Defense Part I: Spending

<STD: Serious Topic Disclaimer>

In the interest of offering content both didactic and accessible, the following Serious Topic will not include proper source support. This means you shouldn’t believe any of it. If you find the discussion compelling, however, I encourage you to pursue a more demonstrable truth.

Especially following the State of the Union address, I’ve been thinking a lot about our defense spending situation. Of course, the whole point of this blog is to argue through things with some degree of measure and reason, and the State of the Union doesn’t do that at all, regardless of the president or his political leanings. The SOTU is designed to inspire, reassure, and galvanize. Nobody should treat anything President Obama said during the address as an argument for any policy point, whether for the sake of supporting or for impeaching that position. It’s just propaganda. I don’t mean this to sound pessimistic—it’s just the nature of the beast. Nobody, in an hour and a half, can make a statement sufficiently comprehensive to be considered a thorough argument of all the issues confronting our nation.

But that doesn’t mean important things weren’t discussed, and our defense spending really is one of the most controversial debates raging at this moment. Balancing our collective budget is politically a hot button issue, and most agree that some amount of revenue increase partnered with some amount of spending cuts will be necessary to accomplish that goal. The trick is finding out where defense spending fits into that equation, and I think the issue presents some unique problems.

 I’ll preface my discussion by saying that I recently had the occasion to speak with someone who is positioned better than perhaps anyone on the planet (except perhaps the current Secretary of Defense and President Obama) to offer insight on this issue. Our discussion informs the following. Actually, things I’ve learned over the past few weeks have changed the way I think about a lot of this stuff.

For a bit of overview: Currently defense spending sits at around 24% of the U.S. budget, but budget cuts undertaken over the past few years under Secretary Gates (two separate cuts of $400 billion each) leave us with a significant anticipated decrease. Washington is working on the much-maligned sequestrations right now, which would cut an additional $600 billion. Considering these cuts and rising costs in other sectors—most notably health care, where costs are projected to skyrocket and which will likely (in my opinion) even increase at a rate exceeding expectations—our defense spending is no longer the monster spending sector archetype a lot of us imagine it to be. Think 15% or so of the budget for the coming years (though, depending on how other sectors develop and how much the sequestrations actually cut defense spending, this number could fluctuate).

So how much do we need to spend on defense? Or, a better question: How should we approach this problem? As with virtually every question, there’s a deceptively simple equation on which we should model our thinking. What are the risks and what are the costs? In other words, if the cost of increasing defense spending is less than the cost we expect to incur if we don’t, then we should increase spending. Conversely, if the costs of decreasing spending are less than the money we cut, then slash away. This is at least the basic model from which we should begin an analysis, but things get a lot more complicated, as we’ll soon see.

The first problem we encounter relates to information availability. U.S. citizens know how much money goes into the defense budget—that information is public domain. Of course, we don’t know what the money buys in a lot of cases (there’s no line item for “secret government base,” of course), but that’s not so important. The point is that one half of our equation—the cost of defense spending—is very easy to measure.

Not so easy to measure are the expected risks associated with changes in defense spending. Imagine a hypothetical country approaching this issue. It could opt to be very defense-oriented, surrounding itself with walls, missile defense systems, mounted guns, tanks, and all the military infrastructure one could imagine. It could invest in nuclear weapons and compile a huge arsenal. It could make military service mandatory. All of these things would certainly make this country very secure (OK, these things might also be the sign of a crazed military dictator and create a very unstable relationship with terrified neighbors, but let’s keep things simple for the sake of this analysis), but would also be very expensive. Another way of looking at this is that the hypothetical country could devote 100% of spending to defense, thus effectively minimizing the risk of loss associated with inadequate defenses. On the other side of the coin, a country could spend 0 on defense, maximizing potential loss, but also bearing no defense costs. If nobody ever attacks, they’re great beneficiaries of their policy; however, if someone does, they’re out of luck (to put things mildly).

That’s just the problem, though: How do we know if someone will attack, or whether our country will be involved when other countries fight? Information gathering is a big part of defense spending for just this reason. But there’s a lot of uncertainty here, and that makes the second half of the cost/risk analysis a lot harder to handle. If our hypothetical country is a place like the United States, most reasonable people don’t feel like their lives are in danger. But if it’s a place like, say, Libya, the opposite might be true. Are these intuitions a good basis on which to judge our spending habits? Probably not, but we don’t have the information to say they’re necessarily wrong.

The alternative is to look at our military intelligence and rely on those risk projections. There a few problems with this approach, as well. They are, briefly: Transparency, agency costs, and risk aversion.

1. Transparency: This one is obvious, and it’s particularly tough because it’s such a necessary problem. The military has to operate with some degree of operational confidentiality, or security could be compromised. Just like police investigators will withhold information from the public to keep the bad guys in the dark and to gauge the value of new data, so the military keeps certain things under wraps. We all know it’s true, and we should all be OK with it to a point—if the military disclosed everything, there might be serious negative externalities, including a serious reduction in the efficiency of our defense spending. I’m a big fan of small government, but some things just make sense, and our system was wisely designed to keep civilians in control of the military (through the president) precisely so that the military could do its classified stuff without presenting a threat to the populace.

Regardless, this makes it really tough for average citizens to gauge what risks we’re facing. People in the Pentagon probably know about attacks on this country that have been foiled silently. Pundits on television probably don’t have that information—nor do a lot of people in our nation’s policymaking circles. But those risks do factor into the model equation.

2. Agency costs: Just like in corporate law (or any scenario where anyone is hired to do something, really), our top military personnel produce agency costs—and I’m not just talking about their salaries. There’s the obvious conflict of interest—people in the military (and especially, perhaps, at the top of it) are probably more likely to support military action, and thus higher spending. These people have made a living out of the business of defense, so it’s natural for them to be devoted to accomplishing those goals. Most people overrate the importance of what they do, and the military is no different. Of course, countervailing this effect are the procedural realities checking military power in the United States (e.g., Congress sets the military budget; the President is Commander-in-Chief).

The fact remains, though, that the people in possession of the best information most relevant to defense spending decisions are also the people who want that spending the most. They also don’t reveal all of that information to us. And the market forces pushing these guys do make efficient decisions are weak: How many people are qualified to be CIA directors, four-star generals, and secretaries of defense? How many of those people have a realistic shot at the job? What about all the political considerations? I don’t think the labor market bears heavily on their decision making process, and I don’t think we have any good way to measure how much efficiency we lose because of these actors’ relative lack of accountability.

3. Risk aversion: This counts as an agency cost in one sense, but also deserves distinction in a more general sense. Defense spending is usefully analogized to insurance spending. A young, healthy person might (in a vacuum) opt not to pay for health insurance. Such a person probably expects to spend very little on health care, so it makes some sense. He’s done his cost/benefit analysis: The likely cost of health care is less than the cost of health insurance. Of course, another important detail is that hospitals may not refuse emergency care. So in the event of some catastrophic health event, this young, healthy person doesn’t go untreated for lack of insurance.

The subject of risk aversion is where the insurance analogy breaks down. We treat the people in charge of military spending decisions like insurance actuaries. They weigh the need for protection against its costs, (hopefully) considering the costs (e.g., the opportunity cost of using that money for defense instead of for, say, education or welfare) and the risks (e.g., the threat of Iranian and North Korean nuclear weapons). Actuaries for an insurance company are looking to maximize profit; they do this by calculating risk as accurately as possible so that the insurance company can insure as many people as possible, charging them as little as possible to ensure their business. If the AllState actuary can find an inefficiency that is making GEICO overcharge its customers, AllState can steal those customers and turn a bigger profit. But of course, charging too little would mean that AllState would operate at a loss. In other words, insurance actuaries are looking to operate at the margin—the most competitive price. Just like WalMart and K-Mart are going to sell the same product for pretty close to the same price—as cheap as they can make it and still profit—so will insurers. But defense spending policymakers aren’t “selling” anything. They have to pitch their budget to Congress, sure, but that process seems (to me, as an outsider) more political than anything else. How can we know they’re properly motivated to act efficiently? Well, in fact, there seems to be a lot of evidence that they aren’t.

Decision makers themselves are somewhat risk averse because they will get blamed for whatever happens, regardless of whether they deserve blame. People lose jobs over this stuff, and nobody wants to get pegged as the one who let a terrorist attack happen, for example. All this is true despite our conscious balancing of costs and risks—we all should know that reducing risks to 0 is probably impossible, and when we cut costs, risks usually increase commensurately (of course, this is all subject to all sorts of other influences, as well). Just like corporate officers are likely to play it safe on investors’ dimes, so are policy makers who don’t want the next big attack to happen on their watch.

So that’s inefficient, but is it necessarily bad? Investors want corporate agents to act efficiently, even when it means taking risks, because investors have diversified risks and they gain the most when their investments operate optimally, even if that means one fails now and again. They’re playing the market in general—not specific corporate entities. But we as American citizens aren’t playing the market; we’ve got all of our eggs in this basket. This brings us to the second point about risk aversion, for which I’ll return to the insurance analogy.

As I said before, even the guy who has no health insurance will get treatment if he goes to the ER—although the cost will be very high. In the real world of defense, however, there is no hospital that can bail us out. This is really scary when you consider that these days the “catastrophic health event” is nuclear war. There’s no safety net to handle that—only preventative measures and practiced responses, and all of that is really expensive. This sort of thinking is somewhat alarmist, yes. Nevertheless, it shouldn’t be ignored, when considering this issue, that included in the “risks” column is a nuclear war whose consequences could be unfathomable—even if the risk is very small. It’s hard not to be risk averse in face of such a worst-case scenario, even if it leads to inefficiency. As a society we consider it moral and responsible to pay for things like life insurance, even when the odds of dying are very low. It’s interesting to me—and I don’t mean the term in a loaded sense—that many don’t seem nearly as invested in insuring our national life, as it were.

So, to review: When we debate what to do with defense spending, we should keep in mind that it’s a question of balancing the toll it takes on our budget with the protective benefits those dollars buy us. The problem with this equation is that the only people who really know how much protection our money buys us—and how much we need it—are probably biased, and in any case there are a worryingly small number of them. It puts the average citizen in a poor position to develop an informed perspective.

Of course, one thing on which everyone does seem to agree is that the $600 billion cut that the sequestration proposes is a bad idea because it doesn’t give the secretary of defense a chance to plan and adjust in anticipation. Hopefully we can get that worked out.