From L5 News, January-February 1985
(before the Challenger disaster)
Over the past months a major controversy has developed over the price NASA charges commercial customers for launching payloads on the shuttle. On one side are potential private space launch entrepreneurs, who have alleged that US taxpayers are being made to subsidize commercial payloads such as communication satellites, wasting money and stunting the infant private launch industry. Opposing them are shuttle users and NASA, who claim either that there is no subsidy or that the subsidy is beneficial to our future in space.
In this debate, the taxpaying public’s interest focuses on whether tax dollars are being wasted, but L5’s interest focuses on longer-term consequences: how does the alleged subsidy affect our goals of space development and settlement? The gut reaction of many space supporters is that of course we want very low shuttle launch prices, and if a subsidy is required, let’s have one. Yet the superficial appeal of this position should make us suspicious: is it wise to disconnect the price of a service from its true cost? Might this not send false signals to budding space industries which could hurt them farther along? Phillip Salin puts this another way: “Pretending that transportation costs are lower than they are does not encourage space commercialization and space entrepreneurs, but rather malinvestment and premature investment of society’s scarce research funds.”
Space supporters can be found on both sides of the issue. Our basic concern is determining which policy best promotes our goals. Which is more important: holding down the price of commercial experiments on the shuttle, or encouraging the development of cheap private space transportation? In this article we will focus on the latter position because its proponents, being less powerful politically, will have more difficulty getting a fair hearing than will NASA and shuttle users.
Shuttle pricing is a complex issue involving masses of detailed numbers that change from day to day and source to source. However, it seems clear from the data now on record that we may face a serious problem for space development: NASA’s low prices appear to be inhibiting one of our key goals. It is impossible to do justice to either position in so short an article, but we hope to get across the importance of this topic to the future of space development and settlement.
Birth of Controversy
The debate first moved into high gear in early 1984 when Phillip Salin was invited to testify before a Congressional subcommittee. Currently a partner in Venture Acceleration, he is an economist and long-time friend of space development. A cofounder and director of the struggling private launch firm Starstruck, he points out that the views expressed here are his own and not those of the company.
On February 28, 1984, Salin testified before the US House Space Science and Applications Subcommittee of the Committee on Science and Technology. Written records include both his original testimony and his later responses to questions. The following excerpts give a general idea of his position:
“The Space Shuttle is now and is expected to always be much more expensive to operate than originally planned in 1971. No one disputes this. The only argument is over how much more expensive it will be than originally planned.”
“[A]s a result of NASA’s proposed pricing policy, the more cost-effective Delta and Atlas Centaur are being subjected to unfair competition by a taxpayer-subsidized Shuttle. Unless immediate action is taken to change the Shuttle pricing policy, these lower cost competitors may be driven out of business.”
“The long run implications of the Shuttle subsidies are even more serious. By discouraging development of more cost-effective private American satellite launchers, NASA is guaranteeing that the only cost-effective satellite launchers will be operated by foreign corporations; following Ariane, we can expect the Japanese to enter this potentially profitable market also.”
“Once Shuttle prices are raised to reflect actual costs, the Shuttle’s true merits will be easier for everyone to see: it has an obvious comparative advantage for manned missions and for complex research missions. Instead of wasting this unique capability on launching ordinary satellites which can be launched more cheaply by others, NASA should be following its Charter and trying to get the maximum amount of research value out of Shuttle.”
“In 1980, I realized that NASA, for all its sincere interest in promoting innovation and progress in space technology, had also, like AT&T, come to confuse itself with an entire industry: the space transportation industry. Instead of conceiving itself as the midwife that would help private firms like Boeing, Rockwell, General Dynamics, McDonnell Douglas, United Airlines, TWA, Pan Am, or Delta Airlines to enter an industry which already was becoming real and potentially competitive, NASA thought it had to do the job itself. It had begun to confuse the business of research with going into business.”
“If we accept NASA’s own projections of costs, budgets, and expected launch manifest of 53 launches between fiscal 1986 and fiscal 1988, the average cost per Shuttle launch during these 3 years will be approximately $250 million.”
“NASA’s proposed pricing policy for 1986-1988 Shuttle launches is not based on full cost recovery, on average annual cost recovery, on average per launch cost recovery, or on any other methodology comparable to what a businesslike firm would have to charge. Instead, NASA’s proposed pricing policy for the Shuttle is based upon a costing concept of its own design, which it calls ‘out of pocket costs.’ According to NASA’s internal analyses, projected ‘out of pocket costs’ for a full Shuttle launch in 1986-1988 will be, on average, $71 million (in 1982 dollars). NASA proposes to charge as its launch price only this $71 million.” [NASA now suggests $87 million as a post-1988 full cost recovery price.]
“[T]he marginal cost of a Shuttle launch throughout the 1980s will be no less than $150 million, and is more likely to be approximately $200 million.”
Salin describes four ways in which NASA errs with its cost numbers: whole categories of costs not explicitly labeled as “operations” are excluded; labor costs are regarded as fixed regardless of the number of launches; estimated future costs are used to set present prices; and marginal costs rather than average costs are used.
On this last point, he states “In addition to creating permanent requirements for taxpayer funding of operations, marginal cost pricing creates unfair barriers to entry by would-be competitors, barriers which are not in the public interest. When one competitor can price at “marginal” costs, due to access to taxpayer funds to make up the difference between average costs and marginal costs, all other competitors will be forced out of business even if their average cost of providing services is lower than the subsidized service.”
“[E]very time a Shuttle full of ordinary geostationary satellites is launched, perhaps $100 million of taxpayer money is unnecessarily wasted…$100 million is enough money to found a credible private space transportation company, from scratch.”
“For manned missions and activities, it is up to Congress how much subsidy it wishes to make taxpayers pay indefinitely. I really have no answer here. But for unmanned, ordinary satellite launches, I believe the price should be raised as soon as possible to at least $150 million/full Shuttle launch.”
“Between high Shuttle operating costs, and high costs of developing new ‘advanced’ transportation, NASA’s effective scientific and research budget will be squeezed down even further. The US will obtain less research from an agency whose whole purpose is to conduct research, and will instead receive inadequate space transportation service from a government agency trying to act like a business.”
In closing his testimony, Salin stated, “The Shuttle is a wonderful machine, but it is not low cost. Yet lowering the cost of space transportation is the main key to opening the space frontier. The next great breakthroughs in space will be economic breakthroughs, providing much lower transportation costs from Earth to Low Earth Orbit. Over the last two decades, first with Apollo, and now with Shuttle, Americans have made great progress pushing the technical envelope of space transportation; now it is time to focus on pushing the economic envelope, by putting private American companies to work on solving the problem. American industry is now poised and ready to do this, provided it is not forced to fight an uphill battle with government-subsidized competition.”
Responses to Congressional Questions
This controversial testimony stimulated many questions from the Subcommittee, not all of them friendly. Later Salin was given fifteen questions to which he responded in writing. The following excerpts are taken from his replies:
In the Preface he states: “Overall, the questions fail to address the most important issue raised in my testimony: how can a Boeing or a General Motors of space transportation ever emerge as long as Congress allows NASA to have the US taxpayer pay heavy operating subsidies for Shuttle, thereby making private space transportation appear uneconomic?
“Back in the 1940s, the idea of marginal cost pricing was quite in vogue among economists; however, there has been substantial progress over the last 40 years in understanding how and why the attempt to pursue marginal cost pricing causes unresolvable analytic problems, political problems, distorted incentives, and potentially huge resource misallocations. The proper basis for pricing government services in a case like this, as I stated in my testimony, is average cost.”
“[A]ny transitory weakness in the emerging competition between Ariane and Delta/Atlas should be attributed, at least in part, to NASA’s attempt to suppress these vehicles and entice payloads onto the Space Shuttle sooner than was in fact economic.”
“I do not object to NASA’s assuming that Shuttle’s per launch costs will drop over the next few years. I object to their basing their proposed prices first on inappropriate accounting categories, and second on the assumption that the costs for these categories will drop extremely far, extremely quickly, all without providing clear, comprehensive, and publicly reviewable support for these figures. The anti-competitive incentive for NASA to ‘promise’ Congress in 1983 that Shuttle costs will be low by 1988, and then be ‘surprised’ in 1988 when it becomes obvious that they aren’t, should by now be clear.”
“[T]here appear to be reasonable grounds for questioning NASA’s projections regarding future cost-reduction achievements, such as how quickly NASA is likely to be able to achieve 75% reductions in its per launch Shuttle operating costs.”
“Right now, NASA is heavily subsidizing commercial customers. Depending on how you look at it, this subsidy is either coming out of NASA’s research budget or out of the taxpayers’ pocket.”
“And until we raise Shuttle prices to at least $200 million, every time a Shuttle is launched carrying primarily commercial comsats, approximately $100 million of taxpayer money has just been unnecessarily wasted, and the incentive to develop more cost effective private space transportation has been unfairly and disastrously reduced to the long run detriment of this country’s economic leadership in space.”
“If an extra launch on Orbiters 1-4 ‘costs’ only $50 million, why are we considering spending $2 billion for another orbiter? Why not fly the old orbiters more often in each year? (Answer: because, as I have argued, due to rising marginal costs per launch, it is currently costing much more than $50 million to try to obtain an additional launch per year from each of the existing orbiters.)”
“To put it bluntly, responding to the possibility of Europeans dumping on American industry [i.e., subsidizing Ariane] by allowing NASA to do so is not the best way to encourage the development of entrepreneurial American space transportation companies.”
“Artificially encouraging a high load factor with an artificially lowered price, when the price is below average cost, in no way serves to promote economic efficiency. Instead, it promotes economic waste. In particular, it promotes perpetuation of waste: since incentives to develop lower average cost competing methods of service are suppressed by the artificially low prices, more economically efficient technologies are not developed. The old, high average cost technologies remain in use, and remain apparent ‘state of the art’ much longer than would otherwise be the case.
“If IBM’s competition had been suppressed by similar pricing methods to Shuttle’s, with losses made up at taxpayer expense, and if IBM had been allowed to thereby force all computer users to use IBM-360s, in order to increase the efficiency of utilization of IBM computers, indeed, IBM 360 utilization and load factors would have gone up. For a while. But in the longer run, neither minicomputers nor microcomputers would ever have been developed — in America.”
NASA Responds
Soon after the testimony, in March, NASA issued a brief response, stating that Salin had a “basic misunderstanding” of shuttle cost structure and pricing principles and summarizing what it called the “general fallacy” of his analysis. NASA objected to Salin’s including advanced propulsion development, tooling, and orbiter production in his cost estimates — Salin’s response is that the propulsion development is actually shuttle-related R&D, and that it along with tooling and orbiter production constitute major and appropriate costs. NASA stated that Salin amortized costs only through 1988, leading to an artificially inflated cost per flight. Salin answers that he amortized over a longer period.
NASA argued that since shuttle costs are incurred up to three years in advance of flight, and the STS has not reached “steady state” operation, Salin’s method “yields totally invalid results.” Salin’s responds by pointing out that NASA’s figures indicate that 60 percent of costs are incurred in the launch year, with an additional 33 percent in the previous year; additionally, he points out NASA will be near steady-state by 1988. NASA states that Salin included costs for military launches from Vandenberg Air Force Base, which should have been excluded — Salin claims that excluding these costs (and launches) would make only modest changes in his cost-per-launch numbers and no change in his conclusions.
NASA states that Salin did not have the data needed to estimate marginal cost. Salin answers that he underestimated marginal cost, and that marginal cost pricing is an incorrect method. NASA concludes “As a result, it is the commercial and foreign customer who is subsidizing the cost of government launches and not the opposite.” Salin replies that these customers paying a share of overhead costs does not constitute a subsidy. Admittedly, NASA’s response was prepared very quickly.
More Numbers
Subsequent to Salin’s testimony, NASA admitted that its estimates of current launch costs are running at $150-200 million. In contrast, its current price for a payload filling the entire orbiter is only $35 million, scheduled to rise to $71 million late this year. The proposed post-1988 “full cost recovery price” submitted to White House is $87 million.
In September 1984, NASA sent to President Reagan a justification of its proposed pricing policy. Salin states that this justification assumes three separate radical cost drops over several years, none of which are volume-related: a major drop in flight operating costs between FY84 and 85, one in solid booster costs between FY85 and 86, and another in launch operating costs between FY86 and 87.
Salin believes that if these are accomplished, NASA’s variable costs become realistic. But this would require a very steep learning curve.
Statistics on aerospace practice show that costs typically fall by a certain fraction as experience increases by a certain fraction; this relationship is called the “learning curve.” Commonly, doubling experience (such as total launches or units manufactured) makes costs fall by ten or twenty percent, corresponding to a ninety or eighty percent learning curve.
Eric Drexler notes that NASA, to meet its own projections, must bring shuttle launch costs down more swiftly — on a learning curve steeper than seventy percent from 1984 to 1988 (about sixty percent for the first two years). Such rapid progress in efficiency would be extraordinary for a competitive company, to say nothing of a federal bureaucracy.
NASA’s costs are high for several reasons. For example, NASA uses at least 12,000 people to make a shuttle launch. In contrast, the agency launches a Delta rocket with only 1,200, while the US military makes comparable launches with only about a tenth that number. Factors like this help explain why the shuttle’s reusability has not given the huge advantage we had hoped for.
Of Salin’s key points, the one he has found most difficult to get across is that the principle underlying the current pricing policy is basically wrong. The price that promotes the maximum usage of the shuttle actually discourages the maximum investment in space transportation. In other words, efficiency in shuttle use is inefficient in promoting investment. They are opposed goals.
Opposing Price Increases
Besides NASA, shuttle users and potential users oppose major price increases. Last July, Orbital Sciences Corporation circulated a draft petition on the subject, which is excerpted here:
“A large increase in the STS price after 1988 would be a profound strategic mistake, threatening the transportation capability upon which commercial development of space depends…A continued stable STS price is the best incentive Government can offer to fulfill this promise…The best strategy for reaching the goal of full cost recovery for the STS is not increasing price, but maximizing use of the STS.”
Leasecraft is a project of Fairchild Space, whose President John Townsend stated that “high shuttle prices in the early years of materials processing industry in space will preclude its success.”
While shuttle users have not compiled a detailed public response to Salin, various companies have produced position papers on shuttle pricing. Some excerpts follow: “One dimension of the Space Transportation System (STS) pricing policy debate has centered on the divergent interests of firms wishing to develop new, innovative technology such as microgravity processing and the firms wishing to privatize federally-developed expendable launch vehicles (ELVs) that have, for the most part, been in use for 15 to 20 years. The ELV interests are lobbying for a sharp rise in prices charged by NASA for STS launch services. NASA is resisting this pressure, which is now being applied through Department of Transportation and Office of Management and Budget.”
“The price charged by NASA for the STS is under political control and subject to large changes in the near future. A large, abrupt increase would send a message to business planners that the cost associated with a new technology venture in space is subject to not only technical and market risk, but political risks as well. A small, gradual price increase by itself may not produce this negative reaction, but a large, abrupt (50-100% in a single year) increase would assuredly send a chill through all organizations trying to advance microgravity projects.”
Rather than pricing to recover some part of the Government’s acquisition and/or operating cost, STS prices should be set to maximize use of the STS and to regulate access to it in an economically optimum way.”
“Subsidizing part of the cost of the STS system is not subverting market forces in a detrimental fashion any more than government aid to infrastructure building in railroads (land grants) or civil aviation (airports, navigation beacons, communications) harmed the national welfare.”
Senatorial Comment
Some US Senators have also expressed their opinion on the subject. In August 1984, nine members of the Senate Committee on Commerce, Science, and Transportation wrote to President Reagan, stating:
“The cost of launching aboard the Space Shuttle should not be a disincentive to these activities [R&D on the shuttle] in the commercialization of space; to the contrary, launching aboard the Space Shuttle should encourage the risk-taking ventures that will be crucial to the successful commercialization of space. As our country moves into an enterprising new era of space activity, we hope that you will consider the role that the Space Shuttle pricing policy can play in the commercial development of space and the ensuing benefits to our nation.”
The letter was signed by Senators Gorton, Trible, Kasten, Kassebaum, Hollings, Heflin, Inouye, Riegle, and Lautenberg. L5 members will recognize friends of space (and of our Society) here, especially Senators Gorton and Trible.
In contrast to the Senators, Rep. Ed Zschau (R-CA), whose district includes both Silicon Valley and Starstruck, agrees with Salin and has described him as “an expert on shuttle pricing” and his work as “very thorough.” In a January 3, 1985 letter published in the Wall Street Journal, Zschau points out the pricing subsidy and accompanying problems for the launch industry. (He also suggests that the space station be built by the private sector and leased to NASA.)
In testimony at a House Science and Technology space subcommittee hearing last summer, Zschau stated that unless NASA stops subsidizing launches, the “market share for the private sector will be diminished.” He further suggested that NASA stick to tasks beyond the capabilities of the private sector.
In June 1984 the Heritage Foundation, a well-known conservative think tank, issued a short, strongly-worded issue bulletin on shuttle pricing, authored by senior policy analyst Milton Copulos. In it he claims that NASA understates shuttle costs and recommends that NASA change its pricing policy to enable the expendable launch vehicle industry to flourish. “Should the space agency [NASA] succeed in blocking more efficient private space ventures, not only will vast commercial opportunities be lost to Americans, but the future of space exploration could also be threatened.”
The US Air Force, while not a participant in the shuttle pricing debate, has made clear its position on the shuttle vs. ELV: it has moved to acquire ten ELVs, thereby reducing dependence on the shuttle. At an American Astronautical Society meeting, Air Force Undersecretary Edward Aldridge stated “We’re pricing ourselves out of the space business because the launch cost is killing us…The cost of space operations must be reduced by orders of magnitude.” Rather than more orbiters, “Perhaps a big dumb booster is the way to go.”
Key Question
It would appear that low shuttle prices may indeed be holding back development of the cheap transportation we need to industrialize and settle space. Is it possible to bring those prices into line with true costs without damaging the new microgravity experimenters? Two possibilities suggest themselves, the first of which is obvious: NASA can charge experimenters lower prices than those charged to satellite launchers. Eric Drexler has noted another possibility which would distort the launch market even less: if governments wish to subsidize commercial experiments, they can issue “launch vouchers” worth a certain amount toward launch on any carrier.
Members wishing to research the issue can request Salin’s testimony (Vol. II of the FY1985 NASA Authorization Hearing Record) from the Publications Office, Committee on Science and Technology, Room 108, House Annex 2, Washington, DC 20515 or call (202) 225-5629. See Rep. Zschau’s letter in the Wall Street Journal (1/3/85) and various issues of Space Business News, e.g. the July 2, 1984 issue. Documentation of the shuttle user position is harder to find; feel free to contact me for further information.
The L5 Society, busy with the challenge of space station funding, has not yet started to investigate the shuttle pricing issue. When we do, we may find a problem broadly similar to that of the Moon Treaty — a case of short-sighted government rules holding back space development.
Chris Peterson is the outgoing editor of the L5 News and a former Secretary of the L5 Board of Directors.
