In an earlier post, I listed some questions for interventionists to consider before advocating their interventions. This is part of my ongoing crusade to get interventionists to think about things as they actually are as opposed to a blank slate. These two modes of thinking I call “status quo reasoning” (seeing the world as it is) versus “state of nature reasoning” (seeing the world as a blank slate).
Some recent research demonstrates the importance of status quo reasoning. In a new working paper at the National Bureau of Economic Research entitled “The Market and Climate Implications of U.S. LNG Exports” James Stock of Harvard University and Matthew Zaragoza-Watkins of UC Davis examine how the US moving from a net importer of natural gas to a net exporter has affected greenhouse gas emissions. For a while after fracking began in the US, natural gas prices became disconnected from other energy prices as the US market was flooded with natural gas and export options didn’t really exist. However, over the course of 2012-2016, approvals were granted and LNG export terminals were built, allowing the US energy companies to export LNG around the world. By tapping into the global market, US natural gas prices “reconnected” to world energy market prices: natural gas prices rose and began moving in tandem with oil and coal, as has been the historical trend.
What is interesting is their discussion on climate effects. In their abstract, they write:
Many supporters of a carbon tax erroneously claim there is no price on carbon. That is the fallacious “state of nature” reasoning I discuss above. In the state of nature, there is no price of carbon. But in the real world, there is a price on carbon. There may not be a monetary price, but there is always some form of price. When natural gas (a considerably cleaner producer of energy than oil and coal) entered the market, carbon use fell. Furthermore, as the price of natural gas coupled with other energy prices, people economized on natural gas usage (which, though cleaner than other sources of energy, still releases CO2), as well as on oil and coal. All in all, the effect of this was a reduction of carbon emissions as if a carbon tax was applied! The market, quite unintentionally, helped solve the externality by imposing a price on carbon.
What’s more, this market alternative to carbon taxes is probably, on net, more efficient than a similar carbon tax. That is, even though this coupling had the same effect as a $30/ton carbon tax, it likely did so with less waste. Whenever we talk about public policy, we must discuss the political process and how the sausage is made. Politics is a messy business, and imposing a carbon tax is no different. Even if we assume that the tax is passed costlessly and without any political shenanigans surrounding it, there are still the administrative costs of the tax (that is, any cost incurred while administering the tax: hiring people to collect it and calculate it, audits, and so on), which reduce its effectiveness. But with this market process, those potential costs do not exist.
In a state of nature, the interventionist could propose a carbon tax too high (or too low). But, taking the status quo into account, in looking at the total effects and not just the marginal effects (as Ronald Coase stressed so long ago in The Problem of Social Cost), we have a much clearer picture of what interventions may be necessary and which may not.
READER COMMENTS
Thomas L Hutcheson
Apr 17 2024 at 5:17pm
I agree that market developments (eg the development of fracking) should affect the optimal tax on net CO2 emissions (lowering it wrt a scenario whihc fracking had not been developed. The same is true of the amazing progress in solar and wind generation costs and battery storage costs. The optimal tax on net emissions would be variable, periodically re-estimated as technology and models linking emissions to harn improve.
Jon Murphy
Apr 18 2024 at 10:44am
I agree, although the implication of that statement is the optimal carbon tax is $0. That point aside, I have little confidence on most analysts’ ability to do such adjustments given how often they cannot describe the current reality.
Knut P. Heen
Apr 18 2024 at 9:09am
I don’t understand why carbon use fell when the supply of natural gas increased without the option to export. Burning anything containing carbon will produce CO2. The difference between hydrocarbons like natural gas, gasoline, diesel, kerosene, etc. is the length of the hydrocarbon chain (which changes the boiling point). Perhaps burning methane will give you slightly more energy per CO2 molecule released, but that difference must be trivial in the big picture. The main benefit of using natural gas is that it contains less sulfur than longer hydrocarbons and coal.
My reading of the paper is that they claim that lack of export possibilities reduced the price (the decoupling) such that Americans consumed more of it. Once the LNG facilities were in place, the price rose and Americans consumed less of it. In that sense, increasing exports worked exactly like a tax on American consumption.
BUT, the gas is going to be burned somewhere in the world anyway. Burning coal to liquify the gas and burning diesel to transport it somewhere before you burn it, increases emissions. A higher price will also increase the production of natural gas for export.
Jon Murphy
Apr 18 2024 at 10:46am
But not everything produces the same amount of CO2. NatGas releases far fewer emissions than oil and coal. So, as prices fell and consumers subsituted from coal and oil to natgas, their emissions fell compared to where they would have been.
Knut P. Heen
Apr 19 2024 at 6:20am
Apparently, there is a substantial difference between burning coal and natural gas, but the difference between natural gas and gasoline is rather small. It seems to be the ratio of hydrogen to carbon which determines the difference in energy released per CO2 molecule. Coal is pure carbon. Methane has a ratio of 4 hydrogen atoms to one carbon atom. Gasoline has a ration of about 2 hydrogen per carbon atom.
If the consumers substituted gas for coal, the price of coal would also have fallen, and there would not have been a decoupling of energy prices. The gas was simply consumed in addition to the coal and oil because it is difficult to swap in the short term. When export became possible, the price of gas rose, and consumers cut back. Gas prices are often out of whack with other energy prices before the necessary infrastructure has been put in place (pipe or LNG facility).
Jon Murphy
Apr 19 2024 at 7:03am
Ceteris paribus, yes. But the ceteris is not paribus. External demand for coal continued to rise.
Dylan
Apr 19 2024 at 9:51am
I don’t know who says there is no price on carbon, except perhaps as a shorthand for saying that the price of carbon is not enough to account for the externality of using it. You need to work backwards. How much warming do “we” want to allow? How much C02 and other GHGs can we emit before we go over that level? How much are we currently on track to emit? Is the later greater than the former? Than, the price of “carbon” is too low. Question, would you have the same objection if instead of a tax, a cap and trade system was implemented instead?
Jon Murphy
Apr 19 2024 at 4:08pm
Just about anyone advocating a carbon tax.
Then that shorthand would be incredibly incorrect.
Yes. My point here is general, not specific to carbon taxes. Carbon taxes were just the example
Dylan
Apr 19 2024 at 5:53pm
Then I think I don’t understand the objection. The way I’ve understood you is that there’s already a “price on carbon” that discourages use compared to a world where there isn’t that price. In the past I remember you arguing vehicle registration fees as one example of a tax that reduces consumption of gas, even though it isn’t explicitly a carbon tax.
I agree with you that this reduces usage on the margin, but I think it is irrelevant to the discussion. For the moment, just agree with me* that there is a ceiling of GHG concentration in the atmosphere that we can’t go above if we want to keep temperature rise to something manageable and that we have a rough estimate of where that ceiling is. The fact that there are things going on that reduce the emissions of GHGs already is irrelevant if the pace of emissions with all of those is still on track to go above that level. All of that stuff is already “priced in” to the market.
*I think large parts of this are up for debate and empirical observation, just trying to narrow the field of debate.
Jon Murphy
Apr 19 2024 at 6:13pm
All that is true, but still affects what the optimal tax (or cap, etc) is.
Say the optimal about of carbon in the atmosphere is X. One can get that with a tax of Y in a state of nature. Put then the optimal tax in the real world will be Y – whatever already exists.
If one argues for Y, one will cause too little carbon to be emitted.
Dylan
Apr 20 2024 at 8:01am
See, I think both of these are just not what is going on. Y isn’t the optimal tax in a state of nature, it is the tax based on the actual world we live in where all of those prices already exist. And, it isn’t assumed there is too much carbon because there isn’t a price on it. Instead, there is the amount of temperature rise that is deemed “safe” and an amount of CO2 concentration in the air that is consistent with that. The existing world, not some state of nature world, is the one that is on track to emit way more than that.
That’s why I asked about an alternative world with cap and trade, where we focus on total actual emissions, and the price of a credit would incorporate everything that happens in the world that either encourages people to emit more or less CO2.
Jon Murphy
Apr 19 2024 at 6:15pm
One thing I should say:
it is often just assumed that the level of carbon is too high, and it is assumed so because people think there is no tax. The argument typically goes: “There is no price on carbon. Therefore, the market has no incentive to produce the optimal level. There is too much carbon, QED.”
Max Molden
Apr 20 2024 at 6:45am
I like your emphasis on ” “status quo reasoning” (seeing the world as it is) versus “state of nature reasoning”,” Jon! I also only just stumbled upon your earlier article.
I want to note that my impression is that this mistake (or an analogous one) is also often committed by opponents of government intervention. (And I certainly include myself among them.) To elaborate, while it may be evident that intervention X is bad, we must engage in “status quo reasoning” when it comes to suggesting where to go next. It may be best to have had no intervention X at all. But once we have intervention X, it is not necessarily the case that the best solution is to simply get rid of it. And I mean that irrespective of political obstacles to disintervention.
My short note is inspired by something Mises wrote — I failed to locate the passage, however.
David Henderson
Apr 24 2024 at 10:59am
Excellent post.
Two points:
First, there is always a price of carbon. Because carbon is in all fossil fuels, and those fossil fuels have a positive price, there is implicitly a price of carbon, even without the effects that you discuss from the NBER study. Or, more exactly, there are multiple prices of carbon, depending on the particular fossil fuel.
Second, I don’t understand why you say, in response to Thomas L. Hutcheson above, “I agree, although the implication of that statement is the optimal carbon tax is $0.” How does his reasoning lead to the conclusion that the optimal carbon tax is $0?
Jon Murphy
Apr 24 2024 at 5:10pm
I’ll follow up with you either by call or by email, but I think I can show mathematically that taking long run transaction costs into account, trying to set an optimal Pigouvian tax leads to a paradox that results in any government intervention having an NPV of <0 even if there is an externality.