Because any form of carbon regulation will have negative effects on a wide range of businesses and has to overcome their opposition, the first, feeble attempts at policy have been riddled with giveaways. That’s a long, sad story, and one I don’t want to belabor here. The upshot, however, is that there has been a preference for systems based on carbon permits with four main characteristics: too many permits are issued (minimizing climate benefits but reducing business burdens), the permits have usually been given away rather than sold (so that businesses don’t have to pay for them but can sell them or pass the impact on to consumers), there have been carveouts for entire industries (where coverage depends on political influence rather than policy effectiveness) and various kinds of offsets are allowed (voiding the need for permits if businesses make favored investments). This is bad policy on every score: ineffectual, opaque, and regressive.
Unfortunately, the reaction of many climate activists has been to confuse symptom and cause. Rather than tracing these bad policies to the current balance of political-economic power, they blame permits themselves, thinking that a switch from permits to taxes will fix the problem. The purpose of this post is to explain why they’re wrong—which can be done rather quickly—and then to make the case that, while there is a large overlap between the two approaches, if your overriding concern is the threat of catastrophic climate change you should go for permits over taxes.
First, look at how each method reflects the political economy of policy-making. At present, business interests have predominant power, and carbon control rules have to placate them. As we’ve seen, this takes four forms under a permit regime: too many permits, permit giveaways, carveouts and offsets. But each has its counterpart under a tax approach: taxes can be set too low, they can be offset or rebated through reductions in other taxes, they can be applied selectively, and there can be exemptions.
Taxes set too low: No one wants to pay taxes. Businesses in particular are in a position to do something about this. Carbon taxes will be subject to negotiation, and when the deal is signed, if businesses have their way—which to this point they have—the taxes will be weak and there will be little reduction in fossil fuel use.
Offsets and rebates: We live in a world of many taxes. There are income taxes, sales and excise taxes, profit taxes and all the rest. If the government institutes a new tax on carbon, businesses will say, let’s make this revenue neutral! Cut other taxes so that our net tax burden doesn’t go up. Formally, this is equivalent to a carbon permit giveaway: the system does provide an incentive to reduce the use of fossil fuels (since they are taxed), but businesses don’t pay the cost. Instead, they can pass it along to their consumers and even come out ahead on the deal.
Carveouts: Nothing says a carbon tax has to apply to all uses of fossil fuels. Business will push hard to have the tax apply industry by industry and then push again to exempt their own. Politicians tend to like this setup too, because more discretionary power over who pays how much tax translates into more campaign contributions, horse-trading heft and influence in general.
Exemptions: Just like businesses can take advantage of loopholes under a permit regime if they make approved investments (for which you should consult the good people at Carbon Market Watch), they can be given tax forgiveness if they do the same things. Anyone who thinks this is unlikely should sit down with any country’s tax code.
The bottom line is that junking permits and switching to taxes won’t do much good if businesses still have to be bought off in order to get a climate bill passed.
But political economy aside, what are the relative merits of the two approaches? First, let’s be clear on how they’re supposed to work. Taxes can be levied on producers or consumers at various points along the fossil fuel cycle; the idea is to reduce the use of carbon fuels by making them more expensive. The more upstream the tax—imposing it on those who bring these fuels into the economy rather than end users—the greater its flexibility and the easier it is to implement. Permits do the same thing, but by requiring a permit for fossil fuel use rather than charging a tax. They are analogous to hunting and fishing permits, which are limited in supply to maintain the population of whatever creatures people are hunting or fishing. As with taxes, the more upstream the permitting, the more flexible the system and the easier it is to monitor and enforce.
At first blush these two approaches are mirror images of each other. Suppose the following diagram represents the demand for fossil fuels, lumping all of them together in a single market:
The initial price is P1 and the initial quantity produced is Q1. A tax raises the price to P2, and demand then falls to Q2. A permit system lowers the amount that can be produced to Q2, and the price rises to P2. Under these assumptions, you can just flip a coin to decide which tool you want to use.
The biggest difference emerges when you introduce uncertainty. Suppose you don’t really know the relationship between price and quantity demanded, especially as you move away from the current market situation. Indeed, this is practically a given. First let’s see what this means for taxes:
A tax sets the price, but the amount by which fossil fuel use will be reduced is unknown. Now look at permits:
Permits set the amount of fossil fuels that can be burned, but the price that clears the market is unknown.
If you could have continuous adjustment of either taxes or permits in order to respond to unanticipated effects as they arise, you would be back at the first diagram, and all would be well. That’s far too great a burden for the political system to bear, however. It’s a huge deal to pass a law regulating carbon, and you are likely to be able to tweak it only rarely. In the real world, uncertainty will stick.
From this perspective, the choice between taxes and permits is about which uncertainty you prefer to live with. If your main concern is that carbon control regulations will be too costly, you should go with taxes: they pin down the cost and let the quantities of fossil fuels fluctuate. If, like the IPCC, your main concern is that we will overspend our carbon budget, you should be in favor of permits, which pin down carbon introduced into the carbon cycle but allow costs to fluctuate.
It is ironic that grassroots climate activists, who claim to be dedicated to meeting carbon goals no matter what, should denigrate permits and be so attached to taxes. My guess is that this analysis, which has been standard in environmental economics for 40 years, will be new to them.
There is a second reason why, in my opinion, permits are better suited to the specific demands of carbon policy. Recall that, if we want to reduce the risk of runaway climate change, we need to meet a fixed carbon budget. The constraint is not, how many tons can be extracted and emitted in any given year, but how many can be funneled into the carbon cycle over the coming decades in total. Any extra ton you allow this year has to be deducted from future allowances if the budget constraint is to be met.
Nothing in the tax approach addresses this condition. If too much carbon is emitted this year, there is no automatic mechanism that will make taxes rise the next; you have to do it by hand. Permits, however, can be designed to be intertemporal from the ground up. You could have some or all of them be undated, in the sense that they would authorize carbon extraction in general but not for any particular year. That way, if you think that more permits are needed this year, you can move some forward from next year—a shift that automatically stays within the long run carbon budget constraint. As we’ll see later, this could even be done by markets, without any conscious political action at all. In practice, I expect that going on a carbon diet will result in numerous unanticipated economic crunches along the way, creating pressures to temporarily relax policy controls. Under a tax system temporary is permanent: extra production in the near term is not made up by offsetting reductions later on. (Remember that the point is to prevent an accumulation of greenhouse gases, not their emission in any particular year.) Under a permit system, temporary could be temporary if the permits are designed properly.
Finally, a word about politics: pick a tool and get a discourse. If the tool is taxes, the discourse will be about how much we want to pay for carbon control. This brings us into the world of the social cost of carbon. Is the tax too high or too low? That depends on whether the benefit we get from cutting carbon a little more or less is above or below the tax rate. Those will be the terms under which politics gets carried out.
If the tool is permits, the discourse will be about how much carbon we want to allow into the atmosphere, and therefore how much climate change risk we’re willing to live with. This brings us into the world of the IPCC, climate science and carbon budget constraints. Are too many permits being issued or too few? That will depend on what our understanding of the carbon cycle and the climate system tells us about greenhouse gas concentrations.
I think climate activists should pick door number two.