: Why The Space Shuttle is Too Expensive to Run, But Too Useful to Lose: Despite the loss of the Space Shuttle Columbia on February 1st, the Space Shuttle is still the most reliable space launcher available. It's also capable of carrying some of the heaviest loads, bringing a large crew along and, of course, landing and being reused. The Space Shuttle crew provides enormous flexibility, especially for tricky repairs or refurbishment of expensive satellites (the Hubble observatory and spy satellites.) But launching the Space Shuttle, or any other model of "booster", is inherently dangerous. These launches stretch existing technology to the limit. And when you do that, you cannot expect zero failures. The exceptionally low failure rate of the Space Shuttle is achieved with a combination of determined engineering, and lots of money. Each Shuttle mission costs over half a billion dollars. 

Most other launchers are much less reliable. The number two in reliability, the Russian Soyuz, is also the main booster for their manned missions. But reliability rates vary a lot, depending on who made the launcher and what for. Not all launchers are equal in reliability. Below are the satellite launchers that have gone up a hundred or more times, and the percentage of those launches that failed.

2% US Space Shuttle 
5 R-7 (Russian Soyuz)
5 Ariane 1-4 (European)
6 Tsyklon (Russian)
7 Kosmos (Russian)
10 Thor/Delta/N1/N2/H1 (US)
11 Titan 2/3/4 (US)
12 Proton (Russian)
13 Kosmos 2 (Russian R-12)
14 Atlas (US)

The Chinese CZ-1 has not hit a hundred launches yet (it's in the 70s) with a failure rate of about 11 percent.

Each type of launcher also had different failure rates depending which sub-model is looked at. Take the American Titan series. The Titan IIID had zero failures in 22 launches, while Titan IIIA had a 25 percent failure rate (in four launches), as did the Titan 4B (eight launches.)

When the cost of a launch, plus insurance, is added up, it's often cheaper to go with an more expensive American or European launcher, rather than a dirt cheap Chinese or Russian one. Russia, in particular, has made people nervous because of financial instability and mismanagement of their satellite launch operations. That said, the most widely used satellite booster, the Russian R-7, has been launched over 1600 times. The next most widely used is the Thor/Delta (nearly 600 launches). The space shuttle has gone up 113 times. 

Even with its greater reliability and capacity, the Space Shuttle is also the most expensive way get stuff into orbit. Satellites sent up via the Shuttle cost $25 million a ton. The Russians and Chinese will do it for $3-6 million a ton. But insurance can more than double that cost if there have been a number of recent failures with Russian and Chinese boosters. This keeps more reliable American and European boosters in business. 

From the beginning of space flight, many engineers have argued for just sending up unmanned space craft. This would be a lot cheaper, and no one would get killed if there was an equipment failure. But humans in space has been seen as an essential selling point when trying to get the needed money from Congress. Even the Soviet Union, a dictatorship, had to confront the publicity angle and put people into space. Moreover, NASA has not been getting the money if feels it needs to keep the Space Shuttle program going. A new Shuttle is needed, and the longer that is delayed, the more expensive it is to keep the existing Shuttle fleet going. The existing Shuttle is a 1970s design and there are a lot of new ideas and materials that could be used in a new design. Since the end of the Cold War, NASA has not been a big priority. But the loss of Columbia will at least get Congress to take a good hard look at the situation.