There is a campaign among the military-industrialists to set a floor on defense spending at 4% of GDP as “the only way we can stop the inexorable slide of national defense.”

This article argues why percentage of GDP is the wrong yardstick.

The Bush Administration requested $541 billion for national defense in FY 2009. […] The Administration also submitted a separate $70 billion placeholder request for ongoing military operations in Iraq and Afghanistan, but Secretary of Defense Gates later provided an updated estimate of $170 billion for FY 2009.18

[…]

This means the United States will spend significantly more, in inflation-adjusted dollars, for defense in FY 2009 than it did during the peak years of the Korean War (1953; $545 billion), the Vietnam War (1968; $550 billion), or the 1980s Reagan-era buildup (1989; $522 billion).19 The United States is also projected to spend more on defense in FY 2009 than the next 45 highest spending countries combined, including 5.8 times more than China (second highest), 10.2 times more than Russia (third highest), and98.6 times more than Iran (22d highest). Indeed, the United States is expected to account for 48 percent of the world’s total military spending in FY 2009.

So why the discrepancy? Military-industrialists say that military spending is a falling proportion of GDP while these numbers say that dollar amount is increasing. The answer is, of course, that GDP is increasing.

His numbers are 4 years out of date, but Ben (of Ben and Jerry’s) explained it well with Oreos

or, if you prefer, an old-fashioned pie chart.

This handy-dandy poster has more up-to-date (2009) numbers:

click the poster for a fancy interactive version