On October 6, Secretary of Defense Mark Esper laid out his vision of what the U.S. Navy’s future fleet should look like. Called Battleforce 2045, it proposed big increases in the number of attack submarines, surface combatants and amphibious warships, combined with an embrace of unmanned vessels aimed at getting the size of the fleet above 500 ships by mid-century.

However, Esper singled out one category of warships for potential cuts: large-deck, nuclear-powered aircraft carriers. After acknowledging that “nuclear powered carriers will remain our most visible deterrent,” the defense secretary went on to say that “we continue to examine options for light carriers that support short takeoff or vertical landing aircraft.”

One option Esper cited was a modified version of the USS America amphibious assault ship, which he noted will carry more than a dozen F-35B jumpjets. He suggested that by purchasing up to six light carriers similar in size to USS America, the Navy could reduce the number of large-deck, nuclear-powered carriers to between eight and 11, reserving the larger carriers for the “high-end” fight while the light carriers took on more prosaic presence missions.

This is a bad idea.

One reason is that if the number of large-deck carriers fell to the lower part of Esper’s range, say eight or nine, the Navy could only keep two supercarriers forward deployed on a typical day. Europe, the Persian Gulf, or the Western Pacific would have to make do with smaller carriers for long stretches.

The number of large-deck carriers the Navy actually operates today, 11, is about as low as it can go while assuring continuous presence in all three regions, given training, transit and maintenance requirements for the Navy’s biggest warships.

But would that really be so bad if there were six light carriers to cover any gaps created by reducing the number of nuclear carriers by two or three? Yes it would, because light carriers can’t actually fill the gap. Light carriers could show up, but they wouldn’t be capable of fighting and winning a conflict of any significant scale.

The Navy has accumulated extensive evidence over the years attesting to the drawbacks of light carriers. Secretary Esper says he is trying to fashion a future fleet that has greater lethality, survivability, adaptability, ability to project power and control sea lanes, and capacity to deliver precision effects over long distances.

The light carrier concept fails on all five of these measures. Let’s consider a few facts. (I should note that Huntington Ingalls Industries HII +2.1%, the company that builds the Navy’s large-deck carriers, contributes to my think tank; if Esper’s vision came to fruition, it would likely build the light carriers instead, plus a host of other vessels in the plan including subs and surface combatants.)

Light carriers are less agile. Large-deck carriers are nuclear-powered, whereas light carriers patterned on USS America would be powered by fossil fuels. This has big operational consequences. Large-deck carriers of the Nimitz and Ford classes have a maximum speed of 35 miles per hour, which is 40% faster than the 25 mph max usually cited for America. Thus the large-deck can cover a lot more ocean, moving up to 700 miles in a day. It also has unlimited endurance at sea, whereas a light carrier based on USS America would necessarily be tethered to its supply chain of fuel-carrying tankers.

Light carriers are less lethal. Large-deck, nuclear-powered aircraft carriers host an air wing that includes multiple squadrons of F/A-18 or F-35 strike fighters, a squadron of electronic attack aircraft, a squadron of radar planes, and various rotorcraft. The total number of aircraft typically exceeds 70. So when Secretary Esper says a light carrier might host “more than a dozen” F-35s, that is a small fraction of the striking power afforded by a large-deck carrier. There would be much less usable deck space on the light carrier, and much less room for storing munitions. While an air wing hosted on the Ford class of large-deck carriers could accomplish 270 sorties per day in a heightened state of alert, a light carrier would be hard-pressed to sustain one-fifth of that number.

Light carriers are less versatile. Because it hosts a big and diverse air wing, a large-deck carrier can accomplish multiple missions simultaneously—power projection, sea control, fleet air defense, jamming of enemy communications, etc. A light carrier would lack sufficient aircraft and deck space to perform multiple missions, unless each mission was conducted sequentially rather than simultaneously. Its MV-22 rotorcraft might be modified to provide jamming or aerial surveillance, but the vessel itself would be too crowded to adapt to new challenges that might arise.

Light carriers are less survivable. One reason for the light carrier’s lack of growth potential is that it provides only a fraction of the electrical generating capacity available on large-deck, nuclear-powered carriers. This impacts survivability, because there isn’t enough power for high-resolution radars or weapons like lasers. Moreover, the smaller size of the vessel would make it more susceptible to being disabled by enemy attack. There would be fewer defensive systems, fewer watertight compartments, and fewer opportunities to work around damage to the flight deck. Its air wing would likely be too small to sustain combat air patrols of surrounding air space while also executing other missions.

Light carriers are less sustainable. A warship that depends on fossil fuels for its propulsion must have frequent access to replenishment tankers. Unlike nuclear-powered carriers that require refueling only once in a multi-decade service life, light carriers powered by gas turbine engines could seldom stray far from their source of fuel. Whether undefended tankers could safely rendezvous with light carriers in hostile waters is an open question, as is the matter of how much all that fossil fuel would cost during the light carrier’s lifetime. Nuclear is more flexible and efficient, and—depending on oil prices—potentially cheaper.

Of course, there is more to carrier sustainment than just keeping the ship moving. A deploying Nimitz-class carrier typically carries enough food and supplies to sustain personnel onboard for 90 days. The ship generates all the fresh water it needs by distilling sea water. The long pole in the logistics process is securing sufficient aviation fuel to keep the air wing active, which is why the newest Nimitz-class carriers can store nearly 6,000 tons of fuel. That might not last long in an intense air operation, but the high speed and unlimited range of a nuclear-powered carrier would ease the challenge of replenishing fuel in a war zone, compared with depending on a light carrier.

Secretary Esper’s desire for a more lethal, survivable and adaptable fleet is understandable, especially if U.S. defense strategy remains focused on China’s modernizing military. However, unlike his proposal to increase the pace of submarine production, it isn’t clear how operating fewer large-deck, nuclear-powered carriers and buying light carriers instead would solve any of the problems faced by U.S. military planners.