself made up of 535 members representing different interests and wielding different sorts ofpower. Following the changes made by Congress, the budget must be signed into law by the President.

These, and other, stakeholders have the ability to influence the budgeting process at multiple levels by engagingin a set of strategic interactions. These interactions may be cooperative or competitive in nature. Some stakeholdersmay exercise direct methods (for example, a Presidential veto) to strongly influence the budgeting process. Othersmay utilize more indirect methods, (such as industrial lobbying of members of Congress), to influence the finaloutcome.

In this paper, we focus on the process by which stakeholders within NASA and the U.S. Congress may cooperateor compete to arrive at an affordable and politically sustainable funding level on a repeated yearly basis. In so doing,we use game theory to model the process by which stakeholders interact to affect the NASA budget. Since thebudget must be approved every year, stakeholders must engage in myopic (short-term) and non-myopic (long-term)strategies when defining the nature of their interactions. Brams’ Theory of Moves is therefore used to supplementtraditional game theory, as a means of explicitly considering players’ long-term strategic goals while examiningtheir actions and motivations [2]. We then examine circumstances under which NASA and Congress may exercise“threat power” to motivate political decision making, yielding the counter-intuitive result that NASA’s highvaluation of its human spaceflight programs acts as an incentive for Congress to reduce NASA’s funding requests.This, in turn, undermines the Shuttle program’s political sustainability. This framework is then used to define apolitically sustainable level of funding for human spaceflight in game theoretic terms.

II. Analysis: Theory of MovesIn order to examine the interplay of stakeholders using Theory of Moves (TOM), it is necessary to understand

some of TOM’s underlying concepts. In this paper, we examine the presence or absence of threat power, defined as“The ability to deter or compel an opponent to take action, at a loss to both players, given that the threatener willmake a net profit in repeated play” [3]. It is worth noting that, using TOM, repeated play is differently defined thanin standard game theory. In standard game theory, repeated play indicates that each player will literally re-play astage game multiple consecutive times with the goal of maximizing total or average utility. In TOM, repeated play“…means that there is always later play that enables a threatener to recoup losses it may have incurred earlier incarrying out threats.” [2]. In effect, a threatener may temporarily accept a loss so as to improve the final outcome.Practically speaking, this implies that a player may change position on a normal-form game matrix after the initialmove has been made. Players then alternate, moving sequentially around the matrix, until one player decides to stopmoving thereby ending the game.

Consider the following simple game description: Every year,Congress must decide on how it will budget funds to each of theExecutive Branch agencies within the Federal Government. Thissituation may be modeled as a game played between Congress,who makes the decision to Fund or Not to Fund (F,NF), and theAgency, following the policy direction of the President, whomust decide to Deliver a Service to Congress or Not to Deliver aService (S,NS) to Congress. Generally speaking, the situationmay be described with the normal-form matrix game in Figure 1.

We may use a variant of this game to describe the fundingprocess for NASA after submission of the President’s budgetrequest. In this case, we define the agency in question as NASA providing the service of flying or grounding allvehicles capable of human spaceflight (e.g., the three Space Shuttles). Four specific scenarios are studied in thispaper. The first scenario, represented by Game One below, describes a situation similar to that prevailing before the

(?,?)(?,?)

(?,?)(?,?)

CongressF NF

Age

ncy

NS

S

(?,?)(?,?)

(?,?)(?,?)

(?,?)(?,?)

(?,?)(?,?)

CongressF NF

Age

ncy

NS

S

Figure 1: A generic game matrix for theAgency-Congress game.


loss of the Shuttle Columbia wherein the Administration’s high valuation of the human spaceflight capabilityprovided by the Shuttle program and a Congressional incentive to keep costs low contribute to a budgetaryenvironment marked by incremental budgeting and policy making [4, 5]. The second scenario, represented in GameTwo, describes a change in preferences brought about by exogenous events, such as the loss of Columbia and thereorganization of the Congressional Appropriations Committees. In this situation, NASA may exercise threat powerin order to achieve its desired outcome of obtaining funding from Congress for the purposes of maintaining humanspaceflight capability. The third scenario explores the period of time in between the Columbia tragedy and thereorganization of the Appropriations Committees, when it was not clear whether Congress would endorse theconstruction of a new vehicle to maintain human spaceflight capability. In this situation, NASA’s budget requestwould be sufficiently high that the benefits delivered to Congress by maintaining human spaceflight capability donot offset the costs. Under these circumstances, NASAwould be forced to continue to fly without receiving itsfunding request. Finally, the fourth game describes asituation wherein neither NASA nor Congress wish tomaintain human spaceflight capability. In this situation,both parties agree to terminate the program.

These games are represented using an instantiation ofthe Agency-Congress game mentioned above. In thisNASA-Congress game, NASA has a decision to Groundor Fly the vehicle providing human spaceflightcapability (e.g., the Space Shuttle, or later, the CEV). Inaddition, the President makes a yearly budget request forNASA to Congress. The amount requested is designatedX*. Congress then decides to grant NASA a certain levelof funding, X, which may either be less than or greaterthan X*. A general version of the NASA-Congress isshown in the normal-form matrix in Figure 2.

A. Game One – “Can-do” NASA and Supportive CongressIn defining the parameters of this game, we assign the following preferences to each player using a series of

assumptions regarding NASA’s goals in this situation. First, we assume that NASA prefers flying its vehicle togrounding it. We justify this assumption by noting that many of NASA’s activities (such as the construction of theInternational Space Station and potential Hubble servicing missions) require a functional Space Shuttle to beexecuted. In addition, NASA Administrator Dr. Michael Griffin adds that “…it takes about $4.5 billion to keep theshuttle going, whether you fly any flights or not” [6], a significant portion of which goes to maintenance costs andrefurbishment costs on the ground [7]. Finally, we observe that, at this point in time, maintaining the capability ofhumans spaceflight may be said to be consistent with NASA’s organizational goals, as the nation’s only civil spaceagency [8]. Next, we assume that NASA prefers receiving funding to having its funding cut. Finally, we assumethat NASA prefers flying the Shuttle even when funding is not present. We justify this assumption based uponhistorical observation, wherein NASA has displayed a “can-do” attitude, perhaps compelling the attempt of complexundertakings when the resources required to support them are not present. Indeed, following the Columbia tragedy,NASA was characterized as “an organization straining to do too much with too little.” [9], indicating that NASA hasharbored a preference for action, even in the absence of sufficient budgetary resources. These assumptions aresufficient to define a ranked set of preferences for NASA for each of the four possible outcomes.

Next, we must define the preference ordering structure for Congress. We begin by assuming that Congresspossesses a strictly dominant strategy of not funding NASA. Strict dominance implies that, given a static,unchanging state for the Shuttle (either flying or grounded), Congress always prefers to save its resources, sincespending them will not alter the outcome. Therefore, given that the Shuttle is already flying, Congress will notprovide additional funding for human spaceflight, since the need for human spaceflight is already fulfilled. Maslowdescribes this phenomenon in psychological terms as follows: “…a want that is satisfied is no longer a want. Theorganism [in this case, Congress] is dominated and its behavior organized only by unsatisfied needs. If hunger issatisfied, it becomes unimportant in the current dynamics of the individual.” [10] Van Dyke elaborates, “[Politicalfigures] tend to speak of those values or interests that are threatened or that seem to be in need of attention, and theytend to forget about values and interests that seem to be assured. Sometimes there has seemed to be completeunawareness of certain values and interests and complete insensitivity to developing dangers.”[11] Similarly, giventhat the Shuttle is already grounded, and will remain so under the construct of strict dominance described above,

(?,?)(?,?)

(?,?)(?,?)

CongressFund X>X* Don’t Fund X<X*

Age

ncy

Gro

und

Fly

(?,?)(?,?)

(?,?)(?,?)

(?,?)(?,?)

(?,?)(?,?)


Age

ncy

Gro

und

Fly

Figure 2: A generic instantiation of the NASA-Congressgame. X is the funding level granted by Congress and X* isthe funding level requested by the President for NASA.


Congress will not expend resources on human spaceflight if those resources will not generate human spaceflightcapability. We also assume that Congress derives value from human spaceflight. One justification for thisconclusion is the number of jobs produced by the Space Shuttle program (the American human spaceflightprogram’s flagship):

“The Space Shuttle program occupies 640 facilities, utilizes over 900,000 equipment line items, and directly employsover 2,000 civil servants and more than 15,000 work-year-equivalent prime contractors, with an additional 3,000 peopleworking indirectly on Space Shuttle activities at all NASA Centers. Thousands more are employed at the subcontractorlevel in 43 states across the country. The total equipment value held by the Program is over $12 billion. The totalfacilities value held by the Program is approximately $5.7 billion (approximately one-third of the value of NASA’s entirefacility inventory), mostly at the field centers. There are also approximately 1,500 active suppliers and 3,000 - 4,000qualified suppliers that directly support the Space Shuttle program.” [12]

Therefore, those members of Congress with NASA employees in their district have a distinct electoral incentiveto keep the Shuttle flying, namely keeping their constituents employed and maintaining an electoral advantage.Other incentives include the achievement of foreign and scientific policy objectives, as illustrated by the followingstatement by Senator Barbara Mikulski (D-MD), Ranking Member of the Senate Appropriations Committee,Subcommittee on Commerce, Justice, Science and Related Agencies, which includes NASA:

“The United States of America should always have its own access to space. The space station, too, we need to be able tofinish that, keep our commitment to our international partners, and keep it as a premier research facility. And, of course,then there is Hubble. Everyone knows my position on Hubble. And I believe it's been the greatest telescope inventionsince Galileo himself stood on that rooftop in Florence.” [13]

It is worth noting that the International SpaceStation and the Hubble Space Telescope bothrequire additional Space Shuttle flights if they areto be completed and repaired, respectively.Finally, to fully define the preference structure, weassume that Congress prefers to pay to keep theShuttle flying, rather than to have the Shuttlegrounded. We justify this assumption by observingthat, following the Challenger and Columbiatragedies, Congress has continued to providefunding to NASA for human spaceflight activities,under the assumption that the Shuttle would returnto flight. In fact, following the Challengerexplosion, NASA was provided with additionalfunding in order to build a new Shuttle, Endeavour.These preference orderings, taken together, result inthe normal-form game matrix seen in Figure 3.

A brief inspection of the matrix yields that the Nash Equilibrium solution of this game is at (Fly, Don’t Fund).This conclusion may also be attained by iterated strict dominance of strategies. In other words, NASA can always dobetter by flying than by not flying, and Congress can always do better by refusing to fund than by agreeing to fund.This outcome is consistent with the situation described in the CAIB report, which describes NASA as “anorganization straining to do too much with too little” [9].

1. Game One -- AnalysisHaving defined the game matrix, we now ask whether, given these conditions, NASA possesses any threat

power. Given this payoff matrix, NASA would only attempt to enforce the (Fly, Fund) outcome, since that is theonly outcome that is better than the Nash equilibrium-enforced status quo. NASA could exercise this threat bygrounding the Shuttle for safety reasons, if Congress were to decide to cut the Shuttle’s funding below the amountrequested by the President. This is classified as a deterrent threat since NASA is attempting to prevent Congressfrom taking an action by threatening a retaliation [2]. We note that although this threat is real (Congress wouldsuffer a loss in utility if NASA decided to ground the Shuttle), this threat is not rational, since NASA would alsosuffer a reduction in utility. Therefore, this threat is not credible: NASA does not possess any power to enforce thisdeterrence.

Figure 3: The normal-form matrix representing Game 1.The Nash Equilibrium is circled. The shaded arearepresents Congress’ compellent threat.

(1,2)(2,1)

(3,4)(4,3)


Age

ncy

Gro

und

Fly

(1,2)(2,1)

(3,4)(4,3)

(1,2)(2,1)

(3,4)(4,3)


Age

ncy

Gro

und

Fly


Next we explore whether Congress wields any threat power. We begin by noting that Congress would prefer toenforce the Nash-equilibrium outcome of (Fly, Don’t Fund). Congress would employ a threat to prevent NASAfrom grounding the Shuttle, essentially refusing to move from its position of refusal regardless of NASA’s actions.This type of threat is classified as compellent since Congress is trying to compel NASA to fly by refusing to move.Here we note that this threat is both real (since NASA stands to lose from grounding the Shuttle) and rational (sinceCongress will only reduce its payoffs by deciding to fund). Thus Congress possesses sufficient power to enforce itsdesired outcome, namely the Nash Equilibrium outcome, in repeated play, suggesting the presence of a chronicfunding shortage. We feel that this accurately describes the situation surrounding the Shuttle Program prior to theColumbia tragedy. Congress, lacking any direct incentive to increase the budget for the Shuttle program, would, atbest, simply renew the previous year’s human spaceflight budget, effectively causing a reduction in constant-yeardollars as inflation decreases buying power [4]. To make matters worse, as the Shuttle fleet gets older, one canexpect its recovery and refurbishment costs to increase substantially, leading to a tighter budgetary environment [7].These factors, when combined with the chronic cost overruns and rushed schedule for completion of theInternational Space Station, contributed to the organizational difficulties that eventually led to the Challenger andColumbia tragedies [5, 9].

B. Game Two – “Play-it-Safe” NASA andSupportive Congress

Consider, now, the case in which NASA nolonger subscribes to the “can-do” attitude. Instead, ifNASA is not receiving sufficient funding, itgenuinely prefers to ground the Shuttle, fearing forastronauts’ safety. This change in preferences couldbe brought about by a shock to NASA, such as theColumbia tragedy, wherein priorities within theagency and/or the White House are redefined. Now,fearing for the astronauts’ safety, NASA prefers toground the Shuttle if sufficient funding is not

available. This may be justified as a precautiontaken during a period of extreme sensitivity to therisks involved in flying astronauts in a vehicle thatis perceived as unsafe. These preferences generatethe game matrix seen in Figure 4.

2. Game Two -- Analysis

Again, we begin by identifying the Nash Equilibrium of the game. We find that it is located at the “breakdownoutcome” of (Ground, Don’t Fund), so-called because it is the state resulting from a breakdown of negotiationsbetween the Congress and NASA. This implies that, if each player were to act myopically, thinking only of theirnear term interests, the Shuttle program’s future would be called into serious question. In this case, Congress wouldessentially give up on human spaceflight as an endeavor that does not deliver sufficient value to fully fund thePresident’s request, whereas NASA, fearing for astronauts’ safety in a scarce budget environment, would notmaintain the Shuttle in flight. Analyzing this game for threat power yields a drastically different result. As before,we examine the case where NASA tries to enforce the (Fly, Fund) outcome using a deterrent threat. Here we notethat the threat is real, since Congress prefers to see the Shuttle flying than to see the Shuttle grounded, and the threatis rational, since NASA prefers to ground the Shuttle in the case of no funding. NASA therefore possesses credibledeterrent threat power. In addition, we note that Congress no longer has compellent threat power. Although NotFunding is still a dominant strategy for Congress in the traditional game-theoretic sense, Congress’ threat is nolonger rational since it can do better by acquiescing to the President’s request for more NASA funding. These resultshighlight the effects that slightly changing one player’s preference ordering may have on the outcome of the game.To begin with, it is important to note that NASA, in changing its preferences, affects Congress’ threat power. Inother words, Congress’ original threat is rational only because NASA prefers to keep the Shuttle flying in spite ofinsufficient funding. This yields the counter-intuitive result that NASA’s high valuation of the Space Shuttleprogram allows Congress to under-fund the program. Cast differently, this implies that NASA, as the technicalexperts evaluating the safety of human spaceflight, might be able to elicit increased funding for the program if it can

(2,2)(3,1)

(1,4)(4,3)


Age

ncy

Gro

und

Fly

(2,2)(3,1)

(1,4)(4,3)


Age

ncy

Gro

und

Fly

(2,2)(3,1)

(1,4)(4,3)

(2,2)(3,1)

(1,4)(4,3)


Age

ncy

Gro

und

Fly

Figure 4: The normal-form matrix representing Game 2.The Nash Equilibrium is circled. Arrows represent thedeterrent threat that would be employed by NASA in theevent Congress chooses not to fund.


be demonstrated that such funding is necessary to ensure crew safety and vehicle reliability. It will be noted laterthat there is a practical limit to how much money Congress might be willing to spend.

It is worth noting that the Columbia tragedy has effectively acted to change NASA’s preferences as describedabove, with the effect of generating the breakdown outcome until the Shuttle is returned to flight. NASAAdministrator Michael D. Griffin swiftly took advantage of the threat power this situation awarded him:

“NASA’s new boss made an impassioned case yesterday [May 2, 2005] for speeding up development of a new spacecraftso that the United States will not lose access to space when the shuttle is retired … Griffin wants to fly the proposed newspacecraft as soon as possible once the space shuttle fleet is retired in 2010 -- avoiding a four-year gap in which theUnited States would have no way to launch astronauts … Griffin said he finds that four-year launch gap unacceptable andhopes to have a plan for closing it by mid-July … "CEV needs to be safe, it needs to be simple, it needs to be soon,"Griffin told reporters later in the afternoon…"The six-year gap between the 1975 Apollo-Soyuz mission and the 1981debut of the shuttle damaged both the U.S. space program and the nation", Griffin said. "I don't want to do it again." [14]

“The estimated cost of these new vehicles is from $10 billion to $15 billion through 2015…NASA hopes to pay the tabfrom its scheduled modest budget increases and saving from falling return to flight costs. But one official says that thosereturn-to-flight costs will climb as high as $7 billion over 5 years--$2 billion more than previous estimated. That figurewould leave little room for new ventures, the cost of which have traditionally been underestimated.” [15]

In this situation, the NASA Administrator is advocating that Congress at least approve, if not improve upon,NASA’s budget for the next years, threatening Congress with a lapse in human spaceflight capability that, ifunchecked and under funded, could become debilitating or, at worst, permanent. Acceleration of CEV development,although eliminating this gap, tends to compound the budget issue, making the threat more real and more credible toCongress. In an action that would make the threat explicit, Administrator Griffin is rumored to be considering arequest for a Return-to-Flight supplemental from Congress [16]. It is worth noting that such a request to Congresscannot be made formally without the approval of the Executive Office of the President.

This strategy seems to be working. At the time of writing of this paper, many prominent members of Congresshave expressed their support for NASA’s human spaceflight activities, and seemed poised to approve NASA’sbudget request. As of June 17, 2005, the House of Representatives supported NASA’s budget request:

“House Majority Leader Tom DeLay (R-Texas) today said the Science, State, Justice and Commerce spending bill, whichincludes $16.5 billion for NASA, will continue Congress' work to implement President Bush's bold new vision for spaceexploration. The House of Representatives passed the appropriation bill today by a bipartisan vote of 418- 7...Thefunding bill passed today sets aside $16.5 billion for NASA - $275 million more than last year's bill and $15 millionabove the administration's request. In addition to providing the full request for the Space Shuttle program, this legislationfunds the president's vision for space exploration at $3.1 billion…"The president's vision will be fulfilled, NASA'smission will be accomplished, and mankind's ancient questions will be answered," DeLay said. "That's notoverconfidence; that's a promise."” [17]

Fulfillment of this promise began on July 22, 2005, with the passage of a NASA reauthorization.

“By an overwhelming margin, the U.S. House of Representatives today passed legislation to reauthorize the NationalAeronautics and Space Administration (NASA) that was sponsored by Space and Aeronautics Subcommittee ChairmanKen Calvert (R-CA) and Science Committee Chairman Sherwood Boehlert (R-NY). The bill, H.R. 3070, NationalAeronautics and Space Administration Authorization Act of 2005, was adopted by of vote of 383 to 15. [H.R. 3070]allows NASA to proceed with its plan to retire the Space Shuttle fleet by the end of 2010; and encourages NASA tolaunch the Crew Exploration Vehicle (the Shuttle's replacement) as close to 2010 as possible…A manager's amendmentoffered by Chairman Boehlert was agreed to by voice vote. In addition to making technical and clarifying changes, theamendment…Increases the amount of funding authorized to be appropriated for NASA to support the President's budgetrequest for exploration for fiscal years 2006 and 2007 [and] expresses the Sense of the Congress that NASA should returnthe Space Shuttle to flight as soon as the Administrator determines that it can be accomplished with an acceptable level ofsafety”[18]“The Manager’s Amendment increases funding for FY06 to $16.9 billion and elevates FY07 funding to $17.7 billion.This increase restores the full funding requested by the Administration for Human Exploration…’I am pleased with thebipartisan strides that the House Science Committee has made to work together to achieve the best results for NASA.Such an inclusive compromise indicates that this country can truly achieve its vision of space exploration for generationsto come,’ commented Rep. Sheila Jackson-Lee (D-TX).”[19]


“"This Congress recognizes and embraces the importance of NASA's technological innovation and research, and thisreauthorization provides for it," DeLay said. "Ultimately, this bill does one thing: it gives the men and women of NASA– many of whom I am fortunate enough to represent – the resources they need to make their next giant leap"” [20]

In addition, prominent Senators from both political parties have voiced their support for NASA’s humanspaceflight activities. Senator Kay Bailey Hutchison (R-TX), a member of the Senate Committee on Appropriations;Subcommittee on Commerce, Justice and Science, as well as the Chair of the Senate Committee on Commerce,Science and Transportation; Subcommittee on Science and Space, stated the following in a May 12, 2005 hearingwith NASA Administrator Griffin, effectively belying the reality of NASA’s deterrent threat:

“Where I have questions and concerns about NASA, they revolve around longer-term impacts to our current investmentsin human space flight capabilities. As you know, Mr. Administrator, I am concerned about the possibility of a gapbetween the planned retirement of the shuttle and the availability of the replacement crew return vehicle. I think a five-year gap is unacceptable. I think it is not only a risk to the important scientific research that we are doing, but it is asecurity risk to our country. And I am pleased that you have shared the same concerns. And I know both the chairmanand the ranking member here have also expressed those concerns.” [13]

In addition, coincident with the passage of H.R. 3070, mentioned above, Senator Hutchison and her Democraticcounterpart, Sen. Bill Nelson of Florida, submitted an amendment to the Defense Authorization Bill “expressing theSense of the Senate regarding the critical nature of human space flight to America’s security”[21]. This amendmentincludes the following language:

“(a) FINDINGS.--The Congress finds that--(1) human spaceflight preeminence allows the United States to project leadership around the world and forms animportant component of United States national security;(2) continued development of human spaceflight in low-Earth orbit, on the Moon, and beyond adds to the overall nationalstrategic posture;(3) human spaceflight enables continued stewardship of the region between the earth and the Moon--an area that iscritical and of growing national and international security relevance;(4) human spaceflight provides unprecedented opportunities for the United States to lead peaceful and productiveinternational relationships with the world community in support of United States security and geo-political objectives;(5) a growing number of nations are pursuing human spaceflight and space-related capabilities, including China andIndia;(6) past investments in human spaceflight capabilities represent a national resource that can be built upon and leveragedfor a broad range of purposes, including national and economic security; and(7) the industrial base and capabilities represented by the Space Transportation System provide a critical dissimilarlaunch capability for the nation.(b) SENSE OF THE SENATE.--It is the sense of the Senate that it is in the national security interest of the UnitedStates to maintain uninterrupted preeminence in human spaceflight.” [22]

Senator Hutchison explained her reasoning as follows:

“During our consideration of this bill and during hearings, it became clear that we must think of manned spaceflight interms of national security, as well as science and exploration. For these reasons, I believe it is important that in thecontext of this Defense authorization bill, we express the sense of the Senate that we recognize the important and vitalrole of human spaceflight in the furtherance of our national security interests, and that we reaffirm our commitment toretaining our Nation's leadership role in the growing international human spaceflight community of nations…Let us standunited to recognize the inexorable link and importance of human spaceflight in our national security. I hope mycolleagues will support this important statement that says keeping our dominance in space is a matter of national securityfor our country… The idea that we would consider a hiatus in our opportunities to put humans in space is one that isunacceptable to me and to my ranking member. We hope the sense-of-the-Senate amendment will be adopted toacknowledge and assure that space exploration is shown to be a part of our national security interests. It is essential thatwe not, in any way, ever let our eye get off that ball, that we must have dominance in space if we are going to keep ourpreeminence in national defense.” [22]

Senator Nelson added:

“I join with my colleague, the distinguished Senator from Texas, who serves as the Chair of our Science and SpaceSubcommittee and of which I have the privilege of being the ranking member…What this amendment does--and I wantto say a word about our two colleagues who lead our Armed Services Committee who I think will accept this


amendment--it simply says: It is the sense of the Senate that it is in the national interest of the United States to maintainuninterrupted preeminence in human spaceflight. Why? Why are we saying that? Because we could be in a posture that ifthe space shuttle is shut down in 2010, which is the timeline, and if we did not soon thereafter come with a new vehicle tohave human access to space…that if we don't watch out and we have a hiatus between when we shut down the spaceshuttle and when the new vehicle flies, one originally that was planned by NASA to be 4 years, which meant it was goingto be 6, 7, or 8 years, then we don't have an American vehicle to get into space. If that is not bad enough, who knowswhat the geopolitics of planet Earth is going to be in the years 2011 to 2018. We may find that those vehicles we rely onto get today, for example, to the space station, when we are down with the American vehicle, may be aligned withsomebody else. That is why we want to make sure we have that other vehicle ready about the time we shut down thespace shuttle so we will have human access to this international space station and reap the benefits, once it is fullyconstructed, of all the experimentation and the processing of materials we can uniquely do in the microgravity of Earth'sorbit. That is the importance, in this Senator's mind, of this resolution.” [22]

These statements reflect a high valuation of human spaceflight on the part of Congress, thereby enabling a realthreat in the prospect of the loss of American human spaceflight capability. In addition, this threat is credible sincecurrent plans call for the retirement of the Shuttle by 2010. It is worth noting that the Senate has consideredlegislation that would require that the Shuttle fly until a CEV becomes available, although this legislation has notbeen passed into law [23]. This language seems aimed at reducing the credibility of NASA’s threat, suggesting thatthe Senate is aware of the vulnerability of their position. A requirement to keep the Shuttle flying until the CEV isdeveloped could, in principle, allow Congress to continue behaving in an incremental fashion with regards to humanspaceflight, since a delay in the CEV due to lack of funding would no longer be a present concern. Indeed, ratherthan fund the full development of a new vehicle, Congress could simply provide the additional funding required tokeep the Shuttle flying while chronically under funding CEV development. This would be a return to the pre-Columbia funding paradigm. Senator Hutchison acknowledges this, stating that

“NASA has begun several efforts in the past decade, to develop a replacement vehicle for human space flight, with aview to eventually retiring the space shuttle. Each of them has failed, after considerable expense, to find the technologicalbreakthrough that was necessary for their success. They were focused on new technologies, new systems that werelargely untested, and unproven. We are now out of time, and can no longer afford the luxury of attempting to develop adramatically new and different human space flight capability.” [23]

As will be noted later, these additional dynamics seem to indicate the presence of a spending cap, beyond whichNASA can not reasonably expect to receive funding, even if employing a deterrent threat, although rumorsregarding the intention of the White House to retire the Shuttles even earlier than planned may serve to pre-emptSenate action [24]. It is worth noting that the events surrounding the Shuttle Discovery’s return to flight on July 26,2005, have further served to increase the credibility of NASA’s threat to ground the fleet for safety concerns:

“NASA may never be able to prevent threatening chunks of insulation foam from breaking off the shuttle's fuel tankduring launch, the agency's chief said Thursday, a day after future flights were ordered grounded because of the problemduring Discovery's liftoff. "We are trying to get it down to the level that cannot damage the orbiter,'' NASA administratorMichael Griffin told NBC's "Today." "We will never be able to get the amount of debris shed by the tank down to zero,"he said. With Discovery in orbit, NASA grounded all future flights because a large chunk of foam had broken off theexternal fuel tank in a hauntingly similar fashion to Columbia's doomed mission.” [25]

These responses demonstrate that the value of this spending cap is not well-understood and may therefore beprivate information, uncovered only in the halls of Congress.

3. The Power of Exogenous FactorsThis analysis illustrates the power of exogenous factors (i.e., those that are beyond players’ control, such as

factors that change player preferences) in determining the outcome of a game. As seen in the first example, theNash-equilibrium outcome, reinforced by Congress’ threat power, predicted that human spaceflight would be underfunded. Furthermore, TOM predicts that this outcome would occur on a repeated (in this case, yearly) basis..Following the change in preferences, we note that a new paradigm has arisen whereby NASA, by acting proactively,may ensure the funding profile for human spaceflight. This outcome will be repeated for as long as NASA is willingand able to carry out its deterrent threat. This suggests a new set of parameters governing the incremental fundingprofile following the change in preferences. In the event that NASA’s preferences were to return to those seen priorto the accident, the outcome would revert back as well. This model therefore supports McCurdy’s identification ofincrementalism as the modus operandi within NASA’s internal political structure [5]. In addition, the presence of


short, non-incremental time periods during which the parameters of the incremental model shift may be interpretedas “political shocks” or “policy punctuations” in between incremental periods [26, 27]. From apsychological/informational perspective, NASA is not strictly carrying out a threat or asking for more funds; rather,the Administrator is providing testimony reminding Congress of the ill effects resulting from neglect of a highly-valued capability. This dynamic is particularly important since the NASA Administrator is limited to formallyrequesting the amount put forward in the President’s budget request. Any NASA desire for more funding outside ofthe President’s request cannot be made through explicit formal channels without the NASA Administrator facing asituation potentially damaging to his career.

C. Game Three – “Can-do” NASA andUnsupportive Congress

Next, we consider another change to the set ofpreferences. In this case, we examine the case whereinCongress prefers saving money for other endeavors,even at the expense of human spaceflight, to spendingthe money required to keep humans in space. Thismight occur in the circumstance wherein adetermination has been made by Congress that thevalue delivered in maintaining the Shuttle program issimply not sufficiently high to warrant the fundsrequested by NASA. Alternatively, Congress mightsimply decide that human spaceflight activities are nolonger within the nation’s political interest. Thissituation is described by the game matrix in Figure 5.

4. Game Three – Analysis

In this case, we note that Congress, once again, has compellent threat power. NASA, lacking a threat that isneither real nor rational has no credible threat available, and will simply default to flying the Shuttle withinsufficient funding. This game assumes, as before, that NASA prefers flying the Shuttle without funding. It is worthnoting that outcomes here are indistinguishable from those in the first example, suggesting that NASA may not beable to determine the preferences of Congress if they are receiving insufficient funding. In effect, by refusing to giveNASA the requested funding, Congress sends a signal indicating a low valuation of the human spaceflight program.

Although the Shuttle was grounded in 2003, following the Columbia tragedy, it was only in 2005 that NASA,under the leadership of newly-appointed Administrator Michael D. Griffin, began to successfully utilize theagency’s threat power. Indeed, a comparison of NASA’s human spaceflight budget for FY2004 awarded byCongress with the President’s request shows that the President’s budget request was under funded by $253 millionfor human spaceflight alone [28, 29]. FY2005 tells a different story; whereas the President requested $16.2 billion intotal for NASA, the Senate awarded $16.4 billion, including $800 million in emergency funds. In contrast, theHouse of Representatives only awarded $15.1 billion in total; specifically awarding NASA $959.6 million less thanwhat had been requested for human spaceflight. This constituted a $23.9 million reduction over the FY2004appropriation, strongly suggesting that the House of Representatives did not value human spaceflight enough tofulfill the President’s funding request for NASA [30, 31]. Although NASA, and by extension, human spaceflightprograms, eventually received full funding, this required extraordinary measures on the parts of President GeorgeW. Bush and House Majority Leader Tom DeLay (R-TX) [32]. President Bush even went so far as to threaten toveto any budget that was not congruent with his funding request for NASA, an unprecedented move in space policy[30]. At a time when the Shuttle was grounded and the future of American human spaceflight was uncertain, thesemoves by the House of Representatives called into question Congressional valuation of the human spaceflightprogram, suggesting that NASA simply did not have the threat power necessary to enforce its budget requests. Thesituation changed in February 2005, when Leader DeLay orchestrated the reorganization of the HouseAppropriations Committee. As a result, responsibility for NASA spending was shifted from what had been theVeterans’ Affairs & Housing and Urban Development (VA-HUD) subcommittee to the Science, State, Justice andCommerce Committee [33]. This type of event is a rarity in Congressional relations, and may be considered enoughof a political shock to have changed Congressional preferences in favor of promoting human spaceflight. This, inturn, yielded Administrator Griffin the opportunity to take advantage of the newly available threat power coincidentwith his arrival.

(1,3)(2,1)

(3,4)(4,2)


Age

ncy

Gro

und

Fly

(1,3)(2,1)

(3,4)(4,2)

(1,3)(2,1)

(3,4)(4,2)


Age

ncy

Gro

und

Fly

Figure 5: The normal-form matrix representing Game 3.The Nash Equilibrium is circled. The shaded arearepresents Congress’ compellent threat power.


D. Game Four – “Play-it-Safe” NASA andUnsupportive Congress

Finally, for completeness, we consider the casewherein Congress possesses a low valuation for theSpace Shuttle program and NASA prefers toground the shuttle in the presence of low funding.This game is represented by the normal-formmatrix in Figure 6.

We note that, in this case, neither playerpossesses threat power of any kind. As a result,neither player is able to enforce a desired outcome.In addition, the Nash Equilibrium of this game isalso the breakdown state of (Ground, Don’t Fund).The intuition for this result is that if neither NASAnor Congress is interested in maintaining humanspaceflight, it will be cancelled.

E. Analysis of Signals: Imperfect InformationThese four examples, when taken together, may provide some insight into why NASA has in the past been

unwilling to carry out a deterrent threat. Suppose that NASA does indeed value human spaceflight highly. In thiscase, Congress possesses the threat power to restrict NASA funding regardless of the Congressional valuation.Without complete information, NASA would be unaware of Congress’ valuation of human spaceflight. In this case,NASA takes the risk of attempting to carry out an incredible threat that might result in the loss of human spaceflightcapability. NASA always has an incentive to act in accordance with its valuations. Therefore, NASA’s actions sendCongress a signal indicating NASA’s valuation of the human spaceflight program. On the other hand, in the absenceof a threat from NASA (or exogenous forces, such as a Presidential veto threat), Congress will seek to reducefunding, regardless of its valuation of the Shuttle program. Therefore, any signal sent by Congress is ambiguous,relegating NASA to a position wherein they do not possess sufficient information. If NASA knew Congress’valuation for the Shuttle, and that valuation was high, NASA might be able to “bluff” Congress. If Congress’valuation of the Shuttle Program is low and NASA tries to make a threat while incorrectly assuming that Congress’valuation is high, Congress will “call NASA’s bluff”, resulting in the grounding of the Shuttle fleet. NASA, lackingcorresponding information about Congress and possessing a high valuation of the Shuttle program, will forego thethreat and accept a lower level of funding.

F. Implications for Political SustainabilityThese results may be used to define a politically sustainable funding level in terms of the parameters of the

game. As we have seen previously, the requested funding level, X*, serves as the threshold point around whichCongress makes its decision to fund or not to fund NASA. This value of X* turns out to be critical in definingCongress’ preferences, assuming that Congress has a certain valuation, P, of human spaceflight capability. As inauction theory, one may think of P as Congress’ reservation price; the maximum amount that Congress would bewilling to pay in order to maintain the Shuttle program. If X*<P, that is equivalent to saying that Congress possessesa high enough valuation of the Shuttle program to provide that level of funding under threat. In this case, NASA,knowing that Congress’ valuation of the Shuttle program is high, could reasonably request more funding(presumably until the funding request exceeds P) by employing a deterrent threat. NASA, in the role of technicalexpert, could make a case for the need for more resources. Conversely, if X*>P, then Congress’ valuation of theShuttle program is lower than what NASA requests. In this case, challenging Congress for more funding would, atbest, be ineffective and at worst would lead to the termination of the program. Therefore, P is an upper bound on theamount of funding that NASA might expect to receive in any given year. In effect, P is the maximum value at whichfunding may be sustained, or the politically sustainable funding level. As mentioned previously, NASA hasimperfect information regarding Congress’ valuation. If NASA were able to correctly elicit the value of P fromCongress, this would effectively turn the game into one of perfect information. Therefore, future work in this areashould focus upon a method for correctly determining P on a yearly basis. Congress is not likely to considerNASA’s budget alone, but will generally weigh it against other programs, further complicating matters [34]. Anycalculation of P must take these complicating factors into account.

Figure 6: The normal-form matrix representing Game 4.The Nash Equilibrium is circled.


III. ConclusionPolitical sustainability is intimately tied with goals, values and interests. In particular, a program will be

sustained if it is delivering value to the stakeholders who are contributing the resources necessary to keep it going.Value delivery is a necessary condition, but it is not a sufficient condition. This is particularly true in situations inwhich there are limited budgetary resources and many worthy goals to address with those resources. Such a situationis encountered on a regular basis by any number of government programs attempting to obtain federal funding fromCongress. The above model demonstrates that a program is unlikely to receive additional funding if it is perceivedthat the program can maintain a consistent pattern of operation without it. In particular, this model examines thesituation in which NASA provides Congress (and by extension, the American people) with human spaceflightcapability. Until that capability is put under threat, Congress is unlikely to support a funding increase for humanspaceflight, and may even reallocate funding in the face of more pressing concerns. The intuition for this conclusionis that Congress, already receiving human spaceflight capability at a lower funding level, may tend to take it forgranted. Put simply, Congress will not pay more to receive what it is already getting. If, however, there is aperception that the capability is under threat, Congress will be willing to provide support up to the point where adetermination is made that the benefits no longer outweigh the costs. In order to maintain political sustainability, asuccessful case must be made on a yearly basis for why funding is necessary to accomplish certain objectives. If theobjectives of NASA do not coincide with those of Congress, a dearth of funding will result. Future work shouldtherefore focus upon determining the goals and interests of Congress, and indeed, of all stakeholders involved in theresource allocations process, in order to determine their crucial needs and how best to fulfill these such that theprospect of their loss is sufficient to encourage sustained support.

AcknowledgmentsThe authors would like to kindly thank the following people for their insights and feedback on this research:

Prof. Asu Ozdaglar, Jason Bartolomei, Sandro Catanzaro, Evdokia Nikolova, Charlotte Mathieu, Matthew Silver,and Nidhi Sharma. In addition, the authors would like to thank the MIT Department of Aeronautics and Astronauticsand the MIT Technology and Policy Program for their financial and academic support.

References1. Bush, G.W., The Vision for Space Exploration, NASA, Editor. 2004.2. Brams, S.J., Theory of moves. 1994, Cambridge [England]; New York, NY, USA: Cambridge University Press. xii, 248

p.3. Brams, S.J., Negotiation games: applying game theory to bargaining and arbitration. Rev. ed. 2003, London; New

York: Routledge. xxvi, 297 p.4. Davis, O.A., M.A.H. Dempster, and A. Wildavsky, A Theory of the Budgetary Process. The American Political Science

Review, 1966. 60(3): p. 529-547.5. McCurdy, H.E., The Space Station Decision: Incremental Politics and Technological Choice. 1990: The Johns Hopkins

University Press. 312.6. Hearing on Human Spaceflight, in U.S. Senate Committee on Commerce, Science and Transportation: Subcommittee

on Science and Space. 2005: Washington, DC. p. 16.7. Wertz, J.R. Economic Model of Reusable vs. Expendable Launch Vehicles. in IAF Congress. 2000. Rio de Janeiro,

Brazil.8. Simon, H.A., On the Concept of Organizational Goal. Administrative Science Quarterly, 1964. 9(1): p. 1-22.9. Gehman, H.W., Jr., et al., in The CAIB Report. 2003, Columbia Accident Investigation Board.10. Maslow, A.H., A Theory of Human Motivation, in Motivation and Personality, A.H. Maslow, Editor. 1970, Harper &

Row: New York.11. Van Dyke, V., Values and Interests. The American Political Science Review, 1962. 56(3): p. 567-576.12. Griffin, M.D., Statement of Dr. Michael D. Griffin Administrator National Aeronautics and Space Administration, in

Committee on Commerce, Science, and Transportation:Subcommittee on Science and Space, S. Subcommittee onScience and Space Committee on Commerce, and Transportation United States Senate, Editor. 2005: Washington, DC.p. 7.

13. Hearing on Fiscal Year 2006 NASA Appropriations, in Committee on Appropriations: Subcommittee on Commerce,Justice and Science. 2005: Washington, DC. p. 33.

14. New NASA Boss Pushes to Replace Shuttle. Technology News [HTML] 2005 May 3 [cited 2005 July 4]; Availablefrom: http://www.technewsworld.com/story/43090.html#.

15. Lawler, A. (2005) NASA May Cut Shuttle Flights and Reduce Science on Station. Science Volume, 540-54116. Sietzen, F., Jr. and K.L. Cowing. NASA Studying Unmanned Solution to Complete Space Station as Return to Flight

Costs Grow. spaceref.com [html] 2005 July 24 [cited 2005 July 25]; Available from:http://www.spaceref.com/news/viewnews.html?id=1052.


17. DeLay, T. DeLay: NASA Funding Fulfills President's Space Vision; House Passes NASA Funding Bill by Vote of 418-7. spaceref.com [html] 2005 June 17 [cited 2005 July 4]; Available from:http://www.spaceref.com/news/viewpr.html?pid=17159.

18. House Science Committee. House Overwhelmingly Passes NASA Authorization Bill. spaceref.com [html] 2005 July 22,2005 [cited 2005 July 22]; Available from: http://www.spaceref.com/news/viewpr.html?pid=17476.

19. House Science Committee Democratic Membership. Bipartisan Compromise Yields Positive Results for NASA.spaceref.com [html] 2005 July 22, 2005 [cited 2005 July 22]; Available from:http://www.spaceref.com/news/viewpr.html?pid=17475.

20. DeLay, T. Delay: NASA Prepared to Make Next Giant Leap; House Passes NASA Authorization Bill by Vote of 385-15.spaceref.com [html] 2005 July 22, 2005 [cited 2005 July 22]; Available from:http://www.spaceref.com/news/viewpr.html?pid=17478.

21. Hutchison, S. Senators Hutchison, Nelson Urge Importance of Return to Flight. spaceref.com [html] 2005 July 21,2005 [cited 2005 July 22]; Available from: http://www.spaceref.com/new/viewpr.html?pid=17458.

22. United States Senate. Floor Statements by Senators Huchison (sic) and Nelson Regarding "Sense of the SenateRegarding Manned Space Flight". spaceref.com [html] 2005 July 22, 2005 [cited 2005 July 22]; Available from:http://www.spaceref.com/news/viewsr.html?pid=17433.

23. NASA Authorization FY 2006-2010. 2005.24. Sietzen, F., Jr. and K.L. Cowing. NASA and White House Discuss Early Shuttle Fleet Retirement. spaceref.com [html]

2005 July 13 [cited 2005 July 25]; Available from: http://www.spaceref.com/news/viewnews.html?id=1048.25. Dunn, M. NASA Pessimistic About Solving Debris Woes. space.com [html] 2005 July 28 [cited 2005 July 28];

Available from: http://www.space.com/missionlaunches/ap_rtf_griffin_050728.html.26. Jones, B.D., F.R. Baumgartner, and J.L. True, Policy Punctuations: U.S. Budget Authority, 1947-1995. The Journal of

Politics, 1998. 60(1): p. 1-33.27. Goertz, G. and P.F. Diehl, The Initiation and Termination of Enduring Rivalries: The Impact of Political Shocks.

American Journal of Political Science, 1995. 39(1): p. 30-52.28. Summary of FY2004 Budget Request. 2004, National Aeronautics and Space Administration: Washington, DC. p. 26.29. Summary of FY 2005 Budget Request. 2005, National Aeronautics and Space Administration: Washington, DC. p. 30.30. Reinert, P., White House draws line on NASA's 2005 budgetBush vows veto for plans with too much or too little, in Houston Chronicle. 2004: Washington DC.31. VA/HUD Subcommittee's report on HR 5041. 2004.32. Weldon, D. Congress Fully Funds NASA - $16.2 Billion Funds Shuttle, ISS & Clears Way for Vision to Start.

spaceref.com [html] 2004 November 20 [cited 2005 July 5]; Available from:http://www.spaceref.com/news/viewpr.html?pid=15543.

33. Chairman Lewis Announces Major Reorganization of the House Appropriations Committee and Slate of SubcommitteeChairmen. spaceref.com [html] 2005 February 9 [cited 2005 July 5]; Available from:http://www.spaceref.com/news/viewpr.html?pid=16124.

34. Johnson-Freese, J., Congress and Space Policy, in Space Politics and Policy, E. Sadeh, Editor. 2003, Kluwer AcademicPublishers: Boston. p. 79-103.

Documents

[American Institute of Aeronautics and Astronautics Space 2005 - Long Beach, California ()] Space 2005 - Political Sustainability in Space Exploration: A Game Theoretic Approach