The recent Center for Strategic and International Studies (CSIS) report recommending that the U.S. government find ways to make more launch options available for commercial satellites gives due recognition to the critical national security role these systems play — even if it ruffled a few launch-industry feathers in the process.

The bulk of the satellite bandwidth used by U.S. and allied forces in the Middle East and Afghanistan today is provided commercially, and that will remain the case for the foreseeable future. The military’s capacity needs are growing exponentially, and the private sector is easily outpacing the U.S. Defense Department in deploying new satellites. Meanwhile, the idea of hosting dedicated government payloads aboard commercial satellites is getting solid traction. To cite just the latest example, Intelsat, which is under contract to host a UHF payload aboard the Intelsat 22 satellite for the Australian Defence Force, is investing in an identical payload for another satellite with the intent of leasing it to the Pentagon, which is facing a shortage in UHF capacity.

But it is by no means clear what U.S. government measures, if any, are warranted at this point to make more commercial launch capacity available. The current geostationary satellite launch market is dominated by two providers, Arianespace and International Launch Services (ILS), both of which are enjoying full manifests owing to the 2009 bankruptcy that has sidelined rival Sea Launch and the ongoing fleet replenishment programs of the major satellite operators.

ILS and Arianespace appear to have absorbed Sea Launch’s business with minimal disruption to the industry at large, and they cite this as evidence that the market has finally found a happy equilibrium. Not so, say satellite operators, who argue that a rocket failure could result in costly delays industry-wide and who naturally want more choices and lower prices.

There is evidence that the geostationary satellite launch market could shrink in the next few years as the fleet recapitalization cycle winds down and cash-flush operators begin sniffing about for acquisitions that could lead to cross-fleet consolidation. Meanwhile, Sea Launch is emerging from bankruptcy and eager to resume operations, doubts about its long-term viability notwithstanding; the Russian-built Soyuz rocket soon will begin launching from an equatorial spaceport; and Space Exploration Technologies Corp. is marketing its Falcon 9 rocket at prices the established providers will have a tough time matching. On top of that, Japan recently opened its Tanegashima Launch Center to launches year-round with an eye toward competing commercially with its H-2A rocket.

The market is segmented, to be sure; the Soyuz and Falcon 9 cannot handle the heavy satellites — those weighing 5 tons or more — that comprise more than half the geostationary commercial manifest each year. Still, given current trends, it is hard to recognize an urgent need for more launch capacity — unless satellite operators have no confidence that Sea Launch, which is entirely dependent on commercial business, can survive in the long term.

The operators would like to see China re-enter the commercial market; China’s Long March rocket has built a solid track record of success over the last decade launching domestically built satellites and has the staying power that comes with government support. It’s not clear whether a Long March return — the CSIS report favors such an outcome — would make the commercial launch supplier base more robust or simply drive out an existing provider. On the other hand, access to China’s satellite services market is sometimes tied to using the Long March rocket.

There is the possibility that hosted payloads will further segment the market; that could happen if the customer is the U.S. military, and the technology involved is so sensitive that Pentagon officials are uncomfortable with having it launched on a non-U.S. vehicle. Some commercial launch industry officials dismiss that notion, noting, for example, that Intelsat’s UHF payload is slated to fly on the Russian-built Proton and that an experimental missile warning sensor hosted aboard an SES satellite was licensed to fly on the Proton and now is slated to fly aboard Europe’s Ariane 5. But that doesn’t mean the Pentagon won’t balk at some future hosted-payload arrangement involving a Proton launch, especially if the technology in question is intended for operational as opposed to experimental use. It certainly would be helpful if the U.S. government would clarify its policy in this regard.

The potential hosted-payload issue provides an argument for having the U.S.-built Atlas 5 and Delta 4 rockets at the commercial market’s disposal. Both vehicles were developed with the expectation that commercial satellites would comprise a large part of their respective manifests, and both debuted carrying commercial payloads. Today these rockets, built and operated by United Launch Alliance (ULA), the Boeing-Lockheed Martin joint venture, are used almost exclusively to launch U.S. government payloads.

The CSIS report argues for a change in the U.S. Air Force-ULA relationship that would bring Atlas 5 and Delta 4 back to the commercial market. Certainly the Air Force could do a better job of managing its launch manifests, as current practices make it all but impossible to schedule a commercial mission within a reasonable time frame, even when government satellites holding launch slots are delayed significantly. But range flexibility isn’t the only thing keeping Delta 4 and Atlas 5 out of the commercial market — neither vehicle is cost-competitive with the Proton or Ariane 5, and ULA’s corporate parents would much prefer to focus on the profitable government business.

All of which is to say that assessments of the commercial launch industry continue to vary greatly from supplier to customer. The CSIS study captures key elements of that ongoing debate, and rightly draws the connection between a robust commercial launch market and U.S. national security. While it also makes some sensible recommendations, particularly with respect to the management of the U.S. launch ranges, it does not resolve one way or another the question of how much capacity is enough.