Here's a disaster waiting to happen. InsideDefense.com reports that the U.S. Air Force intends to commission a study on how it can transition its space-launch activities from a "performance-driven" model to a "cost-driven" model that would be much less expensive. A November 2 statement of objectives for the study says, "Recent success of an emerging new entrant space launch provider, along with numerous past studies and technology demonstrations, suggest designing to minimum cost may have significant life cycle cost advantages over a performance-optimized expendable launch vehicle." The statement goes on to note that the new approach examined in the study will help reintroduce competition into the space-launch market.

This study reminds me of debates within the Harland and Wolff shipyard a century ago about whether money could be saved by using lower-grade rivets in construction of RMS Titanic. We all know how that cost-cutting initiative worked out, and based on experience with more recent efficiency moves in the space sector, there's good reason to suspect that the Air Force's "cost-driven" launch model will end up the same way. The biggest reason launch vehicles used by the military cost so much today is that the government customer insists on design features, process standards and testing criteria intended to maximize reliability. You can eliminate many of these cost-drivers and still have a viable launch vehicle, but the risks that it won't perform as planned increase significantly.

Since most politicians and policymakers aren't rocket scientists, they are easily taken in by the beguiling notion that launch costs would plummet if there was more competition in the business. The reality is that there used to be more competition, and the two market leaders were forced to form a joint venture because government demand was low and performance expectations were high. So what the Air Force ended up with was a monopoly that charges a lot of money but almost always gets its billion-dollar birds into orbit. If safety and reliability standards are stripped away to encourage competition, then cost won't be the only thing that plummets: a lot of pricey satellites will fall into the sea rather than reaching their orbits.

The math on life-cycle costs is pretty straightforward. If you save $100 million per launch by moving to a cost-driven design but lose a billion-dollar satellite one out of ten times, then there are no net savings. In fact, there are net costs because loss of satellites requires expensive work-arounds to compensate for the absence of a vital space asset. We have gone through this problem repeatedly in recent years because the better-faster-cheaper initiatives of the Clinton years resulted in spy satellites that didn't work and missile-warning satellites that took years longer than expected to integrate. The resulting work-arounds cost billions of dollars and eclipsed any potential savings. Bottom line: if the government thinks it can save money by awarding contracts to whoever shows up with the lowest-price, technically-acceptable proposal, then there are some mission areas like space launch and cybersecurity where disaster lies right around the corner.

Loren B. Thompson, Ph.D.