By Michael D. Griffin
March 2007
“Prediction is difficult, especially the future,” said quantum physicist Niels Bohr, and no one has since captured the underlying concept quite so cleverly. But having been foolish enough to accept the challenge of speculating upon where the next 50 years will take us in human space exploration, the first question to be answered is, where to begin? What is the global view that can best shape our thinking? It is so very easy to be completely wrong, since a variety of radically different futures in spaceflight can be presumed with equal apparent credibility today.
For example, it might be that, after completing the construction of the International Space Station (ISS) and retiring the Shuttle, the excitement inherent in a new reach outward from low Earth orbit (LEO) will appeal to the next generation, leading to a vigorous, technology-driven program, a plan to reach the surface of Mars by the late 2020’s, and the will to sustain and build upon that early presence. Or, interest in human space exploration could once again be motivated by competition among spacefaring nations, leading to a modern version of the “space race” of the 1960s, producing substantial progress but for reasons unsustainable in the longer term. It could be that the unchecked growth of entitlements, the generational commitment of resources necessary to combat terrorism, and a continued downward trend of interest by American students in mathematics, science, and engineering education, will combine to make the civil space program as we have known it irrelevant to the lives of our grandchildren’s generation. Or the truth could lie in some other direction entirely; maybe human spaceflight in the next 50 years will be dominated by tourism, first suborbital, then orbital, with relatively little in the way of independent government activity.
The one thing of which we can be certain is that in trying to envision the world of 2057, two generations in the future, we will be wrong. We will be wrong in our assumptions about the larger context of world culture and civilization in which space exploration exists, and we will be wrong even in the narrower context that is the subject of our attention here. Even the most cursory review of some of the key events in the development of spaceflight shows the need for great humility by anyone writing an essay on the likely state of space exploration in 2057.
For example, who would have supposed in early 1957 that the Soviet Union, and not the United States, would loft the first artificial satellite into Earth orbit, the first robotic lunar probe, and the first man, into space? And who would ever have predicted that the United States, stung by losses in a competition in which it had not even known it was engaged, would, or even could, respond by carrying out the first lunar landing eight years and two months after declaring the goal? Most then-knowledgeable observers believed that such a feat was unlikely to be achieved much before the end of the 20th Century, if then. Not even the most visionary of hard science fiction authors – Asimov, Clarke, Heinlein – imagined that it could occur as early as 1969. And then, having spent $21 billion (in mid-60s dollars) to develop the transportation system to make such a thing possible, was it even conceivable that such hard-won capability would be utterly discarded within a few years? Who would have imagined it? And yet it happened.
With those thoughts in mind to encourage an appropriate humility, it is nonetheless natural to wonder how we might develop a vision of the future that is the least likely to be terribly wrong. How can we extrapolate today’s world in such a way as to avoid the most outrageously wrong predictions?
Most of the present discussion will focus on the U.S. government civil space program. I will have some comments on the international scene and on the possible role of commercial space, but for much of the next five decades, the U.S. government will be the dominant entity in determining the course of human space exploration. We will, I hope, develop robust international partnerships that will enormously enhance the value of space exploration. And we must do everything possible to provide an accepting environment for commercial space entities, standing down government capability in favor of commercial suppliers whenever it becomes possible to do so. But with that said, the U.S. today is spending more than twice as much on civil space, per capita, as any other nation, and I believe this situation is unlikely to change significantly for some time. Commercial space firms offer great promise but, so far, limited performance. For a while yet, it is the U.S. government, through NASA, that determines the main course of human spaceflight.
Of course, manned spaceflight is broader than exploration, and over the next decades it is to be expected that other entities besides the U.S. government, both commercial and international, will be conducting human spaceflight activities. A spacefaring civilization cannot be the realm only of government employees and government sponsored engineers and scientists, though a bias toward such groups is clearly one attribute of a frontier activity. But if we understand that broader participation is desirable, U.S. human space exploration programs can be conducted so as to encourage, rather than minimize, such. Doing so will, in my opinion, be a key to its survival and prosperity, a point that I will make again in what follows.
But let us now focus our attention on more specific matters. The most straightforward extrapolation is to assume that the future will, on average, be much like the past in regard to key assumptions. Since no aspect of government civil spaceflight is more crucial than the funding allocated to it, let’s consider NASA’s funding history for the last 50 years, and try to make a reasonable yet conservative projection as to what we might receive in the next 50. And then let’s consider what that funding might allow us to do, setting aside unforeseeable political upheavals. To understand where we might go, we must understand where we have been, and I think we need a better understanding of our history than is commonly the case.
Any assessment of historical or projected budgets necessarily must be done in constant, inflation-adjusted dollars. This fact leads inevitably to the question of what inflation index should be used, because long-term assessments are sensitive to that choice. Many choices are possible; the Bureau of Labor Statistics maintains the familiar Consumer Price Index (CPI), applicable to the U.S. economy at large; i.e., the Gross Domestic Product (GDP). However, the CPI is not the best measure of inflation for government spending, primarily because the “market basket” of goods and services applicable to the private and public sectors of the economy are very different. The best use of the CPI in connection with government programs is in the estimation of the constant-dollar “opportunity cost” of government activities to citizens. Government services are purchased by taxpayers with CPI-adjusted tax dollars; money paid in taxes is money not available to consumers to purchase other goods and services.
The Office of Management and Budget (OMB) publishes several inflation indices applicable to different portions of the government sector. For government R&D activities, including those at NASA, the OMB prescribes the use of the so-called “GDP (chained) Price Index” (http://www.whitehouse.gov/omb/budget/fy2008/sheets/hist10z1.xls). Without delving into the merits and shortcomings of various indices, our discussion of inflation-adjusted NASA funding will employ this index. While fiscal analysis across several decades is sensitive to the choice of inflation index, the present discussion is not significantly influenced by the choice of the GDP chained index vs. other OMB indices.
Unless specifically stated otherwise, all fiscal discussions in this essay are couched in terms of Fiscal 2000 dollars, with inflation adjustments according to the OMB GDP (chained) Price Index.
Figure 1 shows the constant-dollar budget for NASA’s first 50 years, 1959-2008, in Fiscal 2000 dollars, and includes the assumption that the agency will be funded in Fiscal 2008 at the level of the President’s request. Data for other fiscal years is historical. The anomalous funding bump in Fiscal 1977 is due to the inclusion of a fifth “transition quarter” in that year, since in 1976 the fiscal year boundary was shifted from 1 July to 1 October, where it remains today. Major events in NASA’s history – the “Apollo Peak”, the post-Apollo aerospace depression, and the supplemental provided by the Congress in response to the Challenger disaster, are all clearly visible in Fig. 1. (Click on image to see full size.
As seen, NASA today is funded at a constant-dollar level slightly higher than the agency’s historical average. With proposed growth in the President’s budget for Fiscal 2008-12 roughly matching the anticipated rate of inflation over the next several years, agency funding is expected to remain slightly above the 50-year average.
In an attempt to offer a reasonable, but conservative, vision for government civil space activities, let us assume that NASA continues, in Fiscal 2013 and beyond, to be funded in constant dollars at the average level of the President’s request for Fiscal 2008-12. This is illustrated in Figure 2, with the average out-year budget assumed to be $14.2 billion in Fiscal 2000 dollars. We in the space community will certainly hope for more, but we should not expect less. More properly, we should expect to perform in such a manner – actually delivering a bold, exciting, efficient and effective space program, instead of PowerPoint charts with hopes and dreams – that policymakers do not want to provide less!
The year-to-year budget profile will show some variability, of course, but we should expect considerably more strategic and fiscal stability than was evidenced in the agency’s first few decades. Minor annual variations should not affect the larger picture; on the five- to 15-year cycle of developmental space programs and projects, it is the average level of funding which is the most significant parameter. The total funding received by the agency over a significant period, a decade or more, together with stability of strategic goals, largely determines what can be accomplished.
Figure 3 offers a different view of historical and projected NASA funding for the past and future 50 years. Funding is aggregated by decade, and incorporates the assumption of a stable constant-dollar budget embodied in Figure 2. Figure 4 provides a similar view, with funding aggregated in 15-year intervals and constant inflation-adjusted funding assumed through 2063. This 15-year assessment period is particularly convenient, since essentially all Mercury, Gemini, Apollo and Skylab development and operations are captured within the first 15 years of NASA’s history.
Figures 3 and 4 offer what might be a new perspective for many. From a decadal viewpoint, the “Apollo peak” in NASA funding, regarded by so many as the agency’s halcyon period, is a myth. In truth, NASA received funding well above its historical average level for only five years, 1964-68, followed by a lengthy and debilitating reduction. But when averaged over decadal or fifteen-year time scales, the nation’s civil space program has experienced no particularly noteworthy funding peaks. The highest historical funding period was actually in the decade (or 15-year interval) centered on the early 1990s, not during Apollo. Further, if we assume funding stability in constant dollars as shown in Fig. 2, the total in every subsequent decade will match that of the Apollo development decade, 1959-68. Expressed in a slightly different way, NASA could carry out a complete Apollo-scale effort every 15 years between the present day and the 100th anniversary of Sputnik.
Let us now address another time-honored belief about the Apollo era. When we talk about an “Apollo-scale effort,” it is important to understand that, contrary to conventional wisdom, we are not talking about an agency devoted exclusively to human exploration. The funding record clearly shows that the “Apollo era” was actually quite a lot more than just that.
In the Apollo development decade of 1959-68, human spaceflight received 63% of the budget. Funding specifically for Apollo from its inception in Fiscal 1961 to its completion in Fiscal 1973 was about $105 billion in Fiscal 2000 dollars. If Mercury ($1.9 billion), Gemini ($5.1 billion) and Skylab ($12 billion) are included, the entire human spaceflight program from 1959-73 received about $125 billion, or 61% of the $206 billion allocated to NASA during this period. Little has changed in this regard; today, the President’s Fiscal 2008 budget request assigns 62% of NASA’s funding to human spaceflight.
The list of achievements in both aeronautics and space science from1959-73 is long and impressive. Aeronautical accomplishments of this era include 199 research flights of the three X-15 rocket planes, the development and flight testing of a half-dozen lifting-body designs, groundbreaking work in computational fluid dynamics, development of the supercritical wing and the digital fly-by-wire flight control system, and (in conjunction with the Air Force) major roles in the XB-70 and YF12A programs. The “Apollo era” was a true golden age for aeronautics research, which was allocated 6% of the NASA budget from 1959-68.
In space science the list of accomplishments is, if anything, even more impressive. The “Apollo era” saw dozens of Explorer missions including the Radio Astronomy Explorer and Atmospheric Explorer series; a dozen Pioneer missions including Pioneers 10 and 11 to Jupiter and Saturn; Ranger 1-9; Surveyor 1-7; Mariner 1-10; the Orbiting Solar Observatory, Orbiting Geophysical Observatory, and Orbiting Astronomical Observatory series, as well as most of the money for two Viking missions to Mars, launched in 1975. The TIROS, NIMBUS, and ESSA series pioneered the development of weather satellites. The “Apollo era” was also a golden age for space science, which received 17% of the NASA budget from 1959-68.
About 10% of the 1959-68 budget was devoted to space technology development, including space communications technology, and 4% was devoted to “Other”; i.e., university support and cross-agency activities.
The summary below shows a “then and now” comparison. In contrast to oft-repeated claims, human spaceflight is not growing relative to other portions of the NASA portfolio, and is not “eating everyone’s lunch.”
Category 1959-68 FY08 Request
Human Space Flight 63% 62%
Science 17% 32%
Aeronautics 6% 3%
Comm & Space Tech. 10% 0%
Cross-Agency Supt. 4% 3%
The historical record provides clear evidence that it is possible to have robust, co-existing programs of human exploration, space science, aeronautics, and technology development in a single agency funded at a level essentially the same as we presently receive. So, what might the future offer?
Let us assume for the present discussion that over the long term, manned spaceflight will continue to receive 62% of the NASA budget. Again assuming inflation-adjusted funding at $14.2 billion/year on average, it follows that human spaceflight will be allocated $8.8 billion annually, or $132 billion in each 15-year period, in Fiscal 2000 dollars.
Next, we must recognize that “the future” really does not, and cannot, start until after 2010. Until then, we are engaged in completing a long-standing commitment to the International Space Station, with no other option besides the Space Shuttle to do it. At present funding levels, we cannot afford to develop new human spaceflight systems without the money which becomes available following Shuttle retirement.
Despite the concerns of those – emphatically including myself – who worry about the gap in human spaceflight between the retirement of the Space Shuttle and the availability of the new Constellation systems, Orion and Ares, we must stay on our present course and retire the Shuttle in 2010, if there is to be a future for human spaceflight. The Shuttle offers truly stunning capability, greater than anything we will see for a long time, but the expense of owning and operating it, or any similar system, is simply too great. Any new system, to be successful, must offer a much, much lower fixed cost of ownership. The Space Shuttle was designed to be cost effective at a weekly flight rate, a goal that was never credible, if for no reason other than the fact that the funding for so many payloads to fly on it was never remotely available. And, if there were a predictable requirement for 50-60 government-sponsored payloads to be flown annually, that fact should be treated as a market opportunity for a private, not government, space transportation enterprise. A government human spaceflight system must be designed to be cost effective at the half-dozen or so flights per year that we can expect to fly.
But, if the bad news is that “the future” doesn’t start until after 2010, the good news is, that is only four years away. And in the 45 years thereafter, by the centennial anniversary of Sputnik, we can expect to receive at least as much money as was necessary for Apollo, three times over. And despite the limited funding for Exploration in today’s NASA budget, we will have a bit of a head start, because we’re making considerable progress toward the deployment of Orion and Ares, even while flying out the Shuttle/ISS manifest. So what will we do with this money?
Most of the next 15 years will be spent re-creating capabilities we once had, and discarded. The next lunar transportation system will offer somewhat more capability than Apollo. It will carry four people to the lunar surface instead of two, and for a minimum duration of a week, rather than a maximum duration of three days. But in all fairness, the capabilities inherent in Orion, Ares I, and Ares V are not qualitatively different than those of Apollo, and certainly are not beyond the evolutionary capability of Apollo-era systems, had we taken that course. But we did not, and the path back out into the solar systems begins, inevitably, with a lengthy effort to develop systems comparable to those we once owned. It will cost us about $85 billion in Fiscal 2000 currency to get to the seventh lunar landing by 2020.
The above assessment is, for many, a bitter pill to swallow. Not only is it depressing for advocates of human exploration to face the fact that so many years will be spent plowing old ground, but there is also the question of why it will take so long. Again, the answer is captured in the funding profile. We are indeed receiving today, in any given 15-year period, the same real-dollar funding as in the 15 years of the Apollo era, but we are not receiving it on the same schedule. The brief, enormous, funding peak of mid-1960s allowed the Apollo systems to be developed and procured in parallel. Today’s systems must be developed serially. And that is why the job will not be done, this time, in eight years. But that is also why we will not incur the disastrous divestiture of talent and technology that occurred in the 15 years after Apollo, between the early ‘70s and the late ‘80s.
In the long run, to return to the Moon or go to Mars and beyond, stability is to be valued more than going in the shortest possible time. As we move forward into our next 50 years, this must be fully understood by both policymakers and the public, or we will forever be answering the question as to why we work so slowly compared to the Apollo generation. Civil space exploration beyond LEO must have the stability in strategy and funding that was lacking the first time around. This will only be provided by policymakers if a clear link is established between predictable results and predictable purpose, strategy, and funding. I believe we will succeed in forging this new paradigm – the opposite of the Apollo “man, Moon, decade” paradigm – but we must devote considerable attention to doing so.
What will be done with the lunar transportation capability that is being developed? By 2020 we will have this capability, and with it choices to make. We can choose between a lunar program devoted to sortie missions, or one devoted to building up a lunar outpost. And we can choose between the level of effort we intend to focus on lunar activities vs. initiating development for Mars missions. In company with other space agencies around the world, we at NASA have focused on an outpost-centered lunar exploration strategy. I believe this will be preferred over a sortie-only strategy for the reasons that it provides a much more effective avenue for international partnership, and because it provides the greatest opportunity to learn on the Moon what we need to know to go to Mars. But, of course, nothing prevents a sortie mission to any location on the Moon that is of sufficient interest to justify the expenditure of funds. So again, let us look at what is fiscally possible.
It is to be hoped and, I believe, expected that the next era of space exploration will be international in scope, in much the same fashion as the development of the International Space Station today. Whatever might be said of the ISS program – and there cannot be much that has been left unsaid – it has pioneered a path to the development of a major international space facility. There are lessons learned in so doing that we will take with us out into the Solar System. These lessons will be the most enduring, and ultimately most valuable, contribution the ISS can make. We will be applying them on Mars, fifty years from now.
The United States is developing the transportation system which will allow access to the lunar surface for the first time in a half-century. This is the highest “barrier to entry” for exploration beyond LEO, one which essentially exhausts the contribution that we can make to a lunar outpost in the next 15 years. If there is to be a lunar presence significantly beyond merely getting there and getting back, if there is to be a human tended outpost, much of the early capability must be developed by international partners. But outpost sustainability, at least in the early years, will largely depend upon Orion and Ares.
I believe that by 2021-22 we will have regained enough experience in lunar spaceflight operations that we will be able to undertake a modest, but sustained and sustainable, program of lunar outpost development and utilization. I will also venture to say that by 2022 the ISS will be definitely behind us. We will have learned from it what we can, but there will come a time when the value of the work being done onboard the facility will be judged not to be worth the cost of sustaining its aging systems, and it will be brought down. I don’t know when this will occur, and I am not sure it is predictable other than in a statistical sense, but I believe that by 2022 or thereabouts it will have happened. And when it does, the resources which have been used for ISS support can be applied to the support of a lunar outpost.
For the sake of argument and nothing more, let us say that in 2022 we will begin a sustained lunar program of exploration and development consisting of three manned missions (two outpost crew rotations and one sortie) and one unmanned cargo mission per year, utilizing three Orion/Ares I vehicles and four Ares V launches. Present projections assume a cargo capacity of six metric tons on a lander carrying four crew members, and twenty metric tons on a cargo lander, at a marginal cost of about $750 million for a human mission and $525 million for a cargo mission. The marginal cost in Fiscal 2000 dollars for this nominal lunar program will thus be about $3 billion.
These marginal costs do not include an allocation of the fixed costs of production and operations which will be assigned to each flight. Let us assume a fixed-cost support base of $1 billion annually, about a third of that for the Shuttle today, equivalent to roughly 6,000 full-time employees at average Fiscal 2000 labor rates. We should all work to make it much less, but this is an appropriately conservative estimate for the present. This yields a sustained lunar program costing no more than $4 billion/year, leaving $4.8 billion annually in the human spaceflight account to be applied to new development priorities.
By the 2020’s we will be well positioned to begin the Mars effort in earnest. The lunar campaign will have stabilized; a human-tended outpost will be well established; we will have extensive long-duration space experience in both zero- and low-gravity conditions, and it will be time to bundle these lessons and move on to Mars – which does not imply that we will bring lunar activities to an end. Quite the contrary; my prediction is that the Moon will prove to be far more interesting, and far more relevant to human affairs, than many today are prepared to believe. But by the early 2020s, it will be time to assign a stable level of support for lunar activities, and set out for Mars.
The development of the Orion/Ares I/Ares V transportation system is being done in a way that provides a substantial capability for subsequent Mars expeditions. In particular, we expect the Orion crew vehicle (or a modest upgrade of it) to provide the primary transportation from Earth to whatever transportation node is used for the assembly of the Mars ship, and to be the reentry vehicle in which the crew returns home at the end of the voyage. The Ares V cargo vehicle will provide, with no more than a half-dozen launches, the 500 metric tons or so which is thought to be necessary for a Mars mission, based on present-day studies. As a perspective on scale, this mass is about 25% greater than that of the completed ISS.
It is difficult to estimate the non-recurring cost of developing a Mars mission that is initiated some 20 or more years in the future, and especially so when a specific mission architecture has not yet been formulated. But reasoned estimates can be made. A small group co-chaired by Skylab and Shuttle astronaut Owen Garriott and me made an attempt to do so in a study conducted for The Planetary Society in 2004. While necessarily omitting many important details, a reasonable approach based on mission mass, consistent with modern cost estimation algorithms, was outlined. It was concluded that, following a decadal hardware development cycle, nine Mars missions could be conducted over a 20-year period for a total cost of approximately $120 billion in Fiscal 2000 dollars, or $6 billion/year, significantly less than we are spending on Shuttle/ISS today. (If this seems low, it should be noted that the development cost of the heavy-lift transportation system is allocated to the earlier lunar program. The Mars program would pay only the marginal cost of transportation.)
Allocating an across-the-board 30% reserve at this stage puts the cost of a 30-year Mars exploration program at $156 billion in Fiscal 2000 dollars. Of this, approximately $70 billion consists of development cost, with reserve. If $4.8 billion/year is available in the human spaceflight account, then the Mars mission development cycle will require about 15 years. Thus, if we begin development work in 2021, we will be able to touch down on the Martian surface in about 2037, with follow-on missions every 26 months thereafter for the next two decades.
So there we have it, at least for the U.S. civil space program. At present levels of real-dollar funding, by 2057 we can celebrate the 35th anniversary of a lunar base, which will be growing in capability at the rate of 30 metric tons per year, even without assuming any International Partner contribution to logistics, which I believe is overly conservative.
We can celebrate the 100th Sputnik anniversary in conjunction with the 20th anniversary of the first human Mars landing. And we can do all of these things even with what I would consider the pessimistic assumption that we receive no more money, in constant dollars, than we do today. Indeed, there should be money available for missions to interesting Near-Earth Objects, a separate challenge which we will come to understand offers huge opportunities for those seeking to develop a spacefaring civilization.
That’s what I see ahead for the American space program. What about the rest of the world? Both Russia and China have domestic human spaceflight capability today; indeed, the ISS program would be in very difficult straits without Russian crew and cargo services. Other nations or alliances – Europe, Japan, India, Brazil, others – could develop similar capability within a few years of a decision to do so. For advanced nations today, possessing the capability for human spaceflight to LEO is a political, not a technical, decision. But going beyond LEO, to the Moon, is a problem of a different order. And yet, the Moon is a necessary first step outward for any nation seeking a spacefaring future. So let us look at the resources required to pursue such a future.
The development phase of Apollo required about $80-85 billion in Fiscal 2000 currency, about the same as we predict will be required to redevelop similar capabilities. Constellation systems will, as stated earlier, offer substantially more performance than Apollo, but it does seem as if an effort of approximately this magnitude is necessary, no matter what. There is an inherent “knee” of the cost vs. performance curve; it takes a lot of effort to get to the Moon, after which additional capability can be added at somewhat less marginal cost.
So let’s assume a minimum required effort of about $80 billion is required to develop a basic lunar capability. In the U.S., at approximate average aerospace labor rates for Fiscal 2000, this is equivalent to an effort of roughly 600,000 man-years, or 40,000 people for 15 years. Other nations will likely operate in a somewhat “leaner” fashion than is characteristic of the U.S. aerospace culture; I will always remember Max Faget’s comment to me that “we could have done Apollo with a lot fewer people, but we couldn’t have done it with any more.” But it remains likely that an effort similar to Apollo will be required for any nation or society attempting to reach the Moon for the first time, provided it has access to the necessary industrial base and an adequate workforce.
Many nations or alliances can, as a matter of political choice, decide to mount such an effort. Europe has a population 50% greater than that of the U.S., yet spends on a per-capita basis only about a fifth of what we spend on space. A future European generation could choose to do otherwise. India has a middle class population equal in size to the entire U.S. population, and produces engineering graduates equal to the best anywhere. Chinese space agency representatives have remarked publicly that, today, some 200,000 engineers and technicians are engaged in space-related work. And of course Russia could begin the development of a lunar transportation system today, essentially at its discretion, given its existing spaceflight capability and the recent and continuing flow of energy money into that country.
By the mid-to-late 2020’s, at the latest, several nations will have the independent capability to reach the Moon, and will be doing so. My hope is that the various programs can be bent more toward a cooperative than a competitive agenda. I believe that nations will find it to be in their interests to cooperate in lunar exploration and development, as they do in Antarctica today. But it will also be true that each nation to develop key elements of space infrastructure, especially transportation but also navigation and communications assets, will be unlikely to set them aside in favor of reliance on others. For the next generation, maybe as much as two decades, the U.S. may well be the only nation capable of reaching the Moon on its own. But much beyond that, and I suspect that we’ll be there with others. The Moon will be within the grasp of a significant number of advanced nations. It will be the next big leap, a voyage to Mars, where international cooperation is a requirement, rather than an option.
What will be the role of commercial space entities in human exploration? By “commercial space,” I mean space business enterprises which develop a marketable capability while dealing at “arms length” with the government; i.e., largely without the financial backing and close government supervision which has historically characterized the space industry. The government will, at least initially, still be the major customer for such enterprises. Whether or not an enterprise is part of the commercial space arena depends not on the identity of its customers, but on the nature of its interactions with that customer.
I expect that the role of commercial space in human space exploration will be significant, and possibly transforming, over the next five decades and beyond. We at NASA are presently engaged in an effort to determine whether it is possible for a commercial firm to develop orbital space transportation capabilities without the close supervision of the government. The latter approach, through what are commonly known as “prime contracts” with industry, has been the traditional approach over the last five decades for state-of-the-art projects in the defense and aerospace industry. It produces successful outcomes with reasonable certainty and at great expense.
I believe it is obvious to most that, if a desired product lies within the state of the art, it can be provided with substantially greater efficiency by the commercial sector than by the government. There is little comparative data obtained under controlled conditions to support this claim or to estimate the efficiency factor involved. But, to me, the limited data and my own experience points to an efficiency factor of three to seven in favor of the commercial sector. Whatever the factor, the likely cost benefit to the government of commercial procurement of space goods and services, once it is possible, cannot and will not be ignored. But, again, the crucial assumption is that the intended product lies well within the state of the art. When this assumption cannot be met, close government involvement will continue to be required. Commercial firms simply cannot be successful if engaged in a research upon whose success their revenue depends.
Some have opined that the scale and difficulty of spaceflight is such that it will remain an inherently governmental enterprise for the foreseeable future. I do not share this view. For me, the question is more properly “when,” not “if,” the state of the art in astronautics will permit a private enterprise to develop a successful orbital transportation capability without the direct support – and the accompanying onerous and expensive oversight – of a government prime contract.
We at NASA are attempting to determine whether this date has in fact arrived. By providing “seed money” in the form of Space Act Agreements for two Commercial Orbital Transportation Services (COTS) entities, we hope to stimulate the attainment of entrepreneurial commercial space transportation. If such capability is successfully demonstrated, we can then procure such services in a manner more characteristic of the economy at large than is the usual case in the government-driven aerospace sector. We at NASA are prepared to stand down government systems as and when commercial capability becomes available.
Whether or not the specific COTS initiative is successful, the commercial space business model will eventually become so. A long-term government sponsored space exploration program carries with it the implicit demand for many tons of cargo logistics and crew transport, offering a stable and tempting market niche for industry. Some enterprises will be surely successful in their attempts to service this market, and from there commercial space activity will bloom. In addition to transportation, space exploration implies the need for communications, navigation, power systems, and other support infrastructure. These requirements will be targeted by specific firms as services to be provided commercially, rather than by government.
I believe that the future for U.S. civil space exploration that I have outlined here can be attained with the resources that will be available to NASA by means of conventional government appropriations and acquisition strategies. But I also believe that this is just about as much as we can achieve with those resources, unless we can effect real changes in our methods of doing business. If we want to do more, if we want a richer future, if we are unsatisfied by the relatively modest program of inner solar system exploration I have envisioned here, there must be a change in how we go about it. Embracing the possibilities inherent in commercial space transactions is one such method.
What else do we have to do to bring about this future? Most of what we need to accomplish the goals set forth here has already been discussed, implicitly or explicitly, in connection with budgetary issues, but it may be helpful to concentrate some attention on the matter.
The most important factor for future success is stability in purpose, strategy, requirements, and funding. Apollo funding was unstable in both directions. The huge rate of early growth allowed the Apollo goal to be met; the abrupt cessation of funding as the goal drew within sight produced strategic damage that remains unto the present day.
To be successful, program managers (whether in government or industry) need stability. Additionally, they need the knowledge that there will be such stability; defensive planning is inherently wasteful.
Stability of purpose, a result of agreement upon priorities, is as important as funding stability. Managers must have reasonable and effective control of what is done with the resources – people, money, and time – entrusted to them. If funding is in fact stable, then additional money will not be available to solve problems which are, inevitably, encountered in any state-of-the-art development program. Managers must have the latitude to sacrifice or defer lower priority efforts in order to protect more important ones. This in turn requires, at a minimum, broad agreement on what those priorities are. When this cannot be obtained, every programmatic overrun and every minor budget variation produces divisive political infighting over what will be sacrificed, and what will not. A common result is that nothing is sacrificed and all programmatic content is preserved, but at a slower pace. This produces an inherent inefficiency in the execution of all programs, resulting in more overruns, etc., in a degenerating spiral. It is difficult, and hugely wasteful, to carry out a program in such an environment.
There is another aspect of stability that is equally crucial to bring about the future outlined here. It involves, once again, a lesson to be gained from the past. This is the absolute necessity of fully utilizing the systems we develop, at huge expense, rather than discarding them in favor of something which is appealing because it is new. This aspect of stability has had a direct impact on NASA’s ability to maintain stability of both purpose and funding for decades.
We must treat our space systems as we have always treated our airplanes. Successful aircraft designs, from general aviation airplanes to the highest-performance military fighters, are evolved, upgraded, and used for decades. Just as with DC-3s, B-52s, and many other aircraft, we need to understand that Orion and Ares will be flown by the grandkids of the first astronauts who take them into space. We simply cannot again afford the strategic distraction, the wasted money, the squandered talent, and the lost time of building a new human spaceflight system, and then using it for only sixteen missions.
Once again, a look at the budgetary history provides a sobering lesson for the future, a sobering view of “what might have been.” Let’s recycle to the early 1970s, a time of budgetary starvation for NASA, a time when we did not yet have the Space Shuttle, but did still have the Apollo systems – the Saturn I-B and Saturn V, the Apollo command/service modules (CSM), the lunar lander, and the Skylab system. All of these things were in existence in 1973, having been created in that seminal first 15 years of our agency’s history.
Make no mistake; these systems were far from perfect. They were expensive to develop and expensive to operate. Our parents and grandparents, metaphorically speaking, did not really know quite what they were doing when they set out to accept President Kennedy’s challenge to go to the Moon. They learned as they went along. But what they eventually built worked, and worked well. And it could have kept working at a price we could afford.
Let’s look at some recurring costs in dollars then and now. All costs include both hardware and mission operations, and are at the high end of the range of possibilities, because they take no advantage of stable rates of production. Fiscal 2000 costs are approximate, obtained by inflating programs in the aggregate, rather than tracking and inflating separate expenditures of real-year dollars.
Element Real-Year $ M FY 2000 $ M
Apollo CSM 50 160
Apollo Lunar Module 120 400
Apollo Lunar Mission 720 2400
Saturn I-B 35 120
Saturn V 325 1100
Skylab Cluster 275 925
Let’s assume that we had kept flying with the systems we had at the time, that we had continued to execute two manned Apollo lunar missions every year, as was done in 1971-72. This would have cost about $4.8 billion annually in Fiscal 2000 dollars.
Further, let us assume that we had established a continuing program of space station activities in Earth orbit, built on the Apollo CSM, Saturn I-B, and Skylab systems. Four crew rotation launches per year, plus a new Skylab cluster every five years to augment or replace existing modules, would have cost about $1.5 billion/year. This entire program of six manned flights per year, two of them to the Moon, would have cost about $6.3 billion annually in Fiscal 2000 dollars. The average annual NASA budget in the 15 difficult years from 1974-88 was $10.5 billion; with 60% of it allocated to human spaceflight, there would have been sufficient funding to continue a stable program of lunar exploration as well as the development of Earth orbital infrastructure. I suggest that this would have been a better strategic alternative than the choices that were in fact made, almost 40 years ago.
After a time, as NASA budgets once again improved, we would have begun to concentrate our lunar activity around an outpost, and we would have used cargo missions to emplace the outpost equipment. A modified Apollo Lunar Module descent stage, with extra fuel and cargo replacing the ascent stage, could have been used for the purpose. The Saturn V could deliver two such vehicles with a single launch. So, over time, we could have built up an early lunar outpost, or smaller ones at different places of interest. By the present day, using what we had with minimal modifications – and I will remind us all that the Soyuz systems of that era are still flying – we would have a vast store of experience and a significant amount of lunar infrastructure. When the civil space budget eventually improved, as it did, we would have been well positioned to begin development of a Mars mission. And in the meantime, without doubt, we would have continued to modify, refine, and incrementally improve the old Apollo designs, to the point where they would have provided greatly enhanced effectiveness by the present day.
If we had done all this, we would be on Mars today, not writing about it as a subject for “the next 50 years.” We would have decades of experience operating long-duration space systems in Earth orbit, and similar decades of experience in exploring and learning to utilize the Moon. This essay on “the next 50 years” would be quite different than the one I am offering here. I think most of us will agree that it would have been a better one.
Now, nothing is as easy as planning in hindsight, nor as permanent as a lost opportunity. I offer the “alternative history” above not to throw stones at policymakers long departed from the scene, but to inform future decisions. If we ignore these lessons, we will surely repeat them.
The vision of the next 50 years in space that I have outlined here is not a flight of fancy. It does not require a course change from present understandings, nor does it require extensive development of costly new technology. It is a logical, incremental, stable, sustainable plan that can be executed with realistically attainable budgets. For these reasons, I believe that it will be done, and done as envisioned here. We really can celebrate the 100th anniversary of Sputnik with the 20th anniversary of the first human landing on Mars. It is up to us to make it so.
-- Michael D. Griffin is administrator of the National Aeronautics and Space Administration.
Watch this space every week for new Mars imagery. Five spacecraft are operating at Mars. They are the Mars Reconnaissance Orbiter (MRO), the Mars rovers Spirit and Opportunity, the Odyssey orbiter and European Mars Express spacecraft. The new Phoenix lander is also set for launch Aug. 3. 
