Those figures seem to come from here:

https://www.cbo.gov/publication/43993#section0

In particular, the $3.2bn/year is a tax write-off sort of 'subsidy', not a grant sort of subsidy, which I believe solar has received recently (section 1603).

Without doing more digging, I can't tell whether what those oil and gas write-offs are for. Exploration? Research?

Why is that an important distinction here? Both direct subsidies and tax credits/rebates/write-offs have the same net effect on the government's coffers.

The Wikipedia article cites another article that has the breakdown (over a period of years):

1. Foreign tax credit ($15.3 billion)

2. Credit for production of non-conventional fuels ($14.1 billion)

3. Oil and Gas exploration and development expensing ($7.1 billion)

https://www.eli.org/sites/default/files/eli-pubs/d19_07.pdf

Have you read that pdf you linked to?

Similarly, the study counts funds used to support carbon capture and storage programs1 as a fossil fuel subsidy, despite their potential to reduce the emissions associated with burning coal.

If that's counted against oil and gas 'subsidies', that doesn't seem right.

Foreign Tax Credit ($15,300) - IRC Section 901. This is a generally applicable credit that is intended to enable taxpayers earning income or profits abroad to avoid double taxation.

Wait - so a good 40% of the total is actually a non-subsidy that is the consequence of the US having a tax treaty with a foreign country? Not being able to double-tax oil and gas companies is considered a 'subsidy'? The US has a tax treaty with just about every civilized country in the world that has provisions to avoid double-taxation because to do otherwise would be pathological. Counting that is 40% of the 'subsidy'?

The foreign tax credit is an oil subsidy since the 'creative' oil companies started structuring their royalties to foreign governments as taxes, so they can deduct them against profits.

If Exxon has to pay a royalty of $1/barrel to Nigeria for access to their oil fields, rather than account for the royalty 'properly', they structure the deal to pay a $1/barrel tax to Nigeria. Then they can claim that they paid taxes instead of royalties and get the offset from the foreign tax credit.

I don't doubt you but can you explain this in greater detail? I have filed for the foreign tax credit and it disqualifies most non-income taxes. I would presume royalties would be covered by this.

Royalties are excluded, and the IRS has clamped down more in defining what exactly counts as income tax, but it's still common for oil and gas companies to structure their payments as taxes to get the credits. For instance, in Saudi Arabia, there is an 85% corporate tax rate on oil/gas companies and a 20% corporate tax rate on the average business. It sure looks to me like that's a 20% tax rate plus a 65% royalty rather than an 85% rate, no?

http://taxfoundation.org/blog/treating-royalties-governments...

<citation required>

> Why is that an important distinction here? Both direct subsidies and tax credits/rebates/write-offs have the same net effect on the government's coffers.

It has a different effect on the entity being taxed. A tax write-off increases the rewards for an already profitable proposition, but tax write-off don't normally incentivise someone to do something that is a loss (unless the write-off is transferrable to other profits, but that's a separate question). Subsidies, however, can make a loss into a profit, incentivising people to do it. That's exactly the point of the original solar subsidies.

Now that the solar industry appears to be relatively well bootstrapped, it's not unreasonable to remove the straight subsidies. I think solar still enjoys many if not all of the same tax write-offs as other energy sources.

Add the military spending to ensure control of oil producing countries and cheap oil, and it's more like $1 trillion.

Looking at the gross subsidy is misleading. Federal gas tax revenues alone are about $35bn/yr.

Wait, are we talking gasoline for cars here, or natural gas for energy production? Because those are very different things. It's entirely possible for gas and oil power plants to receive subsidies while gasoline for cars is heavily taxed.

The excise tax is intended to pay for transportation infrastructure though, not to disincentivize gasoline consumption.

Whatever the intentions of the people imposing the tax, the effect is to disincentivize gas consumption. My point is, if the government taxes certain activities by $x and subsidises the same activity by $y, quoting only $y doesn't give the full story.

As far as I know gas tax isn't paid by the producers, but sure we can combine gas producers and users for this discussion. Money taken or given to either one has the same effect on the demand for gas production.

When that $x in taxes is paying for the infrastructure used, then the gas producers/users are actually getting $y+x from the government. The $x, being a service charge, cancels out, and the only thing left is $y. $y is the difference between the world we live in vs. a world where the government paid no special attention to the industry. It's the most important number.

Even granting your argument, if the government takes $x as a service charge for infrastructure, and subsidises by $y, that would be a net subsidy of $y, not $y+$x.

But $x is not a service charge. It bares no relation to the amount of anything used. They can't opt out of the road part of the service charge and build their own roads. The tax is not hypothecated. Whatever the rhetoric of politicians, it doesn't bare any resemblance to a service charge.

> that would be a net subsidy of $y, not $y+$x.

That's exactly what I was arguing, yes. That $y is important and $x is irrelevant.

> They can't opt out of the road part of the service charge and build their own roads.

That has no real bearing on the economics of the situation. Paying $x to the government for roads and paying $x to a private contractor for roads work the same way. That's why the net subsidy is $y.

If there is a tax that's applied specifically to gas companies that isn't directly paying for infrastructure they use, then that tax can be subtracted from the subsidies. But the gas tax doesn't fit that bill; if anything it undercharges.

> If there is a tax that's applied specifically to gas companies that isn't directly paying for infrastructure they use, then that tax can be subtracted from the subsidies. But the gas tax doesn't fit that bill; if anything it undercharges.

The gas tax is only one of many taxes that these companies pay - you have to sum all their subsidies and subtract all their taxes. That's the point. If you're not doing that, comparisons with subsidies to other industries are going to meaningless.

You subtract the taxes specific to the industry. Not all their taxes. Are there any such taxes?

I subtract all their taxes. If Exxon pays $100bn/yr in various normal taxes, and the government pays them $1bn/yr in fossil fuel subsidies, that's a net subsidy of -$99bn/yr.

How can it be a fair representation of their situation to only say they receive $1bn/yr in subsidy?

Because something like income tax is a given. We're not comparing against a world with no government. We're comparing against a situation where they get no special attention. You can use whatever word you want for it, but if a company pays half as much tax as everyone else that's a big deal.

You're right to say that it's not fair to only list the $1bn. It should be put in context of the $100bn of normal taxes. But reducing it to "$99bn" is not a fair representation either. It doesn't tell you if they otherwise would have paid 100 or 200.

I'm quite happy to accept the reporting of both gross taxes and gross subsidies. My objection is to only comparing the gross subsidies.

The gas tax is paid for by customers at the pump. It is not a tax being paid by the oil companies.

Who cares? If there's a tax being levied it's going to be passed on to consumers wherever it's applied.

That's only federal subsidies.