Launching humans into space is difficult, dangerous and expensive — and we do it anyway, for good reason.


There is enough energy (kinetic plus potential) in 1 pound of anything in orbit to melt and vaporize every molecule in 1 pound of iron four times over. Energy in an orbiting shuttle — about the same weight and size as a Boeing 737 — is comparable to that required for a 737 to travel around the world (energy input in about 10 minutes for shuttle vs.50 flight hours for 737).

The same energy must be dissipated in getting back to Earth, while absorbing very little of it in the spacecraft.

Energy for suborbital flight is only about 2 percent of that required to reach orbit. Virgin Galactic, XCOR Aerospace, Benson Space Co.and others are planning to offer suborbital rides to passengers willing to pay.

In a typical launch to orbit, only about 1 to 4 percent of total launch weight is payload; 80 to 85 percent is fuel; and the rest is structure, engines, maneuvering fuel, etc., most of which are used or destroyed during a mission. Ratios are even more critical for reusable launchers.

Weight of the parts that reach orbit is critical. There must be highly efficient fuel and engines and very lightweight parts. The resulting design and operating margins are very thin, and parts must be very reliable.


Launch vehicles contain many critical items that are not as forgiving as similar parts in autos or airplanes. Examples of potential failures include stuck fuel valves, autopilot or navigation failures, stuck air vents, auxiliary power unit failures and engine fires — all probably not fatal in an aircraft or automobile, but fatal in a space launcher.

The only way to overcome this is with super-reliable components, redundancy, very careful operations and a great deal of flight experience.

Shuttle systems are redundant in nearly all critical parts, functions and assemblies (from the critical items list). Many systems are doubly or triply redundant. The shuttle main computer is quad redundant.

All new launchers are forecast as better, faster and cheaper — and definitely safer — before they are built. Nobody has ever been killed flying a launcher that never launched.

Launch vehicles need numerous test flights before potential flaws in design, manufacturing and operating procedures are identified and corrected. Some flaws are apparent only during flight. New passenger aircraft require hundreds of test flights before they are certified as safe to carry passengers.

All test flights of both aircraft and space launchers are thoroughly instrumented and include extensive postflight analyses (which are quite expensive).

The high expense of launching prohibits a full test flight program; therefore, all early launches of a new launcher are by definition test flights and are relatively risky. All launchers stay in this flight test mode for their entire lives. Experience is the best way to improve and verify flight reliability and safety.

Unmanned launchers do not have as extensive redundancy and special care as human spaceflight launch systems. Launch escape systems have their own problems and limitations.

Even if a commercial human launcher could attain its goal of 999 successes out of 1,000 attempts — which is unlikely — it is still a very risky endeavor, and is 10,000 times as risky as a trip on a scheduled commercial airline. Launching humans into orbit is inherently dangerous, and likely to remain that way for a long time. Going into Earth orbit is a job for explorers who understand the risk, as do experienced test pilots, not for ordinary citizens or school teachers.

The risk to crews on a mission to the Moon or Mars could be two or three times greater than the risk of launching from Earth to the international space station (ISS). Landings on the Moon or Mars could raise the risk another factor of three.

If a U.S. astronaut is killed in a space mission, it is a bloody nose for the nation. It injures our national pride, and our astronauts pay the ultimate price. It is a tough business.


Technology development without a mission on which to focus the effort is very inefficient and wasteful. Much money can be spent on things that have potential but require additional major development to be used. An example would be money spent to develop ion propulsion; a plasma or ion engine requires a nuclear electric power source to be useful. Many times, an advanced technology project must be redesigned to suit a particular mission.

For up to about four launches per year, fixed costs can be greater than 90 percent of the total cost for a particular launcher. Fixed costs include dedicated launch facilities, control center overhead, and specialized manufacturing and test facilities. Engineering and manufacturing personnel must be kept in place for analysis of flight results and to be on hand to support subsequent flights.

Expendable launchers destroy nearly the entire vehicle each flight. What would it cost to fly a Boeing 737 from Los Angeles to New York twice per year if the aircraft were destroyed each time and a new aircraft were required for each flight? And if the manufacturing facilities were special-purpose and most of their workers were specialized and had to be kept on salary between flights?

Higher launch rates theoretically could reduce the cost per launch and cost per pound to orbit — but this could be a chicken-and-egg situation: Can costs be reduced enough to stimulate enough demand to get to the very high launch rates? Probably not.

Now that we have had several recent years of launching humans without fatal accident, the NASA administrator and deputy administrator, along with the director of the White House Office of Science and Technology Policy, have based top-level policy on their belief that launching humans into orbit has become routine and relatively safe, and a commercial company could provide this service just as well as NASA. This is false and dangerous logic, and the nation is about to pay a huge penalty for this mistake.

With a projected cost to a private citizen of $50 million to $100 million for a few days in orbit, together with a high risk of fatality, there are likely to be few paying passengers. Since NASA will be using no more than three or four flights per year to the ISS, the market will be quite thin.

Given this economic outlook, it will be very unlikely that any company will elect to invest in such a venture. Of course, if the government were to fund development, pay the cost of each flight, and bear the risk of killing people and the associated liability, you may find companies willing to accept government money at no risk to themselves.

Commercial crew launch companies should be required to invest their own money for no less than half of all development costs, and to carry adequate liability insurance not only to cover loss of life but also to reimburse NASA for losing access to the ISS for extended periods. In return for NASA’s supplying development money, they should also guarantee a maximum price to NASA for each flight. Even then, there would remain a potential problem of the company deciding to quit at some point, leaving NASA with no option for U.S. access to the ISS.

Only when commercial operators invest significant amounts of their own money and agree to provide full liability insurance, as quoted by independent insurance companies, will the true cost of commercial crew launches be known. This proposed NASA policy should be examined carefully.

Why do we do it?

So since launching people to space is difficult, dangerous and expensive, and will remain so for a long time, why do we do it?

NASA is one of the few federal government programs about which most people would say, “My federal government does that, and I am proud of it.” The general public supports human space exploration because they are participants in this grand adventure.

A U.S. astronaut on a space mission personalizes the adventure like a robot can never do. Most Americans do not get excited about space science without astronauts. Remember “No bucks, no Buck Rogers”?

The total 2011 NASA budget is less than six-tenths of 1 percent of the federal budget, and the shuttle’s cost is only one-sixth of the NASA budget. Most people think we spend a lot more than that, but they still support NASA. The reason:

  • Inspiration — it encourages young people to study math and science.
  • Astronauts are role models — they represent the best of America.
  • National pride.
  • Some scientific knowledge is acquired better by people than by robots.
  • The United States’ perceived leadership in the world is valuable in many ways, including confidence in security and investment in U.S. companies.


O. Glenn Smith is former manager of shuttle systems engineering at NASA.