Can a carbon tax be effective and equitable policy instrument?
Posted by Daniel Hall on September 24, 2007
Peter Dorman at EconoSpeak doesn’t think a carbon tax would work.
Revenues would be too volatile to finance a government:
…relying on carbon taxes is also a terrible way to finance the government. We are talking about half a trillion dollars or so in revenue, so the percentage of financing would be quite large. Income fluctuates, and that is a problem, but the spending on a particular set of items, like fossil fuels, has the potential to fluctuate even more. Example: suppose we really are facing an oil production peak, and scarcity causes the price to spike? Every 10% rise in oil prices will tend to cause something like a 5% reduction in long run demand (I’m rounding here – and thanks to Gar Lipow for his valuable work in collating the evidence), but this also means less carbon tax revenue, potentially a lot less. This is a serious problem, one that the green taxers have not really confronted.
He says that taxes won’t provide any certainty about emissions:
[It is] an elementary economics error as to suggest that taxes and cap-and-auction are “effectively” the same. In an uncertain world this is false. From a conventional benefit-cost perspective, Weitzman showed long ago that there were important differences depending on the slope of the marginal benefit and cost functions. Translated into common English, if we are uncertain about the long run relationship between the price of carbon emissions and the amount of emission – and we very much are – and if the risk of allowing too much climate change is greater than the risk of economic indigestion from trying to be too green – which seems pretty clear to me – then permits are the right choice. By controlling the number of permits we control our most important impact on the earth’s carbon budget, but allow prices to wander. By setting a tax we control the price but allow the amount of pollution to wander. That’s a big difference: you might say, given the gravity of what is at stake, that it’s the difference between ecological responsibility and irresponsibility.
More of Dorman’s problems and my response below the jump.
Taxes are politically infeasible and will be plagued by special interests:
It is a mistake to get drawn into a debate over how high to set carbon taxes. No one wants to pay taxes. The result will be a half-hearted effort riddled with safety-valves and loopholes. Perhaps this is why the big money is behind a tax approach: they know they will be let off the hook.
It all adds up to a clear argument for cap-and-trade:
…the question is what to do about the problem, and I would strongly encourage those who put ecological responsibility and social justice first to stick to their guns. We should have permits because they put the planet first, and we should auction them and distribute the revenues on a per capita basis because it is fair and economically sound. … When we talk about how many permits to issue… the debate is over how much carbon accumulation, and therefore how large a risk of catastrophic climate change, we are willing to accept. That’s the conversation we need to have.
My responses, in order:
First, no sane person is talking about using carbon tax revenues to fund the entire government. Current U.S. GHG emissions are about 7 billion tonnes of CO2 equivalent. At $10/tonne, that’s $70 billion; at $100/tonne, $700 billion. Considering current prices for the second phase of the EU Emission Trading Scheme are around $30/tonne, I suspect that any price the U.S. puts on emissions in the near term will be nearer $10 per tonne than $100. Current IRS gross collections: $2.5 trillion. I don’t think we need to be worried about whether carbon tax revenues are going to be $100 billion or $95 billion. The difference is spare change that Uncle Sam lost in the couch. Further, Dorman thinks that the fact that revenues will experience long-term decline (from falling emission) is a big problem. But of course in order to sustain those declines — the entire point of an emissions tax, presumably — the tax will have to continue to rise in real terms through time.
Second, certainty about emissions in the near term is not a big concern, as I pointed out earlier today. Beyond the technical point from that post, however, is the notion that it is silly to talk about what precise limit the U.S. should try to place on emissions while China and India don’t have any limits yet. The U.S. should have some policy to limit emissions, but the exact level of reduction is not environmentally important. Further, Dorman seems to imply that we are uncertain about the relationship between a carbon price and emissions — odd, since he seems much more certain when noting how much gas demand falls in response to increased oil prices — but unless Dorman doesn’t believe that demand curves slope downward then I can pretty well guarantee him that more taxes will mean less emissions.
Third, I completely agree that any political negotiation over carbon taxes will be plagued by special interests, giveaways, tax breaks, and other political concessions that will make the system less efficient than it might otherwise be. Anyone who doesn’t think this will happen with allowance allocation in a cap-and-trade system as well, however, needs to start taking their reality pills again.
Fourth, Dorman concludes that cap-and-trade will be more equitable because allowances — or the revenues from allowance auctions — can be distributed on a per capita basis. But this is not a distinction between taxes or a cap-and-trade with auctioned permits, because both raise revenues which can be distributed equitably. On the odds of this actually happening in either case, see point 3 above.
Finally, Dorman thinks that cap-and-trade — and cap-and-trade alone — will let us control the risk we face from “catastrophic climate change.” I will not fully address this here, as I have a long post I am planning to write on this issue. For now, it will suffice to say that our ability to control long-term environmental outcomes is probably much smaller than we imagine, particularly when it comes to catastrophic changes. Now, obviously it likely matters a lot whether we allow emissions to continue on a business-as-usual course — with atmospheric GHG concentrations topping 1,000 ppm CO2 equivalent before the end of the century — versus limiting emissions to stabilize concentrations around 500 – 650 ppm. But within that range we will likely be facing a similar set of risks for catastrophic outcomes, and, problematically, we won’t have any clear idea of how much we are reducing our risks by making marginal changes in the final outcome (i.e., stabilizing concentrations at 525 ppm versus 550 ppm).
The moral of the story is to start doing something now, but not worry about exactly how much to do. Cap-and-trade will fill the bill; so will a carbon tax.