But then again, the U.S. economy grew quite a bit during that period. Maybe the share of our resources devoted to defense didn't actually change much. If one looks at defense spending as a share of gross domestic product, you actually see steady decline since the end of the Korean War, with slight upticks for the Reagan administration and the war on terrorism. Here's Eric Rauchway's chart on that metric, using both the Historical Statistics of the United States, a database published by the University of Cambridge, and more recent military expenditure data from the Stockholm International Peace Research Institute:
No, no, no, one might say. That's all wrong. Armed forces don't defend economies, they defend people. Military spending shouldn't increase in proportion to the economy that can fund the military, it should increase in proportion to the size of the population it's meant to protect. That's Wonkbooker Evan Soltas' argument. He produced this chart of GDP spending per capita, which shows that spending grew quickly due to Vietnam, the Reagan administration, and the war on terror, and then began to decline about as speedily thereafter:
So which one of these approaches provides the best way to evaluate military spending trends? None of them, I think. The cost of defending a country from attack shouldn't necessarily vary with population or the economy; after all, a wealthier country and a more populous country isn't necessarily more vulnerable to attack. The best comparison seems to be using inflation-adjusted spending, but the inflation adjustments going on in the first chart above are all screwy. The measure of inflation they use is for the economy generally. But maybe the cost of tanks and bombers has grown faster than the cost of bread and gasoline. If that's true, using standard inflation figures may look like we're spending more on defense than we really are.
Thankfully, the DOD issues inflation numbers (pdf) for defense spending from 1970 to the present. It turns out that defense inflation is a little greater than normal inflation. Defense prices have risen by about 553 percent since 1970, whereas the consumer price index, the most common inflation measure, only grew by about 487 percent. But they're not actually that different:
So the story told in the first chart seems basically correct. Defense spending fell during detente in the 1970s, rose during Reagan's tenure, fell during the 1990s peace and has grown since the war on terrorism started to a level surpassing that of the Reagan years.