How should we measure energy subsidies?

Data Rule 4: Normalize properly; ask, “Per what?”

Robert Bryce has an interesting post on his Substack titled Actually, Solar Is Getting 302 Times More In Federal Subsidies Than Nuclear. I think he makes a valid point, but I disagree with his normalization. Before we get to his computation, let's have a look at the recently released paper, Federal Financial Interventions and Subsidies in Energy from 2016 to 2022.

The report includes exactly what the title implies, with energy industry subsidies divided into four categories:

Tax expenditure estimates, which measure the effects in federal government revenue resulting from preferential tax treatment for particular taxpayers

Direct expenditures to non-federal recipients (that is, both energy producers and consumers) in the form of a grant, loan, or other financial assistance award

Research and development (R&D) support in the form of a grant, loan, or other financial assistance award made to non-federal recipients

Loan guarantees (a form of credit subsidy) that provide financial support for innovative clean energy technologies that typically do not qualify for conventional private financing because of their high technology risks

Given these categories, Figure 1 provides the total tax expenditures by energy type for 2016, 2021, and 2022. I chose the first year of the data and then the last two years to get a sense of variability and changes.

Figure 1. Data from EIA

There are a few noteworthy features to mention. First and foremost, hydropower receives nothing. Second, wind and solar are receiving much more than oil and gas. The 2016 negative subsidy for oil and gas is attributed to two factors: Temporary 50% expensing for equipment used in the refining of liquid fuels and Expensing of exploration and development costs. Third, nuclear subsidies are insignificant in comparison. It's fascinating to see the actual tax expenditures, but these are totals, and we have to wonder if we're normalizing correctly since we aren't normalizing at all. In other words, do we receive more bang for our buck in one category over another? This is Data Rule 4: Normalize properly; ask, “Per what?”

When comparing homicides between two cities, for example, it is not reasonable to compare the number of murders; instead, we compare murder rates per 100,000 population. The reason for this is that a larger city is more likely to have more murders. We use per capita normalization for a variety of variables in order to establish fair comparisons. Bryce takes the subsidy number from Figure 1 and normalizes it for BTUs produced by the source in the same year as the subsidy. This appears to make sense at first glance. If we get more energy, then a greater subsidy may be worth it.

Figure 2 is a reproduction of his graph, but with the addition of the additional energy categories. He also rounded the BTU to quadrillions and kept only one decimal. As a result, our numbers varied slightly. I also normalized by trillion BTU rather than quadrillion BTU because the EIA used trillion BTU and quadrillion scales dollar values by 1000, making them appear worse.

Figure 2. 2022 data from the EIA.

According to this statistic, Figure 2, Bryce is correct in claiming that solar subsidies are 300 times bigger than nuclear subsidies (4,086,692/13,639). In fact, solar subsidies per BTU are higher than for every other energy source, including wind (solar is 4.4 times higher) and oil and gas (solar is 135 times higher). This metric has three flaws, in my opinion.

To begin, the BTU data in the same EIA report is for energy generated in 2022, although subsidies do not always correlate to energy production in the same year.  Meanwhile, the various energy sources are in different phases of development. The 2022 solar energy generation is based on installed energy, which is still minor for solar. In essence, money may be spent to increase capacity that does not currently exist. In the case of nuclear, we've only installed two new reactors since 1996, with one more on the way, so there's not much extra capacity to subsidize. Oil and gas is a mature business that generates a lot of energy, and subsidies are used more to keep it that way. In short, nuclear, oil and gas, and solar are all at various phases of development. We're not really comparing apples to apples here.

Second, and this is where Data Rule 4 comes in, we must normalize by some type of lifetime energy generation for the money spent. Here's a made-up example: Assume we send $100 for solar, which will produce 10 BTUs of solar energy each year for the next 25 years. I would argue that the $100 spent is equivalent to $100/(10*25) = $0.40 per BTU. On the other hand, my $100 may only purchase me 125 BTUs of oil and gas energy this year. That works out to $0.80 per BTU. In other words, I need to know how much electricity we will obtain in the future from our solar investment. We would like to see a similar calculation for nuclear and other energy sources.

Third, we might want to consider other external costs. Fossil fuels are effectively being subsidized with the costs to clean up climate-related disasters and added health care costs due to pollution. Should we charge for nuclear power based on the likelihood of a disaster? Should the offshore wind business incur additional expenditures as a result of the detrimental effects on the fishing industry? If the federal government picks up any part of the tab for any of these cases, it is effectively a subsidy.

I believe Bryce is on the right track in striving to properly normalize and quantify how the United States spends energy subsidies based on a return on investment. Society should be aware of how much money is being spent and what is being received in return. Unfortunately, there is still work to be done here, and Bryce's work may easily be misconstrued as solar being prohibitively expensive. In addition, he definitely demonstrates a bias when he says:

Subsidies for the landscape-destroying wind energy business

Still, there is a point here. People can still make different value judgments after we have a detailed cost-benefit analysis of different energy sources. Even if wind energy were the most cost-effective type of electricity, Bryce may decide it isn't worth it because it destroys the scenery.

Please keep in mind that I am not advocating for one energy source over another, but rather that the measures used to assess them are insufficient and thus inaccurate. Personally, I believe we are not spending enough on nuclear, and as I have already stated, I do not believe we can run the world on wind and solar alone, regardless of cost.

Please share and like

Please help me find readers by forwarding this article to your friends (and even those who aren't your friends), sharing this post on social media, and clicking like. If you're on Twitter, you can find me at BriefedByData. If you have any article ideas, feedback, or other views, please email me at briefedbydata@substack.com.

Thank you

In a crowded media market, it's hard to get people to read your work. I have a long way to go, and I want to say thank you to everyone who has helped me find and attract subscribers.

Disagreeing and using comments

I'd rather know the truth and understand the world than always be right. I'm not writing to upset or antagonize anyone on purpose, though I guess that could happen. I welcome dissent and disagreement in the comments. We all should be forced to articulate our viewpoints and change our minds when we need to, but we should also know that we can respectfully disagree and move on. So, if you think something said is wrong or misrepresented, then please share your viewpoint in the comments.