> if you want to destroy, say, every enemy military base simultaneously, the number of conventional precision weapons needed for that would be cost prohibitive

> if you want to destroy, say, every enemy military base simultaneously, the number of conventional precision weapons needed for that would be cost prohibitive For the U.S. and similarly wealthy militaries, the math doesn't seem to support that statement. You don't need to destroy every base, just enough to effectively eliminate the enemy's military capability. Let's say that's even 10,000 targets - that's not a lot of munitions for a war. At $2 million a pop - rounding up Wikipedia's price for a high-end cruise missile - that's $20 billion, a pittance in war. The Iraq War, not a major one, cost over $1 trillion AFAIK (up to the first pullout, that is; it's not over yet). For poor countries, who lack international networks for targeting and communication in addition to the munitions and launch systems, I think the statement holds up. It's also cost-effective for another reason: 1 nuke aimed at Seoul or Tokyo or LA is enough to deter attackers in almost any situation.