Secretary of Defense Ashton Carter has served in a variety of positions during on-and-off stints at the Pentagon, dating back to the administration of a president who shares his surname. As such, he has a robust familiarity with the problems that plague the department. And while he is undertaking a number of initiatives in the few months he has left in his position before a new administration takes over, the most consequential might be those dealing with the U.S. military’s personnel policies.

Late last year, Secretary Carter announced a first tranche of proposals as part of his “Force of the Future” initiative. The proposed changes in military personnel policies were a somewhat watered down version of more sweeping proposals that Under Secretary of Defense for Personnel and Readiness Brad Carson had crafted and were on Carter’s desk in summer 2015. However, Carter might have a chance to see one of the proposed changes left out of his November announcement—a revision of the 1980 Defense Officer Personnel Management Act—implemented in this year’s National Defense Authorization Act (NDAA).

The 1980 law mandates a practice commonly known as “up or out.” It placed caps on the number of officers at any given time serving as majors and colonels, or their U.S. Navy equivalent. The caps meant that officers must meet the requirements for promotion in a certain period of time or have to leave the force. Carter wants to add “1-2 years eligibility in a rank” to provide greater flexibility in the career paths officers choose. As the secretary of defense told Stars & Stripes in an interview last month, “in the way [the law] is written, it over-restrictive… Really good people don’t take advantage of opportunities to broaden their skills and make themselves better leaders for the future because they are afraid of letting a deadline pass.”

Secretary Carter is not alone in his call for revising U.S. military personnel policies, but there is also a great deal of resistance. A number of rumored policy changes included in the original Force of the Future blueprint from last summer that did not make the cut when Carter announced his proposals in November 2015. Some of the proposals were simply seen as too costly, while other omissions were—according to some, at least—the product of resistance by the military services. Similarly, the proposal included in this year’s NDAA is facing skepticism in some corners and opposition in others. In the same Stars & Stripes article featuring Carter’s interview, retired officers and officials with the Association of the United States Army and Association of the United States Navy, respectively, worried about the “unintended consequences” of the proposal—such as younger officers becoming complacent if they see their superiors languishing in a rank for an extended period of time.

The NDAA proposal is part of the Senate’s version of the defense bill, which still needs to be reconciled with the House of Representatives’ version. And it is a cautious proposal, allowing a five-year trial period where the caps would be lifted—with the Pentagon required to report back on the impact of the changes.

As discussed here previously in regard to the initial Force of the Future announcement, Carter’s proposed changes have their pluses and minuses. The question is why they should be implemented uniformly across the military services. The same could be asked of the revisions to up or out requirements.

The military services have different personnel needs due to their different missions, functional specializations, and organizational cultures. Like most bureaucracies, they cherish their autonomy and like to protect their turf. While still requiring legislative changes, encouraging personnel changes in one or two services, rather than all, might lead to less institutional resistance to change. The Air Force and Navy were apparently less resistant to the initial Force of the Future proposals because some of the proposals fit with their (relatively) more technologically-focused labor needs than the ground forces.

More importantly, implementing the policy in one or two services rather than across the board would also provide a basis for comparison in judging the impact of the changes. Services that find up or out conducive to their needs might choose to maintain the practice. If they see the other services achieving greater success with more flexible promotion policies, they might choose to mimic them or even to craft innovative policies in response. Across-the-board changes often leave the military with less flexibility, even when the proposed change is supposed to encourage it.