Investigation: Energy subsidies

[rufuspollock]

Investigate the distortions of fossil fuel subsidies on the energy market

See the Pitch Presentation

[Originally started: summer 2012]

[rufuspollock]

@vndimitrova did the original work on this ...

[rufuspollock]

From @jwyg:

Saw this story yesterday about coal power subsidies [1], which reminded me to ask whether anyone has looked into data on energy subsidies in their country?

Perhaps this could be an interesting data expedition, or mini-project?

http://www.independent.co.uk/news/uk/home-news/consumers-to-pay-dirty-coal-power-subsidies-for-years-8724925.html

Olav Anders Øvrebø mentioned

The IMF recently published a good report on energy subsidies: http://www.imf.org/external/np/sec/pr/2013/pr1393.htm

To which @jwyg responded:

Olav: I'm astounded at the following claim in the IMF report you forwarded:

"energy subsidies amount to a staggering $1.9 trillion worldwide—the equivalent of 2½ percent of global GDP, or 8 percent of government revenues."

It might be interesting to further explore the aggregated/estimated figures in the tables in the appendices. [1]

The IEA also publishes this: http://www.iea.org/subsidy/index.html

But I can't seem to see a way to get the raw data.

Anyone else interested in digging into this?

[eldang]

Some relevant details from reading the IMG report:

• That $1.9 trillion is specifically for post-tax subsidies. They seem to be counting pre-tax subsidies (i.e. the more direct, easier-to-quantify sort) as a separate $480 billion.
• It's also probably an underestimate in some sense, because they don't seem to be accounting for indirect support like subsidising the infrastructure to deliver energy.
• On the other hand, it seems to include a lot more than just direct tax reductions on energy providers. Going by figure 5 (p14 of http://www.imf.org/external/np/pp/eng/2013/012813.pdf ) the worldwide direct tax subsidy seems to be estimated at somewhat less than the pre-tax subsidy, whereas the "externalities" category seems to be more than half of the total subsidy. Or in the terms of the appendices (see p43), the "revenue component" seems to be somewhat less than pre-tax subsidies, while the "corrective tax" to price in externalities seems to be more than half of the total.
• I've started digging around in the references, and the appendices to https://www.imf.org/external/pubs/ft/wp/2006/wp06247.pdf (starting at p24) outline a reasonable-sounding method for estimating the gap between prices actually charged and what a unit of energy "should" cost. I'm not certain, but I would guess that this is what the authors of the more recent report used to estimate the revenue component. It does seem to fit with the vaguer description on p46 of the newer report.
• The externalities estimate is necessarily more complicated. Here are the references that they seem to be using as sources:
  • https://www.imf.org/external/pubs/ft/wp/2011/wp11168.pdf & http://www.econ.wisc.edu/~scholz/Teaching_742/Parry-Small.pdf for gasoline - though they both seem to ignore the cost of road building, and the UK & US one seems to ignore road maintenance, so if anything they're probably underestimates.
  • http://www.epa.gov/otaq/climate/regulations/scc-tsd.pdf for the cost of CO2 emissions, though on p45 they list estimates from half to 3 times as much as the one they picked, without giving a rationale for why they chose that particular one.
  • http://www.nap.edu/catalog.php?record_id=12794 (I haven't found a free source yet) for SO2 emissions, which does seem to be sort of a gold standard.
  • http://www.oecd.org/environment/mortalityriskvaluationinenvironmenthealthandtransportpolicies.htm to determine the local pollution estimates across countries

I hope this is some help! Beyond this, I think contacting the authors (or the "Approved By" people on p2?) seems worthwhile. I can sort of see how to recreate some of their calculations using the estimates and methods in those references, but it would be an awful lot of work just duplicating what they've presumably already done.

To my mind, a really interesting data expedition would be to go beyond this paper, and try to quantify indirect subsidies like those for road & pipeline building, streamlined planning approval, etc.