The Space Shuttle Columbia can be operated cheaper than most popular expendable launch systems if its launches are provided at marginal rates. That is, the additional cost of operating Columbia, after the other vehicles' costs have been accounted for in support of the Space Station program, is about an order of magnitude lower than the average Shuttle cost per flight (about $500M at the currently budgeted seven flights per year). Assuming free enterprise were allowed to pay only the marginal cost of operating Columbia for their use, a demand could be created (at least for one reusable Orbiter) for engineering higher and higher flight rates at lower and lower costs per flight.
It should be clearly recognized, however, that the use of Columbia only makes sense for the start-up and initial take-off phases of the enterprise. The Shuttle system, even with the enhancement suggestions envisioned, will reach a limit that will drive the nation toward advanced space delivery systems. The use of a Shuttle for starting a lunar enterprise, therefore, is not the answer for space delivery, but rather our next opportunity.