What about the Other Reasons for Shuttle Approval?

There were other reasons put forward for going ahead with the full capa­bility shuttle, and the success or failure of the shuttle in satisfying those reasons must be included in an overall assessment of the shuttle as a Nixon space legacy. Although secondary factors in the presidential-level decision to approve NASA’s full capability shuttle, the claims that NASA had made from 1969 on—that the shuttle could be launched on a routine basis and that it would significantly lower the costs of space operations—became the publicly offered reasons for developing the shuttle. The space shuttle failed to match those claims. As then-NASA Administrator Mike Griffin com­mented in 2010, “what the shuttle does is stunning, but it is stunningly less than what was predicted.”23

In his prepared statement announcing shuttle approval, President Nixon said that the shuttle would “revolutionize transportation into near space by routinizing it.” At the time the shuttle decision was made, the intent was to launch the shuttle 40 to 60 times per year once it became fully operational. This of course never happened. The most launches of the shuttle in a single calendar year ended up being nine in 1985, with the average annual launch rate over the 30-year lifetime of the program being 4.3 launches per year. The shuttle could not be launched on demand; rather, it took a lengthy and labor-intensive process to prepare each shuttle mission for launch. Rather than a vehicle capable of frequent and routine operation, the shuttle turned out to be, in the words of the Columbia Accident Investigation Board, a “complex and risky system” and a first generation “developmental vehicle” which required great care to operate safely.24

The Nixon January statement also said that the space shuttle would “take the astronomical costs out of astronautics.” In March 1972, as the final shuttle design was announced, NASA Administrator Jim Fletcher claimed that once the shuttle became operational, its incremental cost per launch would be $10.5 million (almost $60 million in 2014 dollars). But NASA in 2012 estimated the cost of each shuttle launch, depending on how it was cal­culated, as between $814 million and $1.266 billion, up to 20 times higher than the 1972 estimate. NASA in December 1971 had said that the full – capability shuttle would carry payload into orbit at a cost of $118 per pound ($691 in 2014 dollars); in Congressional testimony, NASA listed the cost of using the shuttle to deliver cargo to the space station as $21,268 per pound in 2011 dollars.25 By any measure, the shuttle did not “take the astronomical costs out of astronautics.”

One may ask why NASA’s 1972 estimates of shuttle operating costs were so far off the mark, while the estimates of shuttle development costs made at the same time were close to the actual amount. NASA cost estimators had significant experience in forecasting the costs to develop systems such as the Mercury, Gemini, and Apollo spacecraft, various launch vehicles, and scien­tific spacecraft. Based on that experience, they were able to estimate the cost of designing, developing, and testing the final NASA shuttle design with enough precision to allow Fletcher and Low to promise the White House that the cost was likely to be approximately $5 billion and certainly not more than $6 billion. The actual development cost was $5.5 billion in 1971 dol­lars, which were the basis for the NASA cost estimate.

None of those involved in the shuttle decision were as concerned about operating costs as they were about keeping the annual budget and total cost of shuttle development below politically acceptable levels. There was little White House or NASA leadership attention given to the quality of the cost-per-launch forecasts being put forward. NASA had little experience in estimating the costs of repetitive operation of a space system, and thus its estimates of shuttle operating costs were very uncertain. NASA, coming off its Apollo successes, was a technologically confident, perhaps even arrogant, organization, believing that it could incorporate advanced technology in such areas as propulsion, thermal protection, and on-board electronic sys­tems into the shuttle design in ways that would make the vehicle able to be operated inexpensively and on a routine basis. The per flight cost estimate provided to the NASA leadership in late 1971 was based on assumptions of 50 flights per year, vehicle self-checkout, fully reusable thermal protection, and long engine life without major repair. While that estimate may have rep­resented the best NASA could do in 1971, it ended up being based on the wrong technological assumptions and far off the mark. The shuttle required many hours of human labor for its checkout, a number of the shuttle’s ther­mal protection tiles had to be repaired or replaced after each mission, and the shuttle’s main engines required extensive refurbishment between uses. Shuttle program manager Robert Thompson characterized the NASA cost per flight estimate as an “optimistic guess.” That “guess” was nowhere near the actual cost once the shuttle began flying; rather than the shuttle signifi­cantly lowering the cost of space launch, it became an extremely expensive system to operate. Because of the need for extensive checkout and refurbish­ment between flights, the shuttle that was developed had no chance of being operated at a 40 or more launches per year flight rate; in addition, the antici­pated demand for that many shuttle launches a year never materialized.26