defense watch

August 2006

Global Industrial Trends Spell Doom for Defense

By Sandra I. Erwin

The well-publicized intent of the U.S. military to remain on technology’s cutting edge appears to overlook the predictions of the Pentagon’s own soothsayers, who warn that the Defense Department soon may be unable to acquire critical microelectronics.

Simply put, a large portion of the industry that manufactures the most sophisticated microchips — which are key to modern weapon systems — resides offshore, and even those suppliers that remain in the United States view the military as an undesirable customer.

The problem was laid out in excruciating detail in a 118-page report that the Defense Science Board published last year.

“The Department of Defense and its suppliers face a major integrated circuit supply dilemma that threatens the security and integrity of classified and sensitive circuit design information, the superiority and correct functioning of electronic systems,” the report says.

As surprising as this may sound, this is one problem that the Pentagon cannot solve by throwing money at it. Even the largesse of U.S. defense spending cannot compete with the thriving global market. The Semiconductor Industry Association estimated that global sales for semiconductors in 2004 were $214 billion. The Defense Department’s share of the demand is 1 to 2 percent.

William Howard, a former Motorola executive who chaired the DSB panel, says the implications of current industry trends for the Defense Department are “perverse and far reaching.”

These trends are “directly contrary to the best interests of the Department of Defense,” Howard tells a seminar hosted last month by the Center for Strategy and International Studies, in Washington, D.C.

Only two U.S.-based manufacturers — Intel and IBM — are able to produce the cutting-edge semiconductors that the Defense Department requires. But Intel supplies standard “off-the-shelf” chips and has no interest in allocating any of its production capacity to make military-unique sensitive electronics. Only IBM has agreed to segregate a portion of its production line to military-unique classified devices.

During the past decade, the government attempted to boost its in-house capacity by setting up facilities operated by the National Security Agency and the Energy Department’s Sandia laboratory. But those efforts largely failed, says the DSB report.

The economies of the industry dictate that large volume is key to profitability. The relatively small military buys make the Pentagon an unattractive customer.

It costs at least $26 million to design a leading-edge circuit, says Howard. To recoup its costs and make a profit, a company has to sell millions of chips. For the Pentagon, one hundred thousand would be a huge order. As a result, since the 1980s, the Defense Department increasingly has shifted to commercial off-the-shelf technology as much as possible. But there is still a need for microelectronics in classified and sensitive programs that require special designs and secure manufacturing facilities. If the Pentagon had to rely on foreign firms to design and make these secret technologies, the security of classified information embedded in chip designs could be compromised. Risks include “Trojan horses” and other backdoor features that could be used by an adversary to disrupt military systems.

Because no commercial firm makes radiation-hardened circuits, the Defense Department subsidizes two companies: BAE Systems and Honeywell, to operate those production lines. Intel refuses to make radiation-hardened Pentium chips because these devices are on the “munitions list” and cannot be sold to foreign countries without a State Department license.

For commercial companies, he adds, the Defense Department is the “customer of last resort … the worst customer in the world to deal with.”

In addition to security restrictions, another disincentive is the Defense Department’s chronic inability to forecast its future demand. “When you ask defense acquisition people what they will need in the next five to 10 years, the answer is ‘I don’t know,.’” Howard notes.

More broadly, the lack of national policies and top-level direction for how to reverse the decline in U.S. microchip manufacturing is an even more alarming concern, says George Scalise, president of the Semiconductor Industry Association.

The U.S. government in many ways still naively believes that we are in the 1950s, leading the world into the next generation of technological excellence. The United States still holds the top spot, but China is quickly catching up, Scalise says. “We have to figure out what we need to do to maintain our lead,” particularly as we head into the “nano-era” in 15 to 20 years, when current technology will be obsolete.

He blames the eroding U.S. competitiveness on the absence of financial incentives for manufacturers to open facilities here and restrictive immigration policies that keep talented students and professionals from coming to the United States. “This is not industrial policy. It’s about choosing to compete,” Scalise says.

As to what the future holds for the Pentagon’s ambitious plans to field a high-tech force, experts such as Howard are not optimistic. “They assume that contractors will fix the problem.” They talk about “freezing the technology” so they don’t have to worry about the rapid pace of advances in the industry. Those people, says Howard, “belong in the apparel industry.”

These circumstances bring to mind a truism attributed to former President John F. Kennedy, who was credited with inspiring the nation to shoot for the Moon. “There are risks and costs to action. But they are far less than the long range risks of comfortable inaction.”

Please email your comments to SErwin@ndia.org

Back To Top