You get what you pay for.”

That is what multiple industry officials have said of the embarrassing, third failed intercept test in nearly five years for the $41 billion U.S. anti-missile system developed over more than 20 years to defend the U.S. against North Korean ICBM attack. But proponents of the Ground-based Midcourse Defense (GMD) charge that the failure is directly related to five years of diminished funding for development.

The last successful kill by GMD, which is managed by Boeing, was in December 2008; program proponents charge that recent failures are directly related to five years of diminished funding for testing and redesign. The lackluster July 5 trial—a virtual repeat of the 2008 test—was hastily added to the development plan by Defense Secretary Chuck Hagel in March as a show of force related to North Korea's third nuclear test, which took place in February. But, if Hagel hoped to showcase U.S. deterrence as a response to that saber rattling, the test was nothing short of a whimper because some industry insiders believe a hardware problem might have prevented the intercept.

“We have confidence in our system,” Hagel said during a March 15 press conference in which he announced plans for the July 5 flight demonstration and to buy another 14 Ground-Based Interceptors (GBI) as an added measure of security against North Korean attack. “We certainly will not go forward with the additional 14 interceptors until we are sure we have the complete confidence that we will need.” So, it seems, not only are the interceptors that are already fielded of potentially questionable value against an attack, based on the results of the recent test. But plans to buy newer missiles are likely on hold pending a laborious review of the flight-test failure.

For that trial, a GBI already fielded and on alert was launched from Vandenberg AFB, Calif., against a Lockheed Martin LV-2 target lofted from the Kwajalein Atoll in the Pacific. Except for use of a newer target missile, the geometry of this trial was the same as that of the successful December 2008 kill, and it should have been an easy redo.

“We're concerned when any test is unsuccessful . . . and we're concerned now,” Pentagon Press Secretary George Little said. “There are going to be glitches from time to time, and clearly there are some glitches here . . . . We'll try to find the root cause of the problem here. But we are confident in what is, I think, by any measure, the most robust missile defense architecture in the world.”

The U.S. Missile Defense Agency (MDA) did not answer questions about the sequence of events but acknowledged that an intercept was not achieved. Spokesman Rick Lehner did not say whether the interceptor, which includes an Orbital Sciences booster, managed to execute its anticipated flightpath and deploy the Exoatmospheric Kill Vehicle (EKV), a Raytheon system designed to destroy a warhead with a direct hit. He also declined to say whether the test involved the use of countermeasures from the target, which would confuse the radar and optical sensors used by the missile defense system to hone in on a mark.

Multiple industry sources suggest that the booster flew as expected until it failed to deploy the EKV, eliminating the chance for a kinetic kill.

If this is the case, the critical question is whether the failure is a result of a systemic design flaw that could jeopardize the functionality of the interceptors at Vandenberg and at Fort Greely in Alaska; these missiles on alert are the only line of defense against a North Korean intercontinental ballistic missile attack.

What these officials hope is that a hardware failure is the result of a single manufacturing defect or human error. Although embarrassing, this scenario would be the lesser of the evils because it would have fewer implications for the missiles already in silos.

At the very least, an immediate operational upshot from the test is likely a shift in the so-called shot doctrine. The Army, which operates GMD, has rules that dictate how many interceptors or “shots” are dedicated to each target to ensure a kill. The more reliable the system is thought to be, the fewer interceptors designated per target.

Now, the GBI with the “Capability Enhancement-1,” or CE-1, kill vehicle has achieved eight of 14 intercept attempts.

An increase in the number of missiles assigned to each target diminishes the number of ICBMs the system could handle in a short time, according to one missile defense industry official. “What this test says is that the next time it may work or it may not work,” another official says. “Deterrence is modestly degraded.”

The failure, while vastly disappointing to those in the industrial and government sectors, was not a surprise to all. “At the end of the day, you get what you pay for,” this official says.

The dismal performance is the result of more than five years of diminished funding for GMD development, he says. Funding earmarked for development was actually used to buy GBIs—even though the design was immature—or to fund other activities, such as building a third GMD deployment site in Poland, a plan that was subsequently derailed.

The offset was a gutting of what should have been a consistent program of testing and improving the nascent GMD design, including attention to the GBI, EKV and command-and-control links, the official says.

“What failed . . . was the culmination of a five-year process failure—a huge process failure between the Defense Department and the Hill—that has led to a suspension of the normal redesign and reengineering activities on GMD,” the official says. In 2004, President George W. Bush announced limited defensive operations with GMD despite its embryonic status; only years later officials abandoned what would traditionally be an ongoing plan to repeatedly test and redesign the system until it was reliable and producible. The blame for this shift rests on the Bush and President Barack Obama administrations, this official says. Program overseers were forced to simply monitor the interceptors and kill vehicles already built and fix issues that cropped up in inspections.

“We never got to the 'redesign point' there. This idea that you can inspect in quality and mission assurance over the long term . . . is crazy, but that is the path we are on now,” the official says.

This conundrum is unique to GMD among the MDA's major programs. In the case of the Aegis, sea-based regional missile defense system, which employs the Raytheon SM-3IA and now IB, the Navy and MDA took a methodical approach to incrementally testing and designing fixes. Today, Aegis has become the system of choice for combatant commanders in Europe, the Middle East and the Pacific regions.

Likewise, the Army went so far as to stand down testing and production of the Lockheed Martin Terminal High-Altitude Area Defense (Thaad) system, an improved, land-based successor to PAC-3, owing to repeated failures. After a decade of lackluster testing and a substantial redesign, however, the system reemerged and accomplished a starkly improved intercept record. Thaad has garnered enough confidence already to be sought by allies in the Middle East.

Part of the difference in philosophies behind these programs could be attributed to the political nature of GMD. As the only system protecting U.S. territory, a stand-down to allow for redesign is not and was never an option since Bush announced defensive operations in 2004. The very presence of a GMD is required for the U.S. to continue a dialogue and policy of deterrence against such adversaries as Pyongyang and Tehran.

The flight failure comes not only at a sensitive time relative to North Korea's bellicose behavior. It is the first major flight trial under the leadership of MDA's Director, Navy Vice Adm. James Syring, who took over in November. Under his predecessor, Army Lt. Gen. Patrick O'Reilly, who had a stern leadership style and a notorious bent for risk aversion, the agency's technical ranks and morale eroded along with its testing tempo.

Senior Pentagon leaders hoped to improve the latter with three GMD tests slated for this year alone. During the first, the GBI conducted a controlled test flight using a retrofitted EKV Capability Enhancement-2 (CE-2) system designed for improved warhead discrimination over the CE-1 that flew in December 2008 and this month. The CE-2 had failed in two attempts in 2010, and the January flight built confidence that a vibration problem was solved. An intercept trial for CE-2, previously set to take place by year-end, would pit it against an LV-2. It is uncertain, however, when it will take place as program officials are now focused on the failure review for the July test.

The July trial was also the first for Boeing under its new GMD oversight contract, a $3.5 billion deal won in 2011 over a Lockheed Martin team. Boeing's win was attributed to technical merit and aggressive pricing; compared to its legacy GMD contract, Boeing's price plummeted about 35%. But, industry officials say Boeing had to reduce its core of engineering experts in order to reach that price. The result is a “less deep bench” of manpower to put on the problem, one official says.

One solution, according to multiple industry sources, is to “stick to the knitting” and reinstate a costly but effective program to hone the GMD design, allowing for retrofits to existing missiles and improved reliability for yet-to-be-purchased interceptors. This will not come cheaply, however. Such an effort is estimated to be $1 billion annually, a figure likely to raise eyebrows at the Pentagon and on Capitol Hill. “The cost of not improving it versus fixing it is actually trivial,” one official says, noting the lost effort and time spent on failure after failure.

This approach would allow for the Pentagon to have a more reliable fleet of GMDs beyond 2030. And, with higher reliability the Army is likely to allow for fewer GBIs dedicated to a target, enabling a defense against a larger number of threats.

In the meantime, the source says, MDA can focus on developing and fielding a next-generation solution that includes a more reliable booster and kill mechanism for the long-term.