Q&A with Frank Kendall, Under Secretary of Defense for Acquisition, Technology and Logistics

How does the proposed FY 2017 defense budget address the challenges of balancing modernization, readiness and personnel?

I think we struck a reasonable balance.  The Secretary has been pretty articulate about the fact that this is a threat driven budget and it’s driven by changes in the world.  He has a list of five problems. He doesn’t try to prioritize them because we have to deal with them all:  China, Russia, Iran, North Korea, ISIS and extremist groups.  And that means we have to have a mix of current capabilities and capabilities that are ready to be employed very quickly to support current operations.  We also have to worry about high end competitors, Russia and China particularly, but to some degree Iran, that have reasonably capable conventional forces, and in a couple of cases nuclear forces.  Getting the right mix of capabilities and the right balance between what you have today and what you’re going to have tomorrow is what we try to do and I think we do a reasonably good job of it.

What areas of the budget really reflect the Secretary’s strategic thinking?

I think it’s the focus on those five specific problems.  After the Cold War ended we started building budgets around generic regional problems for a while, without being very specific about what they were. Most recently, North Korea was one and Iran was one to some degree.  And we had a strategy that was based on the capabilities to deal with two regional contingencies for a long time.  We modified that a little bit in 2012 with a strategic document that talked about phasing capabilities a little bit more differently.  This one takes us back to the fact that we now have near peer or peer competitors we have to worry about.  Since the last budget, Russia’s aggressiveness has become a greater concern.  I think it’s fair to say that China’s aggressiveness in the South China Sea has become a greater concern.  Kim Jong-un’s actions with regard to nuclear weapons and their recent space launch are all factors.  Plus the fact that the ISIL fight in Syria and Iraq is a new development.  And even though we have the deal with Iran as far as its nuclear weapons program is concerned, they are still doing some things in the region which we’re concerned about in terms of how they are trying to extend their influence in other parts of the Middle East.  So we have a set of very specific problems we’re trying to balance our capabilities for.  I think the threat focused basis for the budget is a return to a previous era in a way and I think it’s the right thing to do in the current circumstances.

What does our industry need to be thinking about in terms of developing over the horizon capabilities?

There’s no question that automation and artificial intelligence are going to be at the center of future weapons systems.  The degree of autonomy I think is somewhat questionable.  When you start moving from humans onboard actively controlling, to humans controlling from a distance, to humans not controlling at all, there’s a spectrum there.  The technologies that are being developed in the IT world, some of the things that are being done with sensing, and the ability to parallel process massive amounts of information to do improved pattern recognition are moving machines closer to human capabilities.  And people will exploit that.  That technology is happening with self-parking automobiles, and the Google car has been driving on the road for some time.  There are people experimenting with things at sea. We’ve been doing unmanned air vehicles for a long time.  We’ve probably had less success with tactical ground vehicles. We’ve tried to do it a few times and never quite got there—at least for tactical behaviors.  But there’s no question of the trajectory that we’re on.  And the question is: Are we going to be on pace or behind?  That’s an area where people should be thinking very carefully about in terms of where to invest in new products and new capabilities.  And it spans a lot of things.  The world of electronic warfare and the world of cyber warfare move at speeds that are incomprehensible to humans, where decisions have to be made and actions have to be taken much faster than the human brain can react in certain areas.  And I think those areas will expand over time.  There is a risk of finding ourselves further behind in that competition than we would certainly like to be.

What was the thinking of DOD’s decision to create an Electronic Warfare office?

It came out of a recommendation from the Defense Science Board about two years ago.  I felt that we had been neglecting electronic warfare.  At the end of the Cold War with the lack of a concern about a peer competitor that the Department had, we had not done as much in that area.  And as budgets got tight there tends to be an emphasis on “platforms” over “capabilities on platforms” like electronic warfare. In 2001, of course, we started a counter insurgency campaign.  So we did work electronic warfare extensively for IEDs.  But of course that is a small subset of the problem.  Meanwhile, others, I’d say notably China but also Russia, were also investing in electronic warfare capabilities to control the information fight and to give themselves a tactical advantage.  And we had fallen behind.  At least we were not investing at the pace we needed to be investing.  So I commissioned the Defense Science Board to take a look at that situation and make a set of recommendations.  One of them was that we reestablish a focal point for leadership at the Department level.  And that’s the reason for the committee (on Electronic Warfare) that Deputy Secretary (Robert) Work asked the Vice Chairman (Air Force Gen. Paul Selva) and me to establish, as well as for recreating an office devoted to electronic programs.

What do you hope to achieve with your new strategy for satellite launch services?

We’re trying to do something that’s a break with tradition.  It’s more of a commercial model.  What the department needs isn’t engines or rockets. It needs the service of putting our satellites into space.  And what’s happening up there is that commercial space is changing.  There’s a big demand function coming from the commercial space sector for launch.  That’s one factor.  Our traditional launch service provider, United Launch Alliance, has performed this service extremely well for a long time, but it’s also been a relatively costly service.  And it is now encountering some competition from SpaceX.  SpaceX is developing largely with the commercial market in mind, but it also wants, as an ambition now, to be certified to do national security launches.  There’s a change on the technology side.  There’s a change on the demand function on the commercial side.  There’s the fact that we want services versus hardware.  And there’s an opportunity in that for us.  The opportunity is to set up business arrangements with launch service providers that basically say, “we will give you an investment in developing your capability.  In return we want future launches from you at a reasonable price.”  And so that’s the type of public-private partnership that we want to strike.  And we’ll do it competitively so the people who want to be in the launch service business will request or bid to get some money from the DOD to close their business case, to develop their systems.  In return we’ll get some guarantees of future launches at a competitive price.  And the idea of this is we would fund, depending on the amount of money required, two or maybe even three launch service providers to put them in a position to give us the launches we’re going to need in the future.  So that’s the idea.  I think it’s a very sound idea. It’s a non-traditional approach but I think it makes an awful lot of sense given the environment.  We’ve been urged by a number of people to do things that are more commercial like.  I think this is a good example of something that is more like a commercial business deal.

The other factor that is complicating the situation is the concern about the Russian RD-180 engine. And that’s a separate issue, but they’ve all gotten merged in together. We’re all in agreement – I think that we want to end our dependency on the RD180 as soon as we can.  But there’s been disagreement about the best path to do that. We would like to have access to RD180s for the next few years while we mature those commercial launch service providers that I talked about a moment ago.  We do not want to be in a position of having to depend on only one launch service provider.  SpaceX is relatively new to the national security launch business.  ULA is very reliable and very proven.  So to discard ULA, and their Atlases if you will, until we get more modern systems seems like a risk to us.  We would prefer to keep them around until we get further down the road to bridge that gap. And that takes a small number of RD180 engines to get there.  The exact number is not definite, but right now we’re at nine.  The authorization bill last year got us to a total of nine.  We think maybe twice that number is what we’d like to have to be comfortable.  But there’s a lot of controversy about this.  Our fundamental concern is having assured access to space.  That’s incredibly important to the nation. We don’t want to see money flowing to people who are being sanctioned in Russia.  If sanctions are extended to those engines than obviously we’ll follow that.  But in the meantime we are very concerned about assured access to space.

The other factor is cost.  There’s one other option which would be to use the existing Delta system for the interim period.  That would be very expensive.  Delta launches cost about $30 million more per launch than an Atlas launch.  And the Delta capacity to support launches is limited.  We’d have to make a major capital investment to increase it.  So that doesn’t look like a good investment, particularly if it precluded getting on with the public-partnerships I’ve talked about. It’s a complicated situation made more complicated and controversial by the Russian engine situation.

How can we make the acquisition system work through your “Third Offset” strategy to achieve the goals of better handling technological change?

We have, for some time, been doing things to make it easier to bring newer technologies into our systems. Modular open systems are ways we do that.  You have to impose that on the design. You have to get the intellectual property to control the interfaces and you have to structure a program so you have that option.  The fact of the matter is that our major capital investments are platforms that are going to be with us for quite a long time.  So we have to build those in a way that allows technology to be inserted.  And, most of the technologies we’re talking about are not end items.  They are technologies in the IT area, for example, in the radio frequency area that can be inserted onto other platforms as modernization efforts as opposed to brand new products.  That’s one of the major things we’re doing.

The other thing we are doing is a lot of experimenting in prototyping.  We put a number of things into the budget this year to basically do proof of principle or technology risk reduction prototyping.  Secretary Carter mentioned  in a speech recently that one thing we’re doing is the Arsenal Plane concept.  The idea is to reduce technical risk so that we can go into development later with less technical risk to give the operational community the chance to develop and experiment with operational concepts, and to position ourselves a few years from now so that we’re ready to make a decision to make a major commitment, assuming that we have the resources.

So that’s what we’re doing.  And I think those are the right steps to take. I don’t think the acquisition system itself is the culprit here. I actually think the acquisition system has been improving steadily over the last several years.  Cycle times are down.  Cost growth and schedule slips are down substantially and dramatically in some cases.  And, we’ve looked at that data every way we can to make sure that those trends are real, that we’re really seeing valid results there.  And, as far as we can tell, we absolutely are.  So we’re doing the right things in acquisition.

The one thing that would be helpful, and I’ve talked to the Congress about this, is the ability to start work sooner.  With our budgeting process, it takes on the order of two years from the time a threat emerges until we can start to spend money on the response to that threat.  That’s way too long. It’s wasted time. So if we had the authority and the resources to start right away, without making a long term commitment, just to get the early front-end work done, it would be enormously helpful.  And if Congress would give us that authority, I think we could do a lot to further the “Third Offset” strategy to reduce the lead time to getting new capabilities in the field.

If we ever had a national emergency arise where we have to throw all the rulebooks out the door, could the acquisition system work on a rapid pace basis and could industry work to build new systems in less than ten years?

It actually takes much less than 10 years.  People point to the F-35 like it is “the” example.  It’s very much an outlier.  We can do things faster and sometimes we do.  The MRAP (Mine-Resisted Ambush Protected) is a good example of that.  What drives a lot of the time in developing a product is, first of all, the complexity of the product and how many new things you have to do.  And that’s where a collection of existing components worked.  We took existing automotive components and put them together in a package, and we didn’t worry a great deal about secondary considerations.  The consideration was the survivability of the people in the vehicle.  So we wanted a truck that people could survive an IED (Improved Explosive Device) in.  That was it.  And we set out to get that and we turned industry on.

We are criticized constantly for our business deals, schedule slips and cost overruns.  When there’s a crisis, you just take the risk. You just get started and cost is not the driver.  Capability is.  You don’t worry about every requirement you might want.  You focus on the things that matter.  I have a couple of examples from the Cold War where threats emerged and surprised us.  One is the explosive reactive armor on armored vehicles that we saw the Soviets field.  Another was the terrain bounce jammer which we detected after they fielded that capability for anti-ship cruise missiles.  In both cases we put crash programs together that involved multiple agencies in parallel trying to find solutions.  And we came up with some things. They weren’t perfect but they dealt with the problem.  So we can do that.

I’m going to go back to the MRAP story because I think it’s really instructive.  The MRAPs that we built originally were built for use in Iraq.  They were done in a hurry with existing components.  When we did the surge in Afghanistan we needed a lot more capability like that in Afghanistan.  And we discovered that the MRAPs we bought for Iraq were not going to work very well in Afghanistan.  And we went out and bought a bunch of additional vehicles that were designed differently.  And what we changed was the size of the vehicle for maneuverability in the terrain there, and also the suspension system.  The original MRAPs were leaf-spring suspensions and they didn’t work very well. And we went to a dynamic suspension system in Afghanistan.  In thinking about that history if we had not been at war, and people were not dying, we would have carefully analyzed requirements for a vehicle to go over a variety of terrain in a variety of conditions; and we would have probably come up with something more like the second vehicle that we built, and done that in the first place.  There’s a role for both.  There’s a role for thinking carefully about what you want and design the reliability in.  One of the things that you don’t get when you go fast is reliability. It takes a while to design reliability in.  Many of the systems we have used in the wars in Afghanistan and Iraq—we have done so at great operational cost because of the support requirements.  We have just been using parts as a way of accomplishing availability because we didn’t have reliability.

Another example where we went fast and just bought more of the prototype we built was Global Hawk.  The original Global Hawks were an advanced concept technology demonstrator.  They were not designed to be production assets.  They were designed to prove out a technology and a concept.  We liked them so we bought more of them.  And they were notoriously unreliable.  So, as a result of that, we were going to cancel the program and keep the U-2 around forever.  What turned out is that was a great motivator for Northrop Grumman.  They worked very, very hard to get the reliability of that system up, and brought the operational costs way down.  And, about two years after that first decision ,we reversed it because the costs had come down enough and the economics split the other way.  It made more sense to keep Global Hawk and retire the U-2.  It was driven by that. You have to decide what you want and go after it.  And there are plenty of other examples.

Right now we are working hard to replace the kill vehicles in our missile defense system for national missile defense.  That’s another case where we took an experimental vehicle and started building more of them.  Then we started failing test flights.  And we said, “Ok, we’ve got to redesign this thing to be more reliable.” And that’s what we’re doing now.  It’s always a trade-off.  There’s a price to be paid for going fast.  Benjamin Franklin was partly right in this regard – haste does make waste.

How can DoD-industry collaboration be improved on the rule to certify adherence on cybersecurity requirements?

We’ve relaxed the implementation period to give industry more time.  We put it out with an immediate (implementation) requirement, and in a couple of steps we’ve backed away from that. Cybersecurity is a pervasive problem, it’s a continuous problem.  We have to increase our standards of protection.  We’ve done a couple of things to do that for unclassified technical information that may not be so sensitive in terms of the knowledge imbedded in it, but it gives a competitor an enormous amount of time and money advantage to have that data.  There’s no question that our industry has been under attack continuously and a lot of stuff has been stolen from us.  So we have to “up” our cybersecurity protection for that data.  That’s what we’re doing now.  We’re happy to work with industry to do it in a reasonable way that doesn’t impose too much of a burden, which is why we’ve backed off on the timing.  But there’s no question in my mind that we’ve got to have greater protection here.

What are your thoughts on DoD’s work with AIA?

I think it’s been a very good professional collaborative relationship.  I can say that of the other associations as well.  We are very open to hearing from industry.  We don’t always think about all the implications of what we want to do.  And we want to avoid unintended consequences that are very negative.  We also want to hear innovative ideas, and the only way you can do that is if we talk.  I’ve pushed very hard on our acquisition people, our contracting and program management people, to have a dialogue with industry right up until the moment when the RFP goes out and you have to stop at that point because it’s a competitive situation.  I’ve had to get past some very nervous lawyers to do that; but it’s absolutely, to me, the right thing to do.  The more we talk to each other the better, as far as I’m concerned; and, I appreciate what AIA has been able to do to further those conversations.