by Tim Fernholz
“It’s like being punched in the belly,” the appropriately named commander at Cape Canaveral, Air Force Brigadier General Randy Starbuck, told reporters.
The two enormous aerospace companies shared similar approaches to the engineering business, a kind of total war on problems. Boeing, one former employee told me, “tends to attack things with an army of people. It sort of crushes a problem. It’s really big, big aerospace.” That has benefits, but an organization that relies on sheer size can develop a kind of institutional arteriosclerosis. “You can tweak the metrics to make things look green for a long time, but you are pushing the problems into the future,” the engineer told me, ominously. “We would deliver stuff, put it on the rocket, but not test it, knowing that when it was tested at the Cape, problems were going to pop up. But that would be in the future. The people who shipped it could take credit and go home. And if you’re smart, you move to another program before that day comes.”
Less than a year later, in April 1999, two more failures occurred, once again with the Lockheed Titan IV rockets the US Air Force relied on. Two minor mistakes, apparently failures in quality control, led to failed missions. In one case, heat protection tape wrapped around the wrong wires prevented two stages from fully separating; in the other, misprogrammed guidance software caused the second stage to run out of fuel mid-flight. Both failures delivered expensive satellites into useless orbits thousands of miles from their proper destinations.
Days later, in May, the Delta III tried to fly for a second time. This time, the new booster did its job, sending the upper stage of the rocket and its satellite into space. But the engine inside the second stage ruptured during flight, due to a poor weld, disabling the rocket and leaving its $145 million satellite payload to become just another piece of space debris circling the earth.
“I think this is probably one of the worst times in the launch history of the country,” the former head of Air Force Space Command told the Washington Post. “Even the old rockets aren’t working, and some of the newer rockets aren’t working.”
3
The Rocket Monopoly
Space preeminence is essential to being a great power in the next century.
— USAF Major General James Armor
This two-year spate of six launch failures cost the US government more than $3.5 billion. Beyond the lost taxpayer dollars, the failures showed a worrying lack of innovation. It was not just the old rockets that weren’t performing; the new rockets developed to replace them weren’t ready to get the job done, either. Where the United States had once boasted of “dominance in launch,” as Gwynne Shotwell, SpaceX’s future president, would put it years later, that capacity was now slipping away. “We owned it in the eighties and the early nineties, and we just let it go,” she said. By the year 2000—that beautiful round number that a generation of science fiction authors used as a watchword for a dawning space age—the United States still hadn’t solved the problem created by the Challenger disaster fourteen years before.
Yet the dreams of science fiction readers hardly mattered to the people who controlled the space program. This was no time for the United States, at that moment enjoying a period of global hegemony, to lose its ability to get to orbit. The military desperately wanted access to space, especially in its new role as something of a global police force in the post-Soviet, pre-9/11 era. Enforcing a no-fly zone over Saddam Hussein’s Iraq, the most active military mission at the time, relied on satellite communications and imaging and positioning technology. Without the ability to maintain a constellation of satellites in orbit, the United States would be in no position to project military force around the world. And, by extension, this meant that the global economy hung in the balance as well: without the ability to secure trade routes and enforce global security pacts, the whole international system could unravel.
Still, the private companies had yet to deliver a viable rocket, despite the government’s $1 billion investment and years of work. Lockheed Martin’s replacement for the Titan IV, a rocket that would be called Atlas V, was still two years away from flying. In August 2000, Boeing’s Delta III attempted its third flight. This time, engineers loaded the rocket with a dummy payload rather than risk a satellite. Neither the rocket nor its engines burst, but it missed its delivery target by a thousand miles. Still, Boeing said the test had succeeded in proving the reliability of its systems.
But that was the last time the Delta III would fly; Boeing mothballed the program, putting what it had learned into the development of a new, more powerful rocket for the EELV program: the Delta IV. Executives had downplayed the challenges of the work and the business case.
“It was sold with the wrong assumptions,” Garvey told me. “It was sold as an upgrade from the Delta II, which is so successful, and we know how to do it, and we’ve been doing it for thirty-five years. That was an improper way to bill it; you sell it that way because you want your customers and insurance companies to buy that. It was actually a very high-risk step.”
Garvey recalls being at the Cape for the first failed Delta III launch. The Boeing team had no real failure response plan. “The marketing guy said that would send the wrong message,” Garvey said, describing a fatalistic “‘we can’t fail’ type of mentality.” When the rocket broke apart, managers scrambled to respond. They needed to identify the problem, but they forced their team to leave all their notebooks and files in the control room to preserve records. These same engineers were sent back to California to figure out what went wrong, without any of their logs.
Still, the companies did learn from their costly failures. In 2002, the new EELV rockets, Delta IV and Atlas V, flew successfully for the first time. By 2004, the rockets had flown six launches, all successful. Only two were for the government—far fewer than the average of ten per year first envisioned in 1998. But the space business tends to ride a seesaw between two priorities: reliability and price. Now that rockets were flying, the seesaw tipped back the other way. Price was going to be the next problem.
Boeing and Lockheed executives came to the Pentagon, hat in hand, to complain that they couldn’t afford to sell their new rockets at the price they had agreed on. The government’s whole plan had relied on being a secondary user of the new rockets, with a robust commercial space industry paying down the costs of development. But that market had simply not materialized.
The biggest, most attractive satellite plays—Iridium, Globalstar, Inmarsat, and Teledesic—had all gone bankrupt at the turn of the century as the dot-com bubble popped and Wall Street balked at capital-intensive technology plays. The business climate for satellite communications was still going from bad to worse. Growing cell phone penetration presented a cheaper, more effective alternative to the bulky and often balky satellite phone. Undersea fiber-optic cables ate into the business of providing international communications links from space.
In another unforced error, protectionist rules put in place by Congress in 1999, which were intended to keep US technology from leaking to the Chinese and bolster the domestic launch industry, had the opposite effect: the European satellite industry blossomed, which in turn boosted demand for European rockets. American firms turned to government-subsidized rockets built in Europe, Russia, and China to fly their birds; for example, the current satellite constellation behind DirecTV, one of the most successful space businesses in the United States, was launched completely abroad. Soon, there was little in the United States for anyone to launch but government satellites, which left the two big rocket makers hanging out to dry.
“If you price for high rates and you only realize low rates, you’re losing money hand over fist,” George Sowers, then a Lockheed engineer working on the Atlas V rocket, says. “That’s what happened to both Boeing and Lockheed.”
Now the rocket makers said they needed to raise prices—or they wouldn’t be able to stay in the business at all.
Government auditors estimated that the lack of a commercial market would add nearly $8 bil
lion to the cost of the EELV rocket program. With other factors included, the expected cost overage added up to more than $13 billion—more than 70 percent over budget. In 2004, this huge increase triggered a legal trip wire designed to prevent out-of-control government spending, which required Secretary of Defense Don Rumsfeld to officially certify to Congress that the program was vital to national security. He did. The Defense Department still claimed that the program was successful, delivering 50 percent cost reductions from the older launcher families. Worryingly, the auditors dryly reported that they “were unable to verify the statements or projections.”
The EELV program had been conceived a decade before this, but its rockets were hardly proven. In fact, government engineers were fretting about excessive flight vibrations in both vehicles that threatened catastrophic failures. But the military had little choice but to accede to the rocket companies’ demands. The US military needed access to space, and it had already sunk a huge amount of money into an effort that had yet to deliver any tangible results beyond two launches.
The first solution proposed was a 2004 modification to the contract so that neither provider would “go into a death spiral of trying to be competitive or face extinction.” Besides paying a fixed-price fee for every rocket bought to launch US satellites, the government would also begin delivering annual cost-plus payments—essentially, a guaranteed profit for the two contractors—to pay for “launch site and factory facility depreciation and amortization (including production tooling), lease costs, launch and range operations, mission integration and assurance, special studies, program management and systems engineering, training, supplier readiness, and transportation.” The initial estimate for each company’s costs, without the profit component, was $300 to $360 million annually.
But this plus-up was not enough. “After a few years of hemorrhaging, both companies told the government, ‘Unless something changes, we’re done,’” Sowers explains. “The solution that was cooked up in the smoke-filled rooms was, well, you know, we don’t want one or the other of you to pull out; we kind of like the idea of two independent systems. Let’s keep them both and put them under one tent.”
Thus was born the great rocket monopoly, United Launch Alliance. The idea was simple: to create a joint venture out of the launch divisions of Boeing and Lockheed so that they could find efficiencies of scale and share facilities and personnel. At the same time, it was also a surprise. The companies were longtime rivals.
“I hated the idea at first,” Sowers says. “I was part of the Atlas program. We were fierce competitors with Delta—we were going to win, dammit.” In the years prior to the merger, Boeing had actually been punished by the government for violating trade-secret rules when one recently hired engineer was discovered with some twenty-five thousand pages of internal Lockheed documents stacked in his cubicle. The inside information would have improved Boeing’s ability to bid against them on rocket launches. The Air Force punished Boeing by giving seven of its launches, worth $1 billion, to Lockheed in 2003. Litigation and criminal investigations continued into 2005, which may have greased the wheels for a merger to put the whole fracas to rest.
But both companies were, in financial terms, “laying on the ground bleeding,” says Sowers, and had to take action. The only problem? Technically, monopolies are illegal. As soon as Boeing and Lockheed announced their plan, the Federal Trade Commission sued to block the merger. The resultant legal thicket was cut down only because the military went to bat for the companies, claiming it was their only way into space.
Even so, the undersecretary of defense, Kenneth Krieg, didn’t mince words about what this meant for the US space program. “The most negative view of the creation of ULA is that it will almost certainly have an adverse effect on competition, including higher prices over the long term, as well as a diminution in innovation and responsiveness. Although the parties assert that the joint venture would generate significant savings for the Department of Defense, our careful review of those savings leads us to conclude that the cost savings, while attractive, are not adequate to support the loss of competition.”
But none of that ultimately mattered, Krieg concluded, telling regulators that “the U.S. can no longer protect national security without space.” He cited one figure: during the 1991 US invasion of Kuwait and Iraq, 9 percent of US munitions were precision-guided by satellite; by the 2003 US invasion of Iraq, 67 percent were. This mattered for pilot safety, since a target that took eight sorties to destroy during the First Gulf War now required just one.
The FTC would ultimately allow the merger, with a few conditions, mostly designed to ensure that ULA didn’t give special treatment to its parent companies, which also built much of the space hardware that flew on those rockets. The Department of Defense promised that new entrants would get a chance to compete for contracts each year but didn’t release the criteria needed for that competition for another four years.
ULA remains divisive to this day. Its advocates point to its sterling launch record, with few failures, that put national security satellites into space on time. The Atlas V has arguably never failed, having logged more than eighty launches, while the Delta IV has suffered only a handful of partial failures. But launch prices continued to rise, with little increase in performance to show for it. ULA rockets still cost hundreds of millions of dollars more than those from European, Chinese, and Russian competitors. At an average cost of around $400 million a launch, ULA’s rockets aren’t much cheaper than the heritage systems they were designed to replace, though they are considerably more effective. And ULA continues to receive its annual subsidy, a cost-plus payment for maintaining its infrastructure, which is projected to surpass $1 billion in 2018.
“This was perfectly fine for Boeing and Lockheed,” Sowers told me. “They were done with the high-risk commercial launch business. They wanted the safety and financial security of cost-plus.” That financial security was real: Though the two military-industrial giants do not release ULA’s earnings, disclosures suggest the companies pulled more than $3 billion out of the joint venture in the first ten years of its existence. This didn’t represent a huge share of either company’s earnings, but the joint venture was in effect a cash cow that provided consistent annual returns without the need for much spending on research and development, or worry about the competition.
In the end, the US military got the guaranteed access to space that it wanted, but the space program itself was stagnant. The government offered little incentive for ULA to become more efficient, and the organization saw little reason to compete for private launches or to develop new technology. Even as the pain brought by the end of the dot-com bubble dissipated and tech investors looked anew at the space business, rising launch costs meant that anyone with a plan to make money flying a satellite looked abroad or simply shelved their idea. Even at NASA, scientists plotting research missions in the solar system and beyond knew that high launch costs would be a limiting factor with every proposal.
These consequences were easy to foresee, but the creation of a monopoly that commanded $32 billion in public spending, and in such a high-tech industry, received little attention. At the time, the national security debate was focused on the deteriorating war in Iraq. The public generally has little exposure to the rocket business—its products are sold to governments or other businesses—so there was no populist outrage against big business.
But what about Congress, which is always keeping an eye on the country’s purse strings? Well, of course, there are many kinds of purse strings. Boeing and Lockheed Martin, having worked for the government in various forms for more than seventy-five years each, knew which strings to pull. In 2006, as their merger proposal was under review by the government, Boeing spent more than $9 million, and Lockheed more than $10 million, hiring DC influence peddlers to smooth their path to regulatory approval. The two companies also gave a combined $4 million to candidates running for election that year. Perhaps most important, the companies’ indus
trial footprint—with major manufacturing facilities in states like Alabama, Florida, Colorado, and California—gave them privileged entrée to those members of Congress who weren’t eager to see hundreds or even thousands of good-paying, high-tech jobs disappear.
The monopoly that bailed out the US rocket was sealed largely behind closed doors.
But there was one organization that protested the deal with special vigor: a little-known rocket company founded just three years before by Elon Musk, who was now a bona fide dot-com millionaire because he had sold two of his companies, dodging the bottom of the tech bubble.
SpaceX had yet to launch a rocket, but it had announced a plan to develop an EELV-class rocket, to be called the Falcon 9, which would compete with Boeing and Lockheed for government contracts. The company’s attorney sent a protest letter to the FTC while the ULA merger was under review, calling its justification for the merger total bunk and demanding that it require ULA to give up its cost-plus subsidies and compete for launches by disclosing the total cost of its work. The company also asked that the Air Force stop allocating launches in five-year block buys—which could effectively lock SpaceX out of the national security market for years—and instead require competitive bidding on each flight.
“Nebulous claims regarding national security appear to trump concerns about the effects on competition—even though competition is critical to promoting innovation, which is critical to protecting national security on a continuing basis,” SpaceX’s counsel wrote to the commissioners, accusing them of “largely ignoring the harms that will be done to competition in the broader U.S. launch services industry if this proposed merger-to-monopoly proceeds.”
