t is difficult to deny what was captured on videotape 81 seconds after the space shuttle Columbia lifted off on its 28th and final mission. The Jan. 16 recording shows a catastrophe in the making-a chunk of insulating foam, about the size of a small suitcase, falling off the shuttle's external fuel tank and smashing at high velocity into the orbiter's left wing. But mid-level managers at the National Aeronautics and Space Administration denied the episode left seven astronauts in mortal danger. They shrugged off numerous warnings from worried engineers and opted not to get spy satellite photographs of what accident investigators later determined was a hole in the wing.
Even after Columbia disintegrated over Texas Feb. 1-very likely downed by superheated air that got into the wing through that breach-some managers expressed disbelief that one or two pounds of foam could have caused so much damage. Members of the independent Columbia Accident Investigation Board strongly believe the debris did play a role, but without pictures of the punctured wing, they have been stymied in proving it. In late June, almost five months after the accident, the board said a series of impact tests with air cannons, foam bullets and reinforced-carbon heat shield wing panels told a "compelling story" about exactly what happened that fateful day.
The panel, led by retired Navy Adm. Harold Gehman, will release its final report on the disaster at the end of August. The report likely will blame more than foam for the destruction of Columbia and the deaths of its crew. It probably will accuse NASA of allowing itself to be lulled into complacency by repeated debris strikes that didn't grievously harm shuttles during dozens of previous missions. Gehman told the Senate Commerce, Science and Transportation Committee May 14 that the NASA safety organization "looks perfect" on paper, "but when you bore down, you don't find any there, there." He said his 13-member panel would produce "an independent analysis and review of not only this accident, but a deep, rich, intrusive inquiry into the entire human space flight program."
Without a doubt, the panel's report will draw chilling parallels to the 1986 space shuttle tragedy and the agency's repeated acceptance of booster rocket seal failures leading up to the O-ring leak that doomed Challenger. After settling on the left wing hole as the primary cause of the Columbia accident, the board refocused its investigation to determine whether NASA's management problems had not been resolved in the 17 years since the Challenger exploded moments after liftoff. "I'm hearing an echo," observed board member Sally Ride at one of the board's public hearings. Ride, America's first female space traveler, served on the presidential commission that probed the earlier calamity. This time, Ride and the other
Columbia investigators began their search for contributing causes by examining whether NASA set itself up for failure when it handed off the $3.2 billion-a-year shuttle program to the private sector in 1996.
With 87 percent of its $15 billion budget dedicated to contracts-the second highest share in government-NASA is a magnet for contracting criticism. "This reliance does not come without problems," says Allen Li, director of acquisition and sourcing management for the General Accounting Office. The agency's ability to monitor its many deals with many companies is a perennial concern. A GAO report issued just a few days before Columbia was lost said NASA's contract management was ineffective and its financial controls were weak and risky.
A March report by the International Federation of Professional and Technical Engineers found the agency faced serious challenges before the Columbia disaster. "A combination of budget cuts, workforce downsizing, and contracting out of key NASA operations negatively affected the safety of NASA's manned space program, its ability to retain and pass along core technical knowledge, and its oversight of the contractor workforce," the union observed. The debate over safety vigilance in the shuttle program, and the degree of oversight necessary to minimize risk, has raged since the shuttle contract was conceived in the mid-1990s, at a time when Congress and the Clinton administration were pressuring many federal agencies to outsource government work.
The deal NASA made seven years ago with the Boeing-Lockheed Martin partnership called United Space Alliance arguably is the government's most controversial experiment in performance-based contracting-a method of constructing agreements with companies that specifies results sought, but not the means of achieving them, and bases payment on the contractors' degree of success in attaining those results. The ninth NASA administrator, Daniel Goldin, awarded the contract to USA without a competition in 1996. Known for his "faster, better, cheaper" management philosophy, Goldin justified the award with an outside review that determined it would cost one-third less to fly shuttles if the government stopped interfering and let one company do all the work. Shuttle program managers embraced the space flight operations contract as a tool for eliminating duplicate effort, shrinking the NASA workforce, and hitting then-President Clinton's budget-cutting targets. Critics argued that NASA was dismantling safeguards designed to prevent another catastrophe.
PUSHING THE ENVELOPE
USA absorbed 15 companies and their 25 associated contracts in the first phase of a six-year, $9.4 billion deal. Last year, NASA exercised one of two available two-year contract options that increased the contract's value to $12.7 billion over eight years. The USA deal pushes the government-contracting envelope. Instead of fixing a fee and specifying a level of effort, it bases payment on performance. Although the method had worked with less complicated services, such as maintaining military bases and other facilities, in the mid-1990s, no agency had yet tried it on an activity that risked human lives and a national asset worth billions of dollars.
The contract promises a bonus for each shuttle mission completed safely and on schedule. There also are rewards for operating under budget-USA can keep 35 cents of every dollar it saves. "Unless you understand the contract . . . you might look at it and say, 'These guys are so anxious to save money they aren't concerned about safety,'" notes former astronaut Michael McCulley, USA's chief executive officer. "The schedule pressure is legitimate . . . but, by the way, the machine has to be as perfect as you can make it." He points out that USA cannot earn cost incentives for any evaluation period in which it does not meet the criteria for safety, whether in flight or in the workplace.
Nonetheless, auditors have admonished NASA numerous times since 1996 to keep closer watch over costs and safety operations at United Space Alliance. At least twice in the past three years, NASA's inspector general questioned the agency's financial arrangement with USA. Once, the IG revealed, NASA gave USA $33 million to cover two canceled flights without ensuring the credit was fair and reasonable. Another time, the IG criticized the contractor's use of almost $469,000 in professional and consulting services that did not meet federal acquisition regulation requirements. Twice in the past two years, the IG concluded that contractor surveillance in the shuttle program was disorganized and that NASA was not following its own rules for daily oversight, instead letting civil servants spot-check the work.
NASA Administrator Sean O'Keefe suggests that his agency's obsession with safety after Challenger evolved into a "control freak mentality" about risk. Rather than anticipating problems, he says, "When we find something, we're all over it like a bad habit. Everybody goes, 'Stop the show, stand right in your spot, don't move, we're going to fix this problem.' Quality assurance is a more sophisticated, more progressive, more modern, more contemporary way of looking at . . . trends, incidents that may point to other factors that tell me something." The administrator says one of the lessons NASA is learning from the Columbia accident is "how to ratchet up the management approach to a more contemporary standard . . . to have the quality assurance mentality more dominant."
In several public hearings during the investigation, expert witnesses detailed a range of inadequacies they said the Columbia disaster exposed. Among the deficiencies was a problem-reporting database that might have blinded engineers to the real hazards posed by launch debris. Witnesses explained that shuttle engineers developed a false sense of security about safety hazards because they inconsistently labeled technical problems. One time, a problem might be described as an "in-flight anomaly"-triggering inspections and corrective actions. Another time, the same or a similar problem might be described as "in-family"-not necessarily normal, but within a set of safety boundaries determined by the worst cases. Debris strikes that caused damage to the shuttle's delicate thermal tiles and fabric were inconsistently labeled.
Witnesses also pointed to lax oversight of engineering decisions that were based on information from faulty analytical tools. Engineers used what investigators described as a "rudimentary" computer model to determine whether the foam debris that struck the orbiter's wing could cause serious damage. The model dates back to the 1960s Apollo era, but never until January had it been used as a decision-making tool during a shuttle flight. Gehman panelists also said members of NASA's management team lacked sufficient training for the life-and-death decisions they are required to make during shuttle flights. And witnesses noted a persistent communication gap between supervisors and rank-and-file employees.
Giving too much responsibility to contractors can distort perceptions of risk, testified Diane Vaughan, author of "The Challenger Launch Decision: Risky Technology, Culture and Deviance at NASA" (University of Chicago Press, 1996). Although shuttles flew 87 successful missions after the first failure, Vaughan, a Boston College sociologist, said NASA's organizational structure continued to interfere with communication, and personnel reductions continued to compromise safety. "If you're not in touch with the everyday engineering data itself, you can lose sight of the fact that [the shuttle] is still an experimental system," Vaughan told the panelists.
Late in the investigation, board members were debating the wisdom of NASA's 1984 decision to designate the shuttle an operational vehicle after just four test flights. "The board is attempting to agree among itself as to how much [of shuttle operations] is exploration and how much is running a trucking line," Gehman said in a June 12 public hearing. "Obviously there's some of each in here." Characterizing the shuttle program as still "experimental," rather than "operational," undoubtedly will prompt questions about the appropriateness of handing to a contractor the operation of a program still in need of close scrutiny.
In the hearings, architects of the shuttle program testified that NASA erred with its hands-off approach. The agency "backed the government out too far" from flight operations, said Robert Thompson, top-level program chief from 1970 until 1981, the year Columbia made its first Earth orbit. Thompson's mid-level counterpart in those years, Owen Morris, accused the agency of reneging on a promise to fill the shuttle operations void with hands-on research and development work to train civil servants to become "smart" technical managers. "They have to come up through the ranks with a few dirty fingernails, maybe even greasy fingers," Morris said.
As the investigation wore on, discussions about NASA management were long on criticism and short on helpful suggestions. Gehman, who presided over the investigation of the Oct. 12, 2000, attack on the USS Cole, which was blown open in Aden harbor in Yemen, initially approached the idea of criticizing NASA management with trepidation.
"We as a board are certainly skittish about making changes to a very complex organization for fear of invoking the law of unintended consequences," he confessed during a hearing April 23.
But NASA was already pondering a major organizational change at the time of the Columbia accident. Agency executives were debating what to do about two studies, one internal and the other by outside experts, recommending NASA complete the transfer of shuttle operations to the private sector. Former shuttle program director Ronald Dittemore, who resigned in April, wrote the internal study. Dittemore thought privatizing the shuttle could help resolve NASA's staffing problems. The average age of the workforce is 47, and almost 30 percent of all full-time permanent employees are in science and engineering jobs. More than 25 percent of those workers are eligible to retire in the next five years. NASA achieved a 50 percent civil service workforce reduction in the shuttle program alone by consolidating contracts. "That's a significant decrease and a significant loss of knowledge and experience," Dittemore told the Gehman panel March 6. Some staffing reduction was justified as NASA reduced its operational role, but it inadvertently weakened independent verification skills in the remaining half of the workforce, he said. Dittemore ruled out rehiring civil servants because "the environment at the time was not bigger government for the sake of the shuttle program-it was to get lean and mean, try to get efficient, try to reduce and still be safe." The only reasonable option, he concluded, was to put together the government and private sector workforces "in such a way that preserved the safety of the program for the long term."
Sean O'Keefe became NASA's 10th administrator in December 2001 and vowed to push further privatization of the shuttle. He was intent on implementing the president's management agenda and its five key principles. One of those principles, competitive sourcing, holds that increasing competition improves performance. For options he could apply to the shuttle, O'Keefe turned to RAND, a Santa Monica, Calif., research institution. RAND thinkers recommended against wholesale privatization or commercialization, saying the shuttles cannot be operated at a profit and would not fetch their replacement value if sold. In a report released in December, the RAND task force charted several "alternate trajectories" for reducing the government's role in shuttle operations. They ranged from further consolidation of contracts to a pseudo-governmental space flight authority to provide human space transportation services. With appropriate shared liability arrangements and a proper fleet maintenance plan, the space shuttle could be readied for a shift to some form of private sector control with very limited government oversight in the next few years, RAND found.
The idea was not well received, especially after the accident. "To throw it over the fence to industry and say, 'We're going to privatize it now, and you have to operate it more cheaply and efficiently, and we're not even going to let you control how we spend money to upgrade it,' is insane," says space policy consultant James Muncy, a former aide to Rep. Dana Rohrabacher, R-Calif. Rep. Dave Weldon, R-Fla., told the The Orlando Sentinel in March that the privatization concept did not have "any legs at all." Weldon was anticipating calls for a move in the opposite direction, toward increased NASA oversight. Before joining the Gehman board, George Washington University space policy analyst John Logsdon predicted NASA would be pressured politically to take back some management responsibility. In February, science policy historian William Kay of Northeastern University told The Miami Herald that NASA would reconsider its privatization move "if for no other reason than NASA might find it convenient to focus [blame] on the company." Kay, author of "Can Democracies Fly in Space? The Challenge of Revitalizing the U.S. Space Program" (Praeger, 1995) also told The Washington Post in February that because so few companies are qualified to do the work, placing shuttle operations in USA's hands merely added a layer of bureaucracy.
LAUNCH CONTRACT LIMBO
On Jan. 31, NASA was taking stock of companies that might be interested in and capable of taking over shuttle operations. If it were a long-term agreement like the USA deal, a new contract would be worth roughly $20 billion. That sum likely would attract not only USA but also its two parent companies and the latest entry in the market, Los Angeles-based Northrop Grumman Corp. with its newest acquisition, longtime NASA contractor TRW Inc., based in Reston, Va. Northrop completed its acquisition of TRW in December, creating Northrop Grumman Space Technology.
Whether it privatized the shuttle or not, NASA wanted to restructure the shuttle contract. Program officials were satisfied with USA's performance but dissatisfied with incentives in the contract. "It doesn't have anything to do with the accident, but we just think there's room for improvement," says NASA Assistant Administrator for Procurement Tom Luedtke. To buy time for a decision, NASA had exercised the first two-year option, renewing the contract until September 2004. "Even people who believe that given the accident and the uncertainties, our best course is to go ahead and exercise the [second] option [with USA], realize that in three years, we're out of contract," he says.
On Feb. 1, NASA tabled its shuttle contract re-evaluation to await guidance from the accident investigators. In June, they were examining the implications of NASA's decision to use monetary rewards to motivate its major contractors to perform. Former USA CEO Russ Turner told the Gehman panel June 12 that if the investigation determines Columbia was lost because of something the company did or failed to do, it will lose $70 million in incentives. "If it turns out they see fundamental issues with how we are doing procurement, the structure of USA, or with other things that deal with how we're currently organized to do business on the shuttle, then obviously, we're going to address those and fix them," Luedtke says.
GLITCHES IN THE DEAL
The structure of the contract already had become a problem for NASA as the agency struggled to maintain a construction schedule for the $100 billion international space station it is building in partnership with several foreign governments. The shuttle currently is the only means of ferrying parts and people to complete assembly and provide crucial services to the station. As construction revved up in 1998, it became clear that NASA and USA had competing priorities. At issue was the size of the workforce. "The low-hanging fruit was harvested pretty early on, and it was kind of getting to a point of diminishing returns. They needed other incentives," says William Readdy, NASA's associate administrator for space flight.
The contract does not require the company to keep extra workers on the payroll to hasten the pace if NASA has a scheduling emergency. It only encourages USA to staff shuttle operations at a level appropriate to assure safety and quality criteria are met and NASA's flight manifest is supported. In 1998, NASA was anticipating a budget shortfall and reduced flight rates because of space station construction delays. Russia was late delivering a key segment of the station to orbit, and other components were late arriving in Florida for launch on the shuttle. Expecting fewer shuttles to fly, USA went to a two-shift operation and laid off about 550 workers. As a result, USA lost the flexibility to make sure shuttles were ready to meet ever-shifting station construction schedules. "In the grand scheme of things, [that] may be a small dollar impact to the shuttle, but profound consequences for the international space station," says Readdy. Within a year of the layoffs, NASA asked USA to staff back up. Mixed management signals such as these have been a source of frustration to contractor executives. "The contractor wants terms and conditions that are aligned to the government's priorities," USA's Turner told the board in June. "The best contract you can have is one that, when you perform well, the customer's happy. Either NASA needs to align itself around the goals and objectives, or you need to change the contract."
In the second phase of the contract, USA could have absorbed 16 more companies and about 60 contracts covering the supply of such critical space shuttle components as the main engines, external tank and solid rockets. But on the way to total consolidation, something happened to cause NASA to reflect further on the wisdom of its deal with USA. In the span of nine months in 1998 and 1999, five unmanned rockets failed during launch. Two were commercial vehicles and three belonged to the Air Force-none were NASA's. But the accidents prompted government and industry to undertake a broad management review that indicated the Air Force's decision to back away and let contractors have more control over expendable vehicle launch processing was ill-advised.
Not long after that, NASA decided to reconsider the long-range strategies spelled out in the anticipated second phase of the shuttle contract. "That really did cause us to pause and take a look to see what we should do," says Readdy. "How should we look at the future of the contract? Should we in fact take a fresh look and say, 'OK, maybe it should be propulsion elements as one contract and vehicle elements and processing as another'?" Despite being denied responsibility for the external tanks, USA continued to struggle with schedule challenges posed by the flyaway foam insulation, Turner testified June 12. The company paid for the necessary repairs. "I was motivated to find a way not to have that foam come off anymore, because it was creating a turnaround issue for us," he said.
USA agrees it was time to re-evaluate the contract and incentives, but company officials who commented for this article were circumspect. "We're 10,500 people focused on one customer," says USA CEO McCulley. "We do not want to appear to be negotiating through the media." NASA and USA might not be far apart on what should happen with the shuttle contract, however. When USA conducted its annual suppliers' conference in Washington in February, it did not send participants to lobby on Capitol Hill. The rare change of agenda partly was in deference to the lost Columbia crew. But McCulley says the company did not feel the need to lobby this year. "In the past, we found ourselves on the Hill, sometimes at odds with what Dan Goldin was saying. We're in complete and total support of Sean O'Keefe. Sean keeps talking about 'One NASA,' and he's right."
Although the shuttle was intended from its inception to be operated commercially, the demand for its services has been small. The Challenger accident dashed hopes of using the shuttle to haul commercial communications satellites to orbit and drove the Defense Department back to single-use rockets for access to space. Today, NASA is the dominant shuttle customer. The RAND task force could not envision an event that would lead to a sharp rise in demand. Daniel Goldin's vision of a fleet of orbiters operated like an airline, possibly with corporate astronauts, has not been realized and probably never will be. If not the lack of a commercial market, then safety and reliability concerns probably will keep the shuttle a government program with government employees. "The launch director, the flight director, the program managers are probably going to continue to be NASA folks. It's still a NASA-led program and we're satisfied with that," says McCulley.
Despite the bleak outlook for privatizing the shuttle, would-be orbital entrepreneurs say it's time to open the space frontier to private enterprise. The only means of U.S. transport to and from the International Space Station, the shuttle has been grounded indefinitely. Small launch companies could jump into that breach. "If there really are profitable market opportunities, providers will exploit [them]," says Edward Hudgins, an expert on privatization and regulatory issues involving space. "One should expect and hope that someday we will not need a NASA."
Beth Dickey is a Washington-based writer who reports about NASA and human space flight.