Despite my self-imposed ban on blogging until I finish that aforementioned much-delayed column, I’ll briefly comment on Ezra Klein’s observations regarding the level of defense expenditures. Ezra writes:
Our wars are ending. Officially, the war in Iraq is over. The war in Afghanistan is drawing down. Osama bin Laden is dead. Typically, at this moment, spending drops by somewhere between 33 percent and 43 percent.
If the sequester goes into effect, the full cut to the defense budget will be about 31 percent. Think about that — during the war on terror, the defense budget increased by more than it did during the Cold War or the Vietnam War, and even with the sequester, the cut to the defense budget will be less than it was after either of those.
(1) Are national security expenditures the only domain in which spending crests but never fully recedes? (I should stress that Ezra is definitely not claiming that this is the case, but bear with me.) If we think of recessions as labor market shocks that trigger automatic stabilizers, it is striking that social spending seems to exhibit a similar pattern. To some extent, this could reflect structural changes: (a) the deteriorating labor market position of less-skilled workers, a group that might see a sharp deterioration during a downturn yet that never fully recovers as employers substitute capital for labor; (b) the stickiness of new means-tested transfers programs, which build constituencies (direct beneficiaries, firms that sell to direct beneficiaries, fiscally-constrained state and local officials relieved that the federal government is picking up the slack through deficit spending, etc.) yet which might also increase implicit marginal tax rates, thus discouraging labor force participation or prompting a reduction in work hours — this in turn might reduce market income; and when it comes to the social spending devoted to the actual delivery of public services, (c) Baumol’s cost disease surely plays a role. That is, high-touch services are more resistant to productivity-enhancing innovation than the production of goods, and so we will spend a larger share of income on high-touch services over time.
(2) Might national security expenditures be subject to these larger forces? We’ve discussed Jonathan Caverley’s thesis that affluent market democracies are ill-suited to counterinsurgency warfare as counterinsurgency warfare is highly labor-intensive while affluent market democracies are more inclined to rely on capital- and firepower-intensive military doctrines, due to the high cost of skilled labor. Looking beyond counterinsurgency as such, the human capital needs of the U.S. military are increasingly difficult to meet, particularly as the armed forces evolve in a more capital-intensive direction and as the demographic composition of the U.S. population changes. Levels of educational attainment are not increasing rapidly, families are shrinking, and young people with the mix of noncognitive and cognitive skills that are attractive to the military have other attractive options in even a sluggish economy. And so the cost of recruiting and retaining high-quality personnel is high and rising. Policymakers have made this problem “worse” in some respects by, for example, sharply increasing the value of educational benefits for veterans, which makes it more attractive to exit the military than to remain. This then means that reenlistment bonuses have to be higher, and so on.