Going forward, we should expect instead to have flat or declining funding for a long time. It is no longer realistic to defer costs that are unaffordable today to another year. Nor does it make sense to accept higher unit costs for forthcoming ships and aircraft and expect that there will be enough money to maintain Fleet size.
The unit costs of military systems have steadily swelled with each new generation that has been fielded in the last 70 years. Defense industrialist Norm Augustine humorously described this trend in his famous book Augustine’s Laws . 1 In 1997, based on trend lines for tactical aircraft unit cost, he calculated that by 2054, it would take the entire defense budget to pay for a single combat aircraft. In 2006 and 2008, the RAND Corporation did studies on the reasons behind both aircraft and ship unit cost increase, which showed a long-term annual rate of growth of about ten percent for ships and tactical aircraft, as well as root causes. 2
By comparing the unit costs of an F-35 to that of an F/A-18, or of a Ford -class carrier to a Nimitz -class, the RAND Corporation found that each new generation is more complex, more capable, often bigger, and therefore more expensive—by a factor of as much as two—than the previous generation, even if estimates at the program’s inception suggested this would not be the case. Capability demand drives cost, but each platform provides the same capacity regardless of price—and under many conditions, capacity is itself a critical capability. Regardless of how proficient it is, a ship can only be in one place at a time. A global strategy requires a global presence, which in turn relies on having a Fleet larger than we can afford if we pay ever-increasing prices per ship.
Achieving the Navy’s target of 306 ships would require an average shipbuilding budget of $18 billion per year over the next 20-plus years, even if unit costs stagnate. The current construction shipbuilding plan for fiscal years 2013 through 2019 is about $13 billion per year, close to the Navy’s 20-year historical average and a level unlikely to be exceeded over the next 20 years. 3 Simple math and more sophisticated cost-prediction models show that with today’s shipbuilding budget, the largest fleet of current ship designs that the Navy would be able to afford is 30 percent smaller than the goal—or about 220 ships. The situation is similar for aircraft. If ship and aircraft capabilities and unit costs continue to climb with the next generation, the Fleet’s size will become even smaller.
But what caused the increasing complexity, capability, and unit costs that are shrinking the Fleet in the first place? Clearly, threat capability is a factor: Weapon systems designed to challenge U.S. Navy access and effectiveness have steadily become more efficient and widely proliferated. The Fleet must be able to defeat threats, but the payloads needed to do this, such as weapons and sensors, continue to get bigger and heavier and demand more electrical power. Assuming that speed and range requirements remain constant, all of these factors will continue to increase ship size.
In the case of amphibious forces, the Marine Corps has significantly increased the size, weight, and capability of its vehicles and aircraft without a commensurate reduction in their numbers. That means that this payload also requires ever-larger Navy ships to embark them. Even the Navy’s “low-end” warship of the future, the littoral combat ship (LCS), represents an increase in size and capability over many of the vessels it replaces, such as mine countermeasures ships and patrol coastal small combatants. Meanwhile, rising aircraft costs have been driven more by growth in sophistication than size; the systems on new aircraft require millions more lines of expensive-to-develop software code than their predecessors did. With fighter aircraft, we have also gone down the expensive path of airframe stealth for survivability.
Today the Navy operates about 50 different types of ships and aircraft with individual design-service lives of 20 to 50 years. On average, about two classes of ship or aircraft annually come up for a decision on replacement at the end of their service lives. Each of these decisions, a multi-year joint bureaucratic process with dozens of participating organizations, is made individually. Typically, as a starting point, the new platform must do everything the old one did, except in the more challenging threat environment of the future. All of the decision-making organizations generally advocate for the next-generation platform to have the desired capabilities unmet by the old one—particularly since any additional unit cost is not their bill. It is no surprise that this process leads to steadily increasing platform and overall Fleet cost.
The future Fleet is being designed ad hoc, one platform at a time, and we cannot afford this. How can we change the trend toward an ever-smaller Fleet of ever-better platforms while maintaining the capability superiority needed to execute our missions? It will take a top-down design to provide a structure in which individual platform requirements can be shaped and disciplined despite all of the pressures. We will have to consider distributing capabilities to a greater extent across a force that is securely networked, at least within line of sight, rather than putting as many as possible on each individual platform and continuing to drive up its size and cost.
We will have to consider separating weapon magazines from the sensors that direct the weapons rather than putting both on the same platform. Another option is increasing reliance on deep-magazine directed energy systems, and on force-wide coordinated soft-kill and counter-targeting techniques, rather than on engaging each threat with ever-larger and more expensive kinetic weapons. We can also think about increasing reliance on penetrating high-threat areas with longer-range weapons or with preprogrammed unmanned systems rather than with manned platforms. Few of these options would rise to the top in the requirements decision-making process for any individual platform. They only start to make sense when considered and competed at a Fleet-wide level.
Developing an overall fleet design to structure and discipline individual platform requirements is no small task. Simply constraining platform cost without dealing with how capabilities might be delivered differently is not sufficient. This is not a once-and-done process, as changes in threat and in our own technology options will never stop. But neither can it be a process that changes the design in some fundamental way every year or two—it will have to influence platform requirements for a long period of time to affect a significant number of new platform designs.
We cannot afford to retire legacy platforms prematurely simply because they are not optimized within our new Fleet design, which will take time to implement and have to be done incrementally. Real and fundamental change in the roles, missions, and interdependencies among platform types, and in the balance between manned and unmanned and between platform and payload, is an inevitable outcome of a Fleet design process. That is the point. Change is hard, and it will have to be authorized and directed by the Navy’s leadership or risk not happening.
A number of ideas for a new Fleet design have been offered recently from outside the Navy’s decision-making mainstream. However, all have had significant flaws, so they have not received serious consideration. They have assumed things such as beyond line-of-sight networking that has no survivable future in the face of adversary counter-space capability; autonomy of unmanned vehicles in executing lethal missions that is beyond the projected capability of software and U.S. rules of engagement to support; and the use of platforms too small to be capable of global deployment and sustained sea-based operations, which is how the U.S. Navy must deliver global naval power. The future Fleet design must be grounded in technical and operational reality, and it has to come from inside the Navy system.
Realistic Options for Change
Developing a rich list of operationally-realistic options supported by rigorous analysis of cost and feasibility is foundational. It could include:
• The use of a common large aviation-ship hull for Navy sea-control/power-projection air wings and for Marine Corps vertical-raid/assault-air wings, reconfigurable between the two missions between the deployments;
• Surface combatants with smaller vertical-launch magazines that can reload at sea from logistic ships or remotely fire weapons carried in supplementary magazines on logistic ships;
• Separate classes of surface combatants optimized for air defense or antisubmarine warfare within a common hull type that can self-defend in peacetime but aggregate to fight offensively in wartime;
• Tactical-combat aircraft that are optimized for endurance and carriage of long-range weapons rather than for penetrating sophisticated defenses carrying short-range weapons;
• Large shore-launched unmanned undersea vehicles that take the place of submarines for preprogrammed missions such as covert surveillance or mine-laying;
• Use of a common hull type for all of the large non-combatant ship missions such as command ships, tenders, hospital ships, ground vehicle delivery, and logistics; and
• Elimination of support models that are based on wartime reliance on reach-back access to unclassified cyber networks connected by vulnerable communications satellites or to an indefensible global internet.
The Navy’s long-term force structure requirement is a 306-ship Fleet of the currently-planned designs, of which about 120 (or 40 percent of the force) would be deployed day-to-day. It would also be able to surge an additional 75 ships (another 25 percent) within two months to meet warfighting capacity requirements. In other words, about 65 percent is employed or rapidly employable.
This sounds good, but the reality is that 30 of these 120 deployed ships would be permanently homeported overseas; 26 would be LCSs that use the rotation of their small military crews to keep 50 percent of that class forward deployed; and 40 would be Military Sealift Command support ships that use rotational civilian mariner crewing to keep the ships deployed 75 percent of the time. The remaining 25 of the forward-deployed force will be large and complex multibillion dollar warships with all-military crews, supported out of a rotation base of 140 such ships.
In other words, we plan to buy and operate five of our most expensive ships to keep one deployed. This is not an efficient way to operate. In times of reduced funding our design must address ways to meet our deployment goals with a smaller rotation base while preserving wartime surge capacity.
Many studies and trials have been done over the years on options for reducing the total number of ships needed to sustain the Navy’s robust peacetime forward-deployed posture. Increasing forward homeporting in other nations always comes up as the first choice. While it is a good one, few countries beyond those that currently support this (Japan, Spain, Italy, and Bahrain) are willing to tolerate a permanent new U.S. shore footprint. Building new shore-support infrastructure in foreign countries to back this results in a large bill for construction jobs outside the United States, which Congress normally finds unappetizing.
Using rotational crews to keep ships forward for extended periods without long deployments for their sailors is an efficient option that works for ships with small crews like LCSs, legacy mine-warfare ships, or Military Sealift Command support ships. Experiments in which this has been done with military crews on large complex warships have not turned out well. This was due both to the logistics of moving large crews overseas for turnovers and the difficulty of maintaining exact configuration commonality within ships of a class so that a crew arriving on a ship overseas has trained before deployment on an identical ship (or simulator) at home. Conversions of ships from military manning to Military Sealift Command civilian mariner crews that routinely rotate individual crewmembers to sustain ships forward are limited by the law of war concerning what military actions civilians can perform, and there are few legal options left for further expansion of this approach.
What is left in the force-generation model of our current Fleet is a force of our most complex warships—aircraft carriers, submarines, destroyers, and amphibious ships—operating with permanently-assigned military crews in the “Fleet Readiness Program” cycle of maintain-train-deploy with a deployed output of one in five. Future designs must address this model and find ways to get more deployed time out of these expensive ships and crews—without exceeding the current objective of having military crewmembers spend no more than 50 percent of their time away from homeport over a complete multi-year operating cycle. The current limiting factor is the period required to train the crew as a team before deployment following the inactivity and crew turnover of the shipyard maintenance period.
Naval aviation is steadily moving toward the increased use of high-fidelity single and multi-aircraft simulation as a means of developing and sustaining operational proficiency with reduced use of expensive live flying. These simulators are funded as part of the overall fielding plan for the aircraft and were also built for the ballistic-missile submarine force to support its Blue-Gold crew manning concept. There is no equivalent model or set of off-ship simulators for major sections of the crews of conventional surface warships (other than the LCS) for nuclear-aircraft carriers or for attack submarines. A Fleet design that bought such simulation capability as part of its ship production programs—the way that aircraft programs do—would have significant potential for improving operational output by reducing the time to train for deployment after maintenance periods.
Today’s Fleet design is the product of many separate and disconnected decisions about the required capabilities of 50 different types of ships and aircraft. While not ineffective, it is definitely too expensive. The budget constraints facing the Navy for the next 20 years are not matched by a projected reduction in the quantity or capability of forces that must be delivered forward every day or surged forward in wartime.
The only way to meet these demands within available resources is to develop a design that provides a structure within which the capabilities of future platforms can be shaped to meet the Fleet’s missions efficiently as an overall force. Doing this will require a systems-level approach to defining what it must be able to do, and will mean abandoning some cherished traditions of what each type of platform should do. The alternative is a Navy no longer large or capable enough to do the nation’s business.
2. Mark Arena, Obaid Younossi, Kevin Brancato, Irv Blickstein, Clifford Grammich, “Why Has the Cost of Navy Ships Risen?” (Santa Monica, CA: RAND, 2006). Mark Arena, Obaid Younossi, Kevin Brancato, Irv Blickstein, Clifford Grammich, “Why Has the Cost of Fixed-Wing Aircraft Risen?” (Santa Monica, CA: RAND, 2008).
3. Eric J. Labs, Congressional Budget Office, “An Analysis of the Navy’s Fiscal Year 2014 Shipbuilding Plan,” September 2013.