Monday, January 23, 2017

Willis is about as wrong as usual

Willis wants us to know that carbon taxes, being regressive, are "cruel." So very, very cruel. This he will prove with data, because he is a Serious Person and not at all a shambolic dishonest embarrassment to the human species.

Immediately, however, we run into the problem that the numbers provided are not remotely plausible. All the numbers look far too high, but the figure on the bottom for thos making less than $20,000 a year is the standout. While poor Americans often spend more of their incomes on basic necessities like energy, the idea that they spend 40% of their income on energy is, frankly, utterly ludicrous.

To get here he cites a number of sources, fucking up in unknown ways to get answers that are wildly wrong:
Someone challenged me on this claim about energy taxes the other day, and I realized I believed it without ever checking it … bad Willis, no cookies. So of course, having had that thought I had to take a look.
The Energy Information Agency (EIA) collects data on this, with the exception of gasoline usage. I got the most recent data, for 2009. (Excel workbook). Gasoline usage figures are here from the Bureau of Labor Statistics. Finally, income averages by tiers are available here from the Census Bureau.
While he laments the loss of his cookie, I would actually offer him two small cookies here. One, he is making an effort to check his intuitions against data. Two, the places he is going for data contain good reliable information: the EIA, the Bureau of Labor Statistics, the Census Bureau.

At that point Willis' cookie supply dries up, because he has mangled the data to get a grossly wrong answer. People who make less than $20,000 per year spend an average of $1,571 dollars a year on energy (slight more if you include gas). So Willis is asking us to accept that the average income of this group is $3,900 a year, and after spend almost half that on energy they are paying for housing, food, and clothes with $200 a month.

The sources Willis cites do not provide numbers for the average income of a household in a given income bracket. Possibly he was using the means as averages, although that would not explain his numbers by itself. The Social Security administration does provide averages, fortunately:

 If you plug those numbers in you get an average income, for those making less than $20,000 per year, of $8,124. That would make the energy costs in this bracket 19% (plus a couple percent for gas), which is quite high compared to expert estimates the very poor spend about 10% of their incomes on energy, but which is less than half the number Willis somehow obtained.

I don't spend as much time chasing the nonsense on WUWT as I used to, because so much of it is repetitive and deadly boring. It's also because those that want to be informed have a much clearer understanding of the workings of ideological alternative realities than was the case when I started this blog in 2010. For reasons I would never chose, the country and the world are much more familiar with the working of denialism than they were then.

This one caught my eye because Willis is trying to be good. He's looking up good sources. He's trying to test his intuitions against the facts.

He fails because he is still operating out of a denier mindset in which he expects to find something all the experts have missed. A simple google search should have altered him that the people who study energy poverty professionally put the cost of energy in the US to the poor at, at worst, 10-20%. He should then have tried hard to figure out where his calculations went wrong.

Willis has never learned how to manage a data set, which is to say, he's never learned the art of doing a little math as you go, checking the numbers against common sense, and circling back to recheck when the answers starting coming out weird. This is basic, habitual skepticism, and the scientists I know do it so adroitly it becomes invisible and almost unconscious, a reflex. Willis clearly doesn't know that he doesn't know how to do this. Despite gestures at an appropriate method of inquiry, the cruel cold talons of Dunning-Kruger still hold him firmly in its grasp.

No comments:

Post a Comment