Last week, McKenzie Eaglen and Rick Berger wrote an article in War on the Rocks about how the Congress and the Office of the Secretary of Defense (OSD) were making gaps in carrier coverage worse. They rightfully highlighted the shortage of carriers. The president, secretary of defense, and combatant commanders all want more carrier coverage than the U.S. Navy’s carrier fleet can provide. However, Eaglen and Berger let the Navy off the hook too easily. The Navy itself bears much of the responsibility for the carrier gap.

Back in the late 1990s, the Navy decided to cease production of the Nimitz-class carriers and move to a new design, now called the Ford class. The new class would have several advantages, especially reduced manning from increased automation and a 33 percent higher sortie rate. The increased sortie rate was driven by perceived needs of a conflict with China where anti-carrier weapons would make it important to get a lot of aircraft into the air quickly. Moving from the Nimitz-class to the Ford-class cost about $11.0 billion (in FY 2017 dollars, after the inevitable cost growth). That included $4.1 billion in research and testing, and $6.9 billion in higher lead-ship costs ($14.7 billion for CVN-78Ford v. $7.8 billion for an upgraded Nimitz follow-on, all converted to FY 2017 dollars using DOD’s “Green Book” deflators). In effect, the Navy spent the equivalent of one and a half carriers on moving to the new class. That loss can never be regained.

Further, the Ford is two years behind schedule. That delay means a longer gap after CVN-65 Enterprise retired in 2012.

Did the Navy make the wrong decision in moving from the Nimitz-class to the Ford-class? It is easy to second guess decisionmakers who were making judgments about programs in an environment of uncertainty about costs and future threats. Nevertheless, with the benefit of hindsight, we can make some useful judgments: The Navy traded off its most in-demand capability for uncertain capabilities in a future conflict. Many of the acquisition problems could have been anticipated and any savings were not enough to offset the high up-front costs. So, yes, with due respect to those who had to make decisions with less information, the Navy probably made the wrong decision.

The classic question in a crisis is, where are the carriers? Since World War II carriers have made their mark through power projection and crisis response. That has to be the highest priority and that requires numbers.

Carriers have limitations, perhaps severe, in the high-end fight. There’s a lot of debate, pro and con, about carriers in anti-access/area-denial (A2/AD) environments, and that won’t be solved here. Moving to the Ford class did not end the debate.  As Jerry Hendrix has pointed out, the short range of the current and projected carrier air wing requires them to get very close to an enemy. What’s the value of a high sortie rate if you can’t get within range?

GAO argues that the Navy should have anticipated that its acquisition strategy for the Ford was highly risky. Too many systems — like the launch system and the reactor — were immature. Manpower and other savings will be real — the Navy estimates $4 billion over the life of the ship compared to a Nimitz-class — but that’s over 50 years. The present value of those future savings  is small (about $1.6 billion in discounted dollars using OMB’s prescribed rates.) Further, as the Navy has learned with the Littoral Combat Ship, manpower savings don’t always fully materialize.

The Navy continues to tradeoff carrier numbers for new capability. In 2013, it proposed retiring CVN-73 George Washington early, and not putting the ship through its midlife extension, in order to save money to build new carriers. The result would have been even fewer carriers. However, congressional pressure forced the Navy to fund the extension and keep Washington in service for another 25 years.

The Navy’s proposal to retire the Washington early would also have been the most expensive course of action for fielding carriers: Building the expensive new carriers and retiring the existing carriers early ($330 million per year for a new build vs. $180 million per year for extending the existing). As the Navy struggles to balance its many budget requirements, one can expect the Navy to make the same proposal in the future with the next carrier overhaul, for CVN-74 John C. Stennis.

Some might be tempted to argue that budget constraints caused the carrier shortage. However, the Department of the Navy’s procurement budget rose from $35.5 billion in FY 2001 to $49.1 in FY 2016, with a peak of $54 billion in FY 2008 (all in FY 2017 dollars, Green Book, Table 6-17). In other words, procurement funding has not been a constraint.

Eaglen and Berger criticize OSD for directing shock testing of the Ford before it deploys. (Shock testing assesses the ship’s ability to withstand nearby explosions. The risk is not that the ship will sink but that key machinery and operating systems will fail.) The test would delay Ford’s first deployment by up to two years and therefore extend the carrier gap. They note that shock testing could be conducted later on either by the Ford or subsequent carriers in theFord class (CVN 79 John F. Kennedy is the next carrier being built of the new class, and CVN-80 Enterprise after that. Carriers are currently built at five year intervals.). This gets at the key question about carriers: Are they primarily for peacetime forward deployments and power projection in regional conflicts or for conflicts against high end adversaries like China and Russia? If the former, then shock testing could be delayed because no forward deployment missions or regional conflicts can seriously threaten the carriers. If the latter, then shock testing is critical in reducing carrier vulnerabilities to enemies who have the ability to attack them. And the Ford-class was justified with combat against China in mind.

There is also the question of fixing any problems that shock testing identifies.  If several carriers have already been built, then fixes must be retrofit, which gets expensive. Conversely, incorporating improvements before the ship is built is relatively easy.

As indicated in my discussion of the Nimitz vs. Ford decision, I tend to side with Eaglen and Berger even if I don’t agree with them on where the blame lies. The U.S. Navy should be allowed to get the carrier out as soon as possible and accept the cost and warfighting risk of delaying shock testing. Crisis response and power projection are everyday events while peer conflicts are unlikely in the near-term.  But it is not an easy decision since both are important. Further, GAO has long flailed the Pentagon for not following “sound acquisition practices” on the Ford program, so any delay in testing will produce more highly visible criticism.

The causes of the carrier gap are many: a world that becomes increasingly threatening, a Congress that cannot align budgets and strategy, an administration that focuses on domestic programs, but also a Navy that made some expensive budget decisions in the past and now must live with the consequences.