Ever since I read through the "Stimulus or Stymied" transcript I have to say I was very puzzled by Valerie Ramey, and not Krugman's critique, which I agree with, but don't really care about for this.
Making a bomb provides nothing of secondary value to the economy. Military goods are not for public consumption. They also don't generally improve other aspects of our economy (the interstate highway system being a notable exception). If we spend $1 billion making a bunch of bombs, that money did employ some people, so the multiplier isn't zero, but it isn't as big as if it were to spread to or enhance other economic activity. (In fact, if a large fraction was employed elsewhere it could even be less than 1.)
On the other hand, if you improve transit then many other businesses and individuals can see a boost. If you rehab the power grid to be more efficient and less prone to failure then there are huge potential economic gains as a result. Massive military spending just doesn't do that.
I understand why she would chose military spending, for the study, but if you look at a "fiscal multiplier" as a measure of economic boost, then no one should expect military spending to be terribly large, particularly if this is to be used to generalize to other spending on things like infrastructure.
Yglesias makes something resembling this point when talking about the new, Japanese nationalist Abe.