Commentary | The Essential Revolution of the NRO: A Second Opinion


I read with interest John P. Stopher’s article “The Essential Revolution of the NRO” [Commentary, Oct. 14, page 19]. I was a CIA officer working in the U.S. National Reconnaissance Office’s (NRO) Program B from November 1967 to August 1985 and served as director of CIA’s Office of Development and Engineering from April 1982 to August 1985. As one of the “Barbarians” that Mr. Stopher discussed I have a definite view of what is needed to make the NRO the organization that many of us think it should be.

There is much in Mr. Stopher’s article with which I agree. Many companies and government organizations have experienced the evolution that he discusses, and many of the changes that have affected the NRO’s effectiveness over the years are, as he discussed, valid. It is his conclusion and recommended fix (an NRO career service) with which I do not agree.

Some of Mr. Stopher’s observations are worth mentioning. His observation that “the competition from the Air Force, the CIA and the Navy was not perceived as particularly beneficial” is, at best, his opinion. I think that many of us in the midst of the competitions (mostly between Programs A and B) to this day believe the competition was highly beneficial. This competition fostered creativity and effective solutions to the serious intelligence issues facing the country and the country benefited from this competition with better solutions than might have been otherwise imagined.

In like manner nearly every independent review of the NRO (and I have been involved in several) has bemoaned the loss of authority and financial control of the NRO program managers. I was program manager of one of the NRO’s largest and most sophisticated (and successful) programs, felt like I was in charge of the program and made decisions that today’s program managers can only dream about. In addition, the program had significant margin, which I was in control of. So what are the real issues that need to be addressed? Here are a few:

  • Programs A/B and C were much leaner organizations than today’s NRO. In Program B we had one lawyer from the Office of the General Counsel who worked with us on contractual and other issues. The lawyer provided advice to the program manager and contracting officer and was not in the contract approval process. The lawyer often worked with us to write language in contracts that allowed us to do things not normal in the Federal Acquisition Regulation.
  • The program manager was in charge of the program (true in A/B/C). He or she was responsible for the budget and was expected to deliver the program on time and within the budget promised. The program manager held and controlled the program margin. This is how we were able to deliver all our programs in less than six years (vs. today’s 10-12 years) and nearly always within the budget promised. Today there is nearly no margin in programs (30-50 percent was the norm in Program B) and if there is, somebody else in the chain holds it.
  • The focus of today’s NRO is delivering what many refer to as “commodity” products. There is no question of the importance of these products to the various customers. This fact, however, has made the NRO risk-averse, not willing to risk any disruption to the delivery of commodity products to users by doing something new and different.
  • The NRO unfortunately has become a captive of the “functional managers” and the resultant requirement process. In the old days, the functional managers did not drive system design and we were able to focus on inventing and delivering systems to solve a national need. If left to the functional manager at the time, some of the important systems that the community enjoys today never would have existed. We built imaging systems to find Soviet missiles (Corona, Hexagon), to understand Soviet weapon systems (Gambit) and to provide near-real-time indications and warning of Soviet and Chinese forces. These systems performed magnificently and arguably were major contributors to the U.S. winning the Cold War. But from a people perspective (which is one of Mr. Stopher’s major points) we in the CIA saw ourselves as intelligence officers first and satellite builders second. Our job was to provide important information on important national issues to the intelligence community. People who see themselves as responding to the “requirements” process of the functional managers might see themselves differently. But this view of ourselves was certainly strongly supported by being in the CIA with the ability to “rub shoulders” with our compatriots in the Directorate of Intelligence. 
  • Monopolies inherently become self-serving and lacking in imagination. The NRO today is certainly a monopoly, vastly different from the Program A/B/C days when competition of ideas for programs was the rule. This competition forced innovation, as a run-of-the-mill solution was certainly not going to win. Indeed, even the contractors were more innovative during those days. Today, it is clear that the NRO will not tolerate any ideas that compete with its view of the world and what programs should be procured. As I noted earlier, the NRO does a great job of procuring systems that produce the commodity products that the intelligence community and military rely on. Without a marketplace for competing ideas, however, there is no incentive for the NRO to innovate.
  • There is no question that Mr. Stopher’s assessment of the way people were managed in the old days is correct, and both the Air Force and CIA have lost the bubble in this regard. But the solution is not to create an NRO workforce, as he proposes. This would not solve the problem and would further remove the staff from understanding the issues facing the intelligence community (at least for the CIA staff). We had a process in the Office of Development and Engineering for training people by assignments — a method for deciding which people would be good system engineers and program managers. We had a succession planning process that identified people for further promotion and planned what assignment or training was needed for them to get those assignments. Nothing stops such a process from being used today. Our processes did not require “certification” of program managers (which does not produce competent program managers). The method that we used of different assignments allowed us to train people much better than any formal training program. Further, we had a stream of people transferring into the Office of Development and Engineering from the National Photographic Interpretation Center, the Directorate of Intelligence and other parts of the CIA, which gave us an inherent understanding of the needs of the intelligence community. Very important, both Program A and B had people in very senior positions who protected their people working on NRO programs and saw to it that their careers were advanced by a long-term relationship with NRO programs. Any perceived issues with the Air Force and CIA organizations are fixable if the NRO director and the director of national intelligence really want to fix them.
  • To me, the most important consideration is what do we want the NRO to be? The NRO is not like NASA. NASA conducts space exploration in support of its own mission. It does programs for its own purposes as it has a specific mission to explore space and advance our understanding of space, its environment, the planets, etc. The NRO does nothing for itself; it has no NRO-only mission. Its job is to produce information for the intelligence and military communities and it happens to do that via overhead collection.

Allowing the NRO to have its own workforce has the real potential to turn it into an organization that sees its role as building satellites. This is a role quite different from how we saw ourselves in the CIA and the Office of Development and Engineering.

So what are the remedies for this? There are a few that would make a difference:

  • Shrink the NRO; it has become too big. Pare down the staff so that the focus is, once again, on program management. In addition, there are too many Systems Engineering and Technical Assistance contractors and Federally Funded Research and Development Center (FFRDC) staff supporting the NRO. Many simply “support” and do not conduct any significant technical work. Plus, they tend to dilute the responsibility of the government staff who need to be held accountable for their jobs. As mentioned, I was the program manager of the largest, at the time, NRO program. We got by just fine with a total of five FFRDC folks who provided specific technical talent that we needed and did not have in the government staff.
  • The Air Force and CIA personnel issues need to be solved. The fact is that both organizations bring talents and skills to the NRO that are unique to their cultures and that the NRO can benefit from if managed effectively. The CIA staff tends to be creative, knows how to do streamlined acquisition and has a history of rapid, on-time, within-cost performance on high-risk programs. An arrangement similar to the Program B days needs to be created. This would benefit the NRO and further cement the NRO/CIA relationship.
  • A major look at the organization of the NRO is needed. Focus on streamlining, creating a culture of effectiveness and innovation by eliminating layers of unneeded management. For example, abolish the position of tower director and have the major program managers report directly to the NRO director. This would put an emphasis on the program manager position and authority. There are not enough programs in the NRO to justify the tower director position; they are there as a by-product of the old directors of Program A, B and C positions. Second, abolish the Ground Enterprise Division and go back to the days when the program manager had real end-to-end responsibility. The legitimate desire of the NRO to have its ground stations better integrated can be done cheaper and better with decent system engineering at the NRO level.
  • There was a time when the NRO was overseen by an executive committee. What programs the NRO did were decided by a committee comprised of the director of central intelligence, the deputy secretary of defense and the president’s science adviser. This ensured that programs served a real national need and were not bound by “requirements.” Such an oversight process could elevate the NRO from the debilitating “functional manager” nonsense and recreate a focus on doing things that are perceived as really important. Such an executive committee might consist of the director of national intelligence, the deputy secretary of defense and the national security adviser.
  • Industry uses technology to increase productivity and shorten product development timelines. When we think of innovation in the NRO, people usually are referring to using new technology to develop new and creative systems. But the fact is that the NRO needs to reinvent how it does business as well. The evolution of the design, analytical and performance tools available today allows a serious reduction in the time it takes to design NRO systems. In addition, many of the program processes and procedures that have been developed over the years can be streamlined. Some of the increased time it takes to develop NRO systems today (versus the old days) is the result of increasing bureaucracy sneaking into the program development process. Innovation here is needed as well.


Robert J. Kohler is former director of the CIA’s Office of Development and Engineering.