Then there is also this, pointed out in a WaPo article coauthored by the economist Joseph E. Stiglitz in 2010:

Writing in these pages in early 2008, we put the total cost to the United States of the Iraq war at $3 trillion. This price tag dwarfed previous estimates, including the Bush administration's 2003 projections of a $50 billion to $60 billion war.
The real cost of the war, however, turns out to be even higher than the estimated $3 trillion! Why? Because the war was, one, funded by debt (Bush cut taxes just as the war began), and, two, a lot of the borrowed war money ended up outside of the economy, and so the Keynesian benefits (war-related production) for standard Americans were small. With no real money returning to the productive side of the economy (which still matters), growth slowed and the growth-dependent financial superstructure collapsed.