Voting by coal-powered states

It's Monday, I'm sick, and I just finished running a colloquium. Hooray! Here's some more quick thoughts on state interests and voting. Today's question: does the list of states that get more than 80% of their power from coal better predict climate legislation voting?

The short answer: No and Maaaaaaybe.

From last post, the list of coal power states is: Indiana, Kentucky, Missouri, North Dakota, Ohio, Utah, West Virginia, and Wyoming. To this, I add Hawaii, the only state which gets most of its power (75%) from oil.


With regard to the House, the picture using this list is actually almost exactly the same as the picture using coal production. 65% of Democrats from coal-power states voted for climate legislation, vs. 85% among states that get less than 80% of their power from coal. This compared to 66% yes votes from coal-producing states. To the extent that there's an effect, it appears to be almost exactly the same level (caveat: this list of states seems to me to be a slightly bluer - or at least more purple - list, so perhaps the effect IS stronger, but is masked by a difference in ideological background noise.)


With regard to the Senate, my best guess at the picture of probable voting should a future bill actually get a vote looks like this:


Likely Yes
Akaka (HI)
Inouye (HI)

Maybe
Brown (OH)
Bayh (IN)
Rockefeller (WV)


Lean No and Likely No
McCaskill (MO)
Byrd (WV)
Conrad (ND)
Dorgan (ND)


To me this looks much like the Senate list from the coal production analysis; there's definitely grounds for claiming an effect though there are more Senators from these states that are willing to talk than I might naively have predicted. It feels to me like there's a hair more effect here - one fewer "yes", one more "no" - but that's just a feeling gathered from reading about these senators' positions and it may or may not be valid. In this analysis, the role of Montana is played by Hawaii.


In what should probably be a standard disclaimer, I should note that I'm under no illusions that any of this is going to plausibly yield anything statistically significant.