Monday, February 21, 2011

Why can't we do math AND strategy?

By now everyone's familiar with the dismissive turn of phrase Bob Gates used to blow off the recommendations of the Simpson-Bowles deficit-reduction commission: "In terms of the specifics they came up with, that's essentially math, not strategy." The defense-defenders love this one, and they've deployed it aggressively: basically every recommended cut to a defense program or the defense topline is derided as "math, not strategy."

Hey, it's a good line. It's probably the same thing I'd say if I were the SECDEF. But it's wildly irresponsible for the Pentagon to say "hey, it's not us, we're doing our part!" as Gates did back in November:
"If you cut the defense budget by 10 percent, which would be catastrophic in terms of force structure, that's $55 billion on a $1.4 trillion deficit," Gates said. "We are not the problem."
(By that logic, shouldn't GE be saying "if you cut the F136 engine, which would be catastrophic in terms of Ohio jobs and the assured reliability of the Joint Strike Fighter, that's $2.9 billion (pdf) on a $553 billion defense budget. We are not the problem"?)

Secretary Gates wants to have his cake and eat it, too: he wants to dismiss critics of his cuts by saying the recommendations are based on strategy, but push back on those calling for deeper cuts by talking about math. But a responsible defense budget and a responsible federal budget must take both considerations into account: math and strategy. If ADM Mullen is right and the national debt really does pose the greatest threat to national security, then isn't the "math" part really, really significant? Strategy, after all, is inseparable from math: strategy is fundamentally about choices, about maximizing benefit with limited resources. The SECDEF said it himself just two years ago!
The United States cannot expect to eliminate national security risks through higher defense budgets, to do everything and buy everything. The Department of Defense must set priorities and consider inescapable tradeoffs and opportunity costs.
Gates's defenders -- and I'm one of them -- will say "but the Department already has set priorities, and this budget request is based on the real requirements of supporting our national security strategy!" And when it comes down to it, I probably agree with him on this one. Beyond this justification, Gates has to defend the requested figure out of simple adherence to his word: he told the service secretaries and chiefs that they'd get their savings back if they found fat to cut; if those dollars instead come out of the topline, just wait and see the foot-dragging when the next round of service-identified "efficiency" cuts comes around.

But just as expected, people aren't happy: conservative commentators, China hawks, the defense industry, and most of their buddies on the Armed Services committees think that $553 billion is just an irresponsibly paltry sum, while peaceniks, budget hawks, and proponents of other vulnerable spending programs are tearing their hair out at the seeming invulnerability of the defense budget. Add to that latter group the New York Times editorial board, which yesterday did its level best to oversimplify the issue and generally aid confusion and misunderstanding. On the subject of the EFV's termination, they wistfully bemoan the $3 billion already wasted -- as if anything the SECDEF might do could get it back. On the Joint Strike Fighter, they note a doubling of per-unit cost while arguing vaguely that "it should be cut back" -- apparently failing to absorb the irony of recommending action that will further increase cost-per-plane. And on the Osprey and Virginia-class sub, they argue for program cuts or cancellations based on oversimplified and outdated operational rationales.

And that's not even the worst part. Try this one on for size:
The Pentagon needs to jettison the ancient formula that guarantees each service its accustomed share of taxpayer dollars. Air and sea power that can be readily mobilized are vital to American security. But for a decade, the Army and the Marines have been pushed to their limits while the Navy and the Air Force have looked for ways to stay useful and justify their budget shares. 
Updating the formula to reflect a more realistic division of labor would wring significant savings from the Air Force and the Navy while protecting the Army and the Marines from the multiple combat tours that have strained service members and their families over the last decade.
I hear a lot about this "ancient formula," but I've never had it adequately explained to me. That said, let's stipulate that there is such a formula for ensuring each service gets its annual share of the budgetary pie; might it not be the case that this makes good sense? After all, you can't turn the entire Defense Department around on a dime, as Secretary Gates has learned, nor should it be easy to do so. This isn't just a matter of institutional interests and inertia  (though those are at play, too), but of preventing the crises of today from fundamentally destabilizing the kind of steady-state modernization that underpins the U.S. military's global freedom of action and deterrent functions.

But back to the editorial -- let's look at exactly what the two paragraphs above really say: "Air and sea power that can be readily mobilized are vital to American security... [but] a more realistic division of labor would... protect the Army and Marines from... multiple combat tours." Uh, what? This is really almost impenetrable. Presumably the writers mean to say that budget shares should more realistically reflect the actual intra-Departmental division of labor over the last several years (not "a more realistic division of labor"), and that cuts to the Navy and Air Force's budgets would allow for an increase in land forces end-strength and result in extended home-station dwell-time for Army and Marine units. But there are a number of very significant problems with this line of reasoning.

  1. The suggestion that the Navy and Air Force have had to "look for ways to stay relevant and justify their budget shares" is based on two serious misunderstandings: the first is that air and sea forces' true value should be found in tactical utility in combat rather than strategic deterrent and guarantor of access, and the second is that "looking for ways to stay relevant" is a bad thing rather than an important driver of institutional adaptation. 
  2. Calls to shift funding from expensive, sophisticated, long lead-time weapon systems to personnel accounts and other investments in land power force structure fail to acknowledge that high-end technical solutions cannot simply be purchased when needed, but are the end-result of sustained investment in the defense industrial base and commitment to developmental technologies.
  3. Further end-strength increases will put additional strain on an already troubled recruiting base and increasingly challenge the Army and Marine Corps to build soldiers and Marines from suboptimal raw material. On top of that, absent significant reform to Tri-Care and military benefits, expansion of the active-duty rolls will carry with it the very same long-term fiscal commitments that are helping to cripple serious efforts at deficit reduction in the domestic portion of the federal budget.
  4. If you listen to senior leaders in both the Defense Department and the broader United States government, manpower-intensive land wars aren't on anybody's top-five list of efficacious foreign policy solutions. U.S. troop levels in Afghanistan are about to start coming down, so what sense does it make to start budgeting for a larger Army and Marine Corps?

This editorial exhibits the very worst tendencies of both budget-cutters and budget-growers: bath math, bad reasoning, and an absence of strategic judgment.

Things aren't as simple as any of the factions in the budget debate want to make them. Beware of those who bring math or strategy, but not both.

6 comments:

  1. "and this budget request is based on the real requirements of supporting our national security strategy!"

    The latter is the problem. It's unaffordable and wasteful (as is much else).

    ReplyDelete
  2. With regard to the 'ancient formula' dictating resource allocation between the services, I don't quite understand your defence of it.

    It may very well be, as you suggest, that the breakdown in funding between the services should not be overhauled just because the ground forces have been under the most strain in recent years. But wouldn't you agree that some some level of adaptation to current realities might be needed? In fact, depending on who's numbers you trust, Army funding actually decreased as a proportion of total funding during the years of its greatest stress.

    And if your main concern is preparing for future contingencies, what makes you so sure that the current allocation between services, pretty much unchanged since the 1990s, is the one to keep hold of, the one most responsive to what lays ahead?

    Critics of the defence budget and defence policy often mount strawmen and beat them with a stick, but I think the inflexibility of resource allocation between the services is a legitimate target.

    ReplyDelete
  3. Gulliver:

    First, I'm dense. Second, you're an SME, and your writing shows this. All that respectively said, I found it hard to glean an overarching thesis to this post.

    To me, the key takeaways were as follows:

    1) There's the risk of imperial overstretch (see Stephen Walt today). The budget matters in national security and is part of national security; it should be taken into account in high-level national security formulation and planning.

    2) The issue of continual consistent service shares of the budget is irksome. First, it ignores that the Navy and Air Force play a role in national security, even though if they're not particularly actively engaged in on-going fighting.

    3) Media coverage, and actual actions, display ignorance of how the defense industrial base function. For example, desires to cut industrial spending on weapons programs based on high per-unit cost of production ignores that cutting those programs increases per-unit cost of production.

    I don't have any particular insight to add this post, other than to note that budget incrementalism and stasis has been noted since, at least, the days of Herbert Simon; that program termination is inherently difficult; and that also, in a way, is it wrong to cut back on a wrong-headed program even if it does increase the per-unit cost of production, if said program is indeed wrong-headed?

    I find little objectionable in those points - with the exception, possibly, of the discussion of per-unit cost of production. But I also know, or think, there are thoughts, ideas and concepts I'm missing. Are we simply in near-conformity of opinion, at least as I've admittedly (over?)simplified the issues, or am I off-base far more than I perhaps realize?

    Thanks
    ADTS

    ReplyDelete
  4. David -- With regard to the 'ancient formula' dictating resource allocation between the services, I don't quite understand your defence of it.

    I shouldn't call it a "defense," exactly, but rather an appeal to try to understand the practical justifications for why such a thing might exist.

    But wouldn't you agree that some some level of adaptation to current realities might be needed? In fact, depending on who's numbers you trust, Army funding actually decreased as a proportion of total funding during the years of its greatest stress.

    And if your main concern is preparing for future contingencies, what makes you so sure that the current allocation between services, pretty much unchanged since the 1990s, is the one to keep hold of, the one most responsive to what lays ahead?


    Sure, adaptation to current realities should be on the table -- particularly "adaptation" in support of ongoing operations. But frankly we're talking for the most part about nibbling around the edges, at least when it comes to procurement and modernization. And the "emergency" war supplementals (or now, the separate OCO account) have been used as a means to adjust the balance of funding -- if only slightly -- in a more equitable and rational way.

    Now if you're talking about broad-based reallocation among the services, I don't have a huge problem with the idea of doing that as part of future-year projections. But again, this has to be about math and strategy: a counterintuitive appeal to surge spending on land force end-strength that would come far too late to meaningfully impact operational realities in a war that's being brought to a close doesn't seem to take either of those factors into consideration.

    If you want to say "let's give 75% of future defense dollars to land forces because we'll need large land forces to execute X strategy or policy, and because unnecessary big-war programs A, B, and C (things that are only being funded right now in order to justify service shares and budget toplines) can be cut, and because all the modernization and procurement programs we think are absolutely REQUIRED to prepare for future contingencies ca be funded within the Navy's and Air Force's new 25% total allotment, or are dramatically less important than the necessary spending on land force end-strength," then that's different, and that's great. But the budget isn't just about now, and it's not even just about next year. Math AND strategy.

    ReplyDelete
  5. ADTS -- I found it hard to glean an overarching thesis to this post.

    So did I. I'm not totally sure I had one.

    To me, the key takeaways were as follows:

    1) There's the risk of imperial overstretch (see Stephen Walt today). The budget matters in national security and is part of national security; it should be taken into account in high-level national security formulation and planning.


    Yes.

    2) The issue of continual consistent service shares of the budget is irksome. First, it ignores that the Navy and Air Force play a role in national security, even though if they're not particularly actively engaged in on-going fighting.

    I would say "the reflexive assumption that the constancy of these shares is a simple result of bureaucratic inertia or quasi-corruption and has no practical justification" is irksome, to be more precise. But yeah (-ish).

    3) Media coverage, and actual actions, display ignorance of how the defense industrial base function. For example, desires to cut industrial spending on weapons programs based on high per-unit cost of production ignores that cutting those programs increases per-unit cost of production.

    True, though trust me on this: I am not interested in mindlessly spitting the Aerospace Industries Association's line at you, and that should be obvious by now from my other writing.

    is it wrong to cut back on a wrong-headed program even if it does increase the per-unit cost of production, if said program is indeed wrong-headed?

    No, certainly not. I'm not trying to make a sunk-costs argument, though I suppose I understand how it sounds that way. On the contrary, we've determined that the Joint Strike Fighter and Osprey programs, for example, are NOT wrongheaded. JSF's presumed utility is one of the major justifications for the cost savings this SECDEF has achieved on F-22, as I'm sure you know, and Osprey has been integrated into the Marine Corps' current air operations and future concepts in some kind of at least remotely useful way. Yes, absolutely, shitcan a program if it doesn't work or if you don't need it, no matter how much has already been spent. See EFV and F136 for examples of me making precisely this argument.

    I find little objectionable in those points - with the exception, possibly, of the discussion of per-unit cost of production.

    All I'm saying here is that it's not intellectually coherent to say "per-airframe costs have shot up on JSF, and that makes it a sucky program that we should cut back"; the irony here is that you may reduce the outlay on that airframe, but you will be buying less capability by multiples -- it's not just an arithmetic reduction. And again, the original number ought to have been based both on math and strategy: i.e. this is how many airplanes we need, and this is how much we expect we can spend on them. Cut that latter number and you either need to explain why the first figure doesn't hold anymore from a capability perspective -- why (as in the case of F-22) the first number you came up with was bullshit (e.g. to fill out a topline request in a time of plenty, give yourself slack when inevitable cuts happen, etc., both of which are pretty much unethical and won't be openly offered as explanation) -- or why the dollar tradeoff makes strategic sense. "We're getting less than we planned for the money we budgeted" isn't a complete argument by itself.

    ReplyDelete
  6. David -- Because it just occurred to me, I'll give an example of the sort of thing that happens when you make dramatic adjustments to procurement plans based on translating current operational requirements into ill-founded presumptions about the character of future conflicts: you end up giving away (yes, literally giving away) a whole bunch of expensive MRAPs and up-armored HMMWVs because 1) they'll be incredibly expensive to return to home station, and 2) you really can't figure out what the hell you're going to use all of them for in the future.

    Am I saying we should refuse to adequately equip our forces for the current fight out of concern that we might be making the wrong long-term decision? Nah, I'm just pointing out why it's important to be right (or least wrong) when you're talking about shifting big dollars around. It matters less (to DoD if not the country) in this instance because this stuff was funded with war supplementals/OCO and didn't require reprogramming of money from the base budget's modernization accounts, but it's a useful lesson.

    ReplyDelete