WASHINGTON — In 2017, the top two officials at the Pentagon — then-Defense Secretary Jim Mattis and then-Chairman of the Joint Chiefs of Staff Gen. Joe Dunford — testified to Congress that the defense budget needs to have 3-5 percent annual growth over inflation each year through 2023 to ensure America’s military success.

Dunford, speaking to the Senate Armed Services Committee in June 2017, went as far as to say: “We know now that continued growth in the base budget of at least 3 percent above inflation is the floor necessary to preserve just the competitive advantage we have today, and we can’t assume our adversaries will remain still."

Three years later, as the Trump administration prepares to unveil its fiscal 2021 budget request on Feb. 10, such growth appears impossible. The budget is expected to be largely flat, as a two-year budget deal reached last summer calls for $740 billion in defense spending in the next fiscal year, up just $2 billion from the enacted FY20 amount.

“The 3-5 percent goal was reasonable enough and absolutely needed,” said Mackenzie Eaglen, a budget analyst with the American Enterprise Institute. “But it is not happening. The defense top line for 2021 is negative real growth, aka declining.”

Susanna Blume, a defense analyst with the Center for a New American Security, said that certain parts of the defense budget, particularly maintenance and personnel costs, grow faster than the rate of inflation. “That’s what’s behind these comments about requiring a certain amount of real budget growth in order to sustain the joint force as it is today,” she said.

But there is a wild card, according to Ellen Lord, the Defense Department’s top acquisition official: a series of reform efforts led by now-Defense Secretary Mark Esper, which so far have accounted for $5 billion in savings.

“We’re getting more and more efficient. That is obviously what Secretary Esper is focused on with his defensewide review, that we are cutting out administrative tasks and a variety of portions of programs to make sure we return those savings to our critical modernization efforts such as [artificial intelligence], hypersonics and so forth,” Lord said during a Jan. 31 news conference at the Pentagon. “We are always having to look very carefully at our budgets and make sure we triage them to focus on the critical few. So we’re always concerned, but we’re always going to work it.”

How much of that expected growth gap can be filled by Esper’s efficiency drive is difficult to pin down. Blume said its “certainly possible that efficiencies could make up some of that gap,” but whether the work that has been done now and is planned in the near term will be enough “are questions we don’t have answers to today.”

Added Eaglen: “Efficiencies alone will not get the Pentagon its 3-5 percent growth in actual dollars to reinvest. The defensewide review only yielded $5 billion, and the way it works with these drills is that the money doesn’t necessarily move from pot A to pot B as a result."

“But that doesn’t mean it is not worth doing. Any money amount is helpful. And the exercise is also about getting the bureaucracy to shift its time, tasks and attention to great power competition as much as it’s about shifting funds into higher priorities that support the strategy,” Eaglen said.

If one of the Pentagon’s big bets work out, that could be a real game-changer, Blume said. Those bets include efforts to replace a Defense Logistics Agency warehouse using a 3D printer as well as attempts by the Air Force to rapidly develop, prototype and produce fleets of planes. If one of them goes well, Blume said, “you can potentially start to bend some of those cost curves.”