Carbon taxes and emissions trading are two ends of a spectrum of possibilities for carbon pricing. One or other end of the spectrum may not be the best place to be.
As carbon pricing spreads around the world (see post from 7th March) governments are differing in whether they choose to pursue emissions trading or carbon taxes. For instance, South Korea is getting ready to implement its emissions trading scheme, for which legislation was passed last year, while the South African Government has recently reaffirmed its commitment to a carbon tax. There many reasons for these differences in approach, which I’ll return to in future posts, but for the moment I want to look at how existing schemes illustrate that the distinction between the two types of instrument is not absolute. Emissions trading schemes can have fixed price components or limits to price ranges, which introduce an element of price certainty, while taxes may allow a degree of trading in the form of offsets or credits generated from outperforming a baseline. Emissions trading and carbon taxes are thus at two ends of a spectrum of possibilities. The best policy choice is not necessarily at either end.
The chart shows major carbon pricing schemes around the world on a spectrum from pure emissions trading on the left to pure carbon taxes on the right. The first thing that’s apparent is that there is a clear preponderance of emissions trading schemes, especially by volume of emissions covered. But many ETSs contain some element of price certainty, and taxes may include elements of trading, so many schemes do indeed in practice lie somewhere between the two extremes.
A spectrum of possibilities between an emissions trading schemes (ETS) and a carbon tax
The EUETS and the Regional Greenhouse Gas Initiative (RGGI) in the Northeastern USA are almost pure emissions trading scheme, with minimal limits on price, although RGGI does have a low reserve price in its auctions (currently $1.98, indexed to inflation), and there are also some very limited auction reserve provisions available in the EUETS. The prospective Beijing ETS appears to likely to follow this model quite closely, although there appears to be a possibility that the regulator will seek to exercise some influence on the price by buying and selling allowances centrally, rather analogous to the role a central bank might play in currency markets, or by some other means. However, the Beijing scheme remains to be finalised and it is not yet clear what form this or the other trial Chinese schemes will take in practice.
Several other schemes diverge from the model of a pure ETS by having elements of fixed pricing. The Australian scheme includes an initial 3-year fixed price period. Before the link to the EU ETS was established late last year it was intended that after the 3-year fixed price period it would include a floor price in the auction and a ceiling related to the international carbon price. (The price ceiling remains in principle following the EU linkage, but in practice will never be triggered). This made its original design much closer to the California and Quebec schemes, both of which contain reserve prices in the allowance auctions to give a price floor, and also price ceilings in the form of reserve tranches of allowances. Floor prices are currently just over $10/tonne, well above the levels found in RGGI, and indexed to rise at above the rate of inflation.
The Alberta scheme is a hybrid between a tax and an ETS. It sets a baseline for emissions from an entity per unit of output (emissions intensity), rather than a fixed emissions cap. If emissions are below this baseline then credits are generated. If emissions are above the baseline then the entity has several options: it can buy credits from those in surplus, it can purchase offsets, or it can pay a fixed price of $15/tCO2 which goes into a fund for clean technology investment. It thus resembles a carbon tax payable above a threshold level of emissions intensity, but with some element of trading allowed.
The proposed South African carbon tax would, if introduced in its currently proposed form, allow some of any remaining obligation to be met through offsets.
At the other end of the spectrum the British Columbia carbon tax is a solely a tax, with no element of trading at present (and only very minimal provisions for offsets).
Political processes introduce dynamics which may ultimately limit the extent to which a carbon pricing scheme focus is exclusively on price and quantity. Decisions about the level of the cap are likely to be influenced by expectations of the prices that will result. And if a surplus of allowances leads to low prices a tightening of the cap may be considered, as in both the EUETS and RGGI at present. Similarly the level of a tax may be adjusted over time to meet an environmental goal, or other policy objectives such as raising revenue.
So policy debates should perhaps be less about a preference for taxes or emissions trading in their pure form, and more about where a scheme should be positioned on the spectrum of possibilities, and how outcomes can be managed over time as the scheme develops. And even if a scheme starts at one end of the spectrum, events may force changes that recognise that in practice there is always a balance to be struck between prices and quantities.
Adam Whitmore 27th March 2013