Agree with you on infrastructure and more importantly the dollar adjustment.
CPI is a nice handy tool for things that could damage the basket of goods in contained in the CPI but it's wildly inappropriate for weather disasters because the weather disasters primarily cause damage to capital infrastructure rather than consumer goods. Case-Schiller, M3, or DOW, S&P 500, would be much better numbers to use for adjusting damage to capital infrastructure. Even then it's kind of a crap shoot as none of those price indexes really accurately reflects the kinds of property damaged by weather.
Is this really climate related? My initial suspicion is that as technology advances, we just have more complicated (expensive) systems and infrastructure that are harder to repair because of the expertise required. We also become more and more reliant on technology every day, but also more and more incompetent when that technology suddenly fails. It is sort of a 'house of cards' effect.
Whereas we definitely have more sensitive equipment inside a given building, we also have improved materials and designs that can better-withstand weather events.
I wonder to what degree events that would have formerly been billion-dollar disasters have dropped off this chart, because power plants, hospitals, etc are resisting forces that would have seriously damaged older buildings.
And on the other side, I wonder to what degree our lax infrastructure spending has exacerbated the cost of these disasters. Truly if we were spending on levees and spillways at engineer-recommendations, recent hurricanes and floods would have caused notably less damage [1].
I guess my only point is that while I concur that it's careless to try to draw conclusions from this data, I don't think we can clearly assume that progress alone shows us which direction the data is skewed and why.
[1] Would we have even had to open the Morganza spillway if Louisiana's levees had been maintained and updated over the years?
The part I found most interesting was the points-of-origin graph, and how closely that correlates with the Tropic of Cancer. But, there's also a historical graph there for your edification.
The historical graph here does not appear to show any upward trend 1950 to present for frequency of EF3-EF5 tornadoes; if anything, they may be becoming slightly less common.
In their report, they found no long term trend for drought conditions. I can find nothing that casts doubt on their research.
If I were serious about figuring out whether the cost figures for natural disasters were attributable primarily to increased costs or primarily to increased severity of natural disasters, I'd use earthquakes as a comparison -- if it's an increase-of-cost issue, then I'd expect to see the same increase-of-cost for historical earthquakes. If the increase-of-cost for earthquakes is substantially less than of the natural disasters in this report, then I'd be inclined to attribute it to a greater frequency and severity of natural disasters.
* If I were serious about figuring out whether the cost figures for natural disasters were attributable primarily to increased costs or primarily to increased severity of natural disasters, I'd use earthquakes as a comparison -- if it's an increase-of-cost issue, then I'd expect to see the same increase-of-cost for historical earthquakes. If the increase-of-cost for earthquakes is substantially less than of the natural disasters in this report, then I'd be inclined to attribute it to a greater frequency and severity of natural disasters.*
On the other hand, we've got a lot better at building earthquake-resistant structures.
Also, the number of earthquakes causing any significant damage in the United States over the past two hundred years is small enough to be statistically iffy.
Oh my god, in 2010 there are infinity more car related deaths than there were in 1880. Are we getting worse at driving, or are cars just less safe?
There are more people, there is more stuff, and there is a little thing called inflation. People live in areas of Florida that were swamp 20 years ago.
This is true, though the CPI is a fairly lousy way of adjusting for inflation, especially over long time periods, and when talking about things which aren't consumer prices.
For instance, the Empire State Building cost $40 million to build. According to a CPI inflation calculator I found, that comes out to $600 million in today's money. But good luck trying to rebuild a 100+ storey skyscraper in New York City for $600 million nowadays. You'd probably spend that much just on Environmental Impact Statements and arguing with the City about whether you're likely to displace any endangered frogs.
Areas that were corn fields 20 years ago now have houses, malls, commercial, mfg plants, etc.
We have a greater population that is spread out. What was one a small crop disaster for farmers is now a very costly disaster for a municipality.
Does the study account for this? Is this a valid concern?
tl;dr: We have more "stuff" to ruin now.