"So, just to clarify, you are certain that the - presumably third rate - actuarial boffins at Munich Re are working in nominal dollars, whenever they state historic natural disaster cost comparisons?"
I never saw any reference to Munich Re actuaries in the post to which I responded, which was:
But, hey, even if you insist that those economic losses are stated in real terms, it implies a real compound annual growth rate of : [(2.97/1.63)^(1/20)-1] = 3.0%pa.
This is no different to the rate of real global GDP growth over that period, so in economic proportion terms the losses from natural disasters are no different over the past 20 years as they were in the preceding 20 years [*]
Source: World Bank
[*] And remember, that's ifthose losses are indeed presented in real, not nominal, terms.