Lessons on restructuring the defense industry from the Virginia submarine program

The price of American submarines is coming down. As announced early this week, the US Navy has signed a contract with General Dynamics Electric Boat (GDEB) and Huntington Ingalls Industries (HII) to deliver, from now through 2023,  ten Block IV Virginia-class submarines for $17.6 billion. What’s remarkable is how that price was reduced by what seems to be a monopoly producer, but really isn’t. And after a success like that, the Pentagon might consider where else the model can work.

The Congressional Research Service reports that the price of the first Block III ship of the class, the North Dakota, will be about $2.8 billion, which is somewhat above its target cost of $2.6 billion. (See also the articles in the Wall Street Journal and Defense News for good overviews.)  Admittedly, not everything has gone smoothly with the transition from Block II to Block III. But even after some unplanned rework, construction of the North Dakota, the first of the Block IIIs, is still ahead of schedule. So even without considering the improvements that will be built into the Block IVs, a price of just $1.76 billion per ships constitutes a huge reduction for a program that is already performing rather well. Even better, the Block IVs are intended to have lower lifecycle costs, in that they should require only three overhauls for every fifteen deployments (today’s ships take four for every fourteen).

What’s remarkable is the work-sharing deal between GD and HII. The submarines are fabricated and assembled between three shipyards: Quonset Point (GDEB), Groton (GDEB), and Newport News Shipbuilding (HII). Ten have been delivered, and another eight are under construction. Each company has been guaranteed a certain work share in past contracts for Virginias, but each retains an incentive to perform well, as the Navy could upend that arrangement at any time, and accord more work share in future contracts to the more efficient shipbuilder. In effect, the Navy has been recompeting its business on every ship without needing the threat of sinking one of its only two suppliers for submarines.

Indeed, it’s remarkable how the arrangement between GDEB and HII is working comparatively better for the government than the joint venture of Boeing and Lockheed Martin in United Launch Alliance (ULA). Ten years ago, heavy satellite launches cost about $400 million. Eight years ago, those two companies were allowed to combine their businesses to make a run at drastically reducing costs. Today, ULA’s launches still cost about $400 million. It’s taking a lawsuit by Space-X to challenge the sole-source award of 36 more shots to ULA. Had Boeing and Lockheed been directed to form a structure with more scope for competition, that might have gone differently.

A similar question arose last week at the Bloomberg Government 200 conference. Linda Hudson, the former CEO of BAE Systems in the US, suggests strongly that more mergers in defense would be necessary. From her experience in tracked armored vehicles, she specifically asked “how do you keep two competitors in a business where there’s barely enough work for one?” But Elana Broitman, the Pentagon’s industrial policy czar, repeated the administration’s consistent policy that another round of conglomerate mergers wouldn’t be tolerated. Here, I side with Broitman: merger to monopoly is generally an unsound idea. But then again, if this cooperative-competitive arrangement between GDEB and HII is working, why not encourage similar industrial restructurings in other segments? 

As I wrote back in August 2013, the study that the Army Department commissioned from A.T. Kearney found massive overcapacity for building armored vehicles in the US—and that was before the cancellation of the Ground Combat Vehicle program. The problem is not that there are two contractors, but that each contractor has much more costly floor space and staff than it needs in the long run. My radical suggestion for consolidation: strong-arm BAE Systems and General Dynamics Land Systems into relocating alongside one another, at a single production cluster, perhaps at the Army’s main depot in Alabama. As I put it then,

     It would rather be the ground vehicle equivalent of Palmdale, California, where Lockheed Martin, Northrop Grumman, and Boeing all maintain industrial operations together employing thousands of people. If any one program flags a bit, otherwise excess staff can cross the street for another job on a rising program, repeatedly cushioning the adverse impact for the industry as a whole. This process should be familiar to the Army and the Marines, as it has been underway in Warren, Michigan since the middle of the last decade. With the consolidation of vehicle procurement management outside Detroit, most military vehicle manufacturers in the US set up engineering centers in what has become a rather affordable place to work.

The bankers won’t like it, for they won’t be earning a fee. The consultants might like it a little more, as someone will need to help rebuild these businesses. But the Pentagon should like it a whole lot, as industries can be secured with a reasonable degree of competition, but without the overhead of lingering Cold War infrastructure.

James Hasik is a senior fellow in the Brent Scowcroft Center on International Security.