At this point, I think it’s pretty well understood that the budget debate largely comes down to my caps vs. your caps. There’s a procedural reason for this: Caps and tax increases are really the only policies the Congressional Budget Office will credit with saving much money over a period of decades, and since no one wants to do much in taxes, everyone is relying on caps. Caps also allow you to be vague about the what, where and how of your cuts, which is a plus for politicians.

The result is that no matter where you look, it’s caps, caps and more caps. Rep. Paul Ryan’s budget caps Medicare and Medicaid spending at the rate of inflation and non-security discretionary spending at pre-2008 levels for five years. The Simpson-Bowles plan caps all federal health-care spending at GDP+1% and cuts and caps both defense and non-defense discretionary spending. The Obama budget caps Medicare spending at GDP+0.5 percentage points per year and adopts the Simpson-Bowles plan’s non-defense cuts and caps.

But bad caps can delay hard choices about what to cut and where to tax and how to reform, and turn deficit reduction into a meaningless exercise in spreadsheet manipulation. So when looking at a cap, you have to ask yourself two questions: First, is it achievable? I could say we’re going to cap defense spending at half its current level and hold it there for the next 10 years, and if the CBO scored my proposal, you’d see it saves a lot of money. But it’ll never happen.

Second, does the legislation include policy or procedures that could achieve the cap? It would be possible to hold health-care spending to something like GDP+1%, but it’s going to require a lot of reforms, some of them quite politically difficult, all of them sustained over a long period of time, to get there. If my proposal doesn’t include policy to get there or procedural reforms that are going to force future Congresses — or someone else — to pass policy to get us there, then it’s just a goal, not a cap. For a cap to be plausible, you need to pass either the policy that makes it plausible now or procedural changes that will force future Congresses to do it, or allow it to be done, later.

One of the reasons I’ve been pretty down on the GOP budget is that it fails both tests. Capping Medicare and Medicaid at the rate of inflation isn’t possible under any combination of policy or process reforms, at least if you still intend to provide health care to people. Something like GDP+1% — which is what Ryan used in the original incarnation of his Medicare proposal, what the Simpson-Bowles commission settled on and what the Affordable Care Act shoots for — is difficult, but at least it’s conceivable.

But even if Ryan moved back to GDP+1%, he’s still lacking any real process or policy mechanisms to get there. He hands Medicare over to private insurers and adds exchanges, but the Congressional Budget Office thinks that’ll make coverage more, not less, expensive. He doesn’t have anything in terms of the delivery system — paying doctors for quality rather than volume, etc. — and nothing in his plan binds future Congresses, or even helps them act, in any way. In fact, he repeals the Independent Payment Advisory Board, which the Affordable Care Act empowered to make difficult reforms without Congress’s explicit consent.

Comparatively, the Simpson-Bowles plan and the Obama budget build on the Affordable Care Act, which uses a mixture of policy changes — particularly, reforming the way Medicare pays for care, separating evidence-based care from care that night be wasteful and developing new ways for Medicare to continuously test and scale cost-saving reforms — and procedural reforms — notably, strengthening the Independent Payment Advisory Board — to achieve their caps. The Obama administration also includes a deficit trigger that makes automatic cuts to both spending and tax expenditures if future Congresses don’t do their part to reduce the deficit. This is a strategy that worked quite well to force future Congresses to make changes in the early 1990s.

So I don’t see this as much of a contest. But whether you buy my analysis of the different caps or not, this is the discussion we need to have. In the end, caps don’t save money. It’s the policies enacted alongside the caps, or the decisions future Congresses make to abide by the caps, that save money.