The recent announcement that the Pentagon will set aside up to 14 Evolved Expendable Launch Vehicle (EELV)-class rocket “core” purchases for open competition beginning in 2015 is an important first step in the long-overdue process of both incentivizing cost reductions in the EELV program and in addressing the gaping flaw in logic in first accepting, and then heavily subsidizing, a two-vendor solution in the Atlas 5 and Delta 4, which share a single point of failure in two variants of the same RL-10 upper-stage engine.

Given the changes taking place in the U.S. government’s approach to securing military launches, the recent proposal to have the taxpayer fund development of a dual launch adapter for deploying GPS satellites from the Atlas 5, which would not debut until 2017, should be viewed in a very skeptical light. There are multiple problems with the proposal, not the least of which is that the numbers don’t add up.

According to United Launch Alliance (ULA), the Boeing-Lockheed Martin joint venture that builds and launches Atlas 5 and Delta 4 rockets under the EELV program, introducing a dual launch adapter for the GPS 3 series of satellites “could save as much as $50 million per satellite” compared with single manifested launches. That figure, however, is extremely questionable because of the change in launch vehicle configuration required to make it happen. The dual strategy would require a transition from an Atlas 5 411, with an approximate cost of $130 million per flight, exclusive of its share of the Launch Capability contract, to the considerably more expensive Atlas 5 551. Based on the published cost of the U.S. Navy’s Mobile User Objective System 1 launch, as well as NASA’s Juno mission, the cost of the more powerful Atlas 551 can be estimated to be on the order of $215 million per launch, also exclusive of Launch Capability charges. Even assuming all 26 potential flights are condensed into 13 launches on the more expensive Atlas, the savings would appear to be closer to $23 million per satellite, and that is before factoring in amortized development costs of the payload adapter. Estimates of $1.5 billion in lifetime savings through cost avoidance versus the single manifested baseline are equally unrealistic.

Rather than further subsidizing ULA through funding the dual adapter, the GPS 3 satellites which it would launch represent instead an opportunity for incorporating new entrant competition.

In recent years, the most commonly cited rationale for continued adherence to the EELV monopoly is the “expensive, almost priceless” value of U.S. Department of Defense payloads. While each may be necessary for national security, some are clearly more expensive and more priceless than others. Compared with the $3 billion per unit Space Based Infrared System early warning satellites or the $2 billion Advanced Extremely High Frequency spacecraft, GPS satellites fall undeniably on the lower end of the spectrum. With a permanent constellation of roughly 30 active GPS spacecraft in orbit, a steady stream of replacements waiting to launch and still others held in storage, the loss of any one satellite, while unfortunate, simply cannot be placed in the same risk category as the aforementioned programs, each of which has a much higher unit cost and a far smaller fleet. By comparison, with the projected cost of GPS 3 satellites under $250 million and a schedule that calls for only two launches before 2016, the fundamentals of the program are as close as military space gets to a commercial market environment. This strongly argues that it is a clear point of entry initially for Space Exploration Technologies Corp. (SpaceX), but also hopefully for Orbital Sciences Corp., in a manner that protects the national security both literally and financially.

If saving increasingly scarce defense dollars relative to GPS 3 launches really is the goal, a more effective solution lies in separating this component of defense launch opportunities from the “up to” 36 core ULA block buy entirely, and instead specifically opening it up to new entrants with fair and open competition for awards beyond 2016, including denial of additional Launch Capability charges, if that portion of the contract persists, for orders awarded to ULA.

Doing so would allow the Air Force to take advantage of the much greater cost differential between a single manifested SpaceX Falcon 9, at roughly $80 million including mission assurance fees, resulting in an additional $27 million per satellite savings compared with the ULA proposal, without the addition of a taxpayer-funded development program. Taking into account further savings from avoiding ULA’s exclusive Launch Capability charges, currently averaging an additional $120 million per flight, the cumulative savings would be even greater.

Pending a successful debut launch campaign, similar benefits are potentially available in the Orbital Sciences Antares launch vehicle as well. The U.S. launch establishment, and particularly NASA, is sorely in need of a replacement for the Delta 2 in the medium-lift category, a requirement for which the Antares is the only immediate alternative. Providing ULA a free launch adapter and damaging Orbital Science’s long-term business case for the Antares, which is conservatively predicated on conducting two or three launches per year, is clearly not in the national interest.

With SpaceX having two orders as part of the Air Force Orbital/Suborbital Program-3, more than 20 commercial launches currently on the manifest and two technically challenging trips to the international space station under its belt, it is simply not credible to suggest that the company lacks either the technical capability or the experience to have a fully vetted Falcon 9 version 1.1 ready to deliver the comparatively small GPS birds safely to orbit well in advance of 2017.

Another reason to forgo publically funding the dual adapter proposal, and instead advance the entry of new competitors into this element of the EELV business, is the redundancy to be gained by selecting from three distinctly separate vendors. In recent years, as the costs have begun to reach levels that Gen. William Shelton, commander of Air Force Space Command, has described as “unsustainable,” the line of reasoning that demands the U.S. accept dramatically higher prices for Defense Department payloads aboard current EELV rockets is the implied superior reliability. That pricing premium, however, must now be considered in the light of the most recent Delta 4 launch, which suffered a performance problem on the second stage due to a fuel leak, resulting in a longer than intended burn to reach the required orbit. Ironically enough, the payload was a GPS satellite. Had it happened to have been a significantly heavier payload, as Shelton recently observed, the most likely result could have been a launch failure. While the RL-10 has returned to flight, even as the specific source for the leak has yet to be determined, the issue draws an unexpected equivalency with SpaceX and the No. 1 Merlin engine shut-down on its most recent launch, challenging the premise for accepting and even subsidizing launch premiums, which are no guarantee of a positive outcome.

As to concerns frequently put forward by advocates of the status quo that the market cannot sustain multiple vendors, it bears recalling that the original justification for proceeding with a two-vendor solution for EELV was that healthy competition would promote ongoing innovation and reduce taxpayer expenditure by spreading fixed costs through commercial orders. More than a decade after the fact, the excuse that the commercial market anticipated in 1998 did not pan out and must finally be put to rest. SpaceX, in winning multiple commercial orders, has demonstrated unambiguously that when innovation leads to reduced costs, the global market is in play.

In the long run, the greater benefit to be derived from forgoing direct support of the dual launch initiative may be the tough love given to United Launch Alliance in requiring that it step up to the plate and deliver more cost-efficient launch solutions for which it is undeniably technically capable. The “dial a rocket” approach has produced numerous versions of both the Atlas 5 and Delta 4, but the number it really needs to ring up is one that is economically competitive on a level playing field, and right now it is clearly a dropped call.


Stewart Money is a freelance writer currently working on a book about SpaceX and the Falcon.