This is why a successful policy against GHG emissions will required three parts; I sometimes think of it as comfortable three legged stool that will support our goals:

Public investment - Public investment provides alternatives for people to respond to price signials with.

Rule based regulation: In the absence of regulations, very large price increases tend to produce very small demand reductions. But many of our consumption sectors really are easy to measure. In Denmark and Norway there are building regulations that regulate maximum consumption for climate control per square foot. In transportation you can measure emissions per passanger mile or per ton mile. In home appliances and office equipment most of the energy consumption is by a  handful of types of appliances and equipment: again fairl straightforward to do energystar type regulations.

An emissions tax in the context of such public investment and regulation can be very effective; also it is neccesary for control industrial emissions because of the variety of sources in that sector are not really amenable to either rule  based regulation or public investment. Not that there are not a few technologies within that sector responsible for a bit emissions chunk. Regulation and subsidies can play a signficant role here, but not as a substitute for emisisons pricing.

In terms of funding public investment - I suggest taking that from military spending rather than depending on emissions taxes. You don't want clean energy technology dependent on dirty power for subsidies. I like Peter Barnes' suggestion of rebating revenues to the public as the Alaska Pipeline does.   That compensates perfectly for the regressive effects of energy taxes on not only the poor but the middle class. No program depends on that revenue, and the rebates fade out as the things they compensate for do - collection of energy taxes. If people start fussing about their rebate checks shrinking, we can extend the taxes to other pollutants.

The key insight of Dr.Monica Prasad is the power of transfering payments from CO2 emissions to support the cleaner energy production.  She says Denmark's carbon tax worked better than other countries' carbon taxes because most of the tax revenue went support cleaner energy production approaches.

The better way is an output emission allowance standard, as follows.  Use delivered electricity or useful thermal energy as the metric.  Give every producer of heat or power the same allowance for every unit they deliver.  Require every producer to obtain and offset every ton of CO2 they produce with an allowance.  Make the allowances fully tradeable and exchangeable between heat and power.  Fine any producer who does not cover CO2 emissions with allowances 150% of the average cost of the CO2 emissions over the prior year.  True up annually.

Add two features.  Publish a schedule that reduces the allowance per unit of electricity and of heat each year to achieve a society wide drop of CO2 emissions.  Then adjust the scheduled alowance in future years for growth (or shrinkage) of economy-wide fossil CO2 emissions.  For example, to adjust the scheduled allowance of CO2 per MWh in 2012, multiply the original scheduled allowance by a fraction which has the 2008 base year economy-wide fossil CO2 emissions divided by the 2011 CO2 emissions.  This insures that the total allowances shrink as designed, withou regard to growth of fossil fuel use or economic growth.

Why is an output allowance system better?
The market sets the allowance price for CO2 and constantly adjusts.  There is no way even the wisest philosopher queen could ever set the right tax rate to accomplish desired CO2 reduction.  Pols set the desired reduction and the market discovers the price per ton of CO2.
There are no frictional losses.  Every dollar paid by the dirty electricity and thermal producers goes to the cleaner producers, avoiding the tempetation for politicians to divert CO2 revenues to other purposes.
Selling excess allowances immediately benefits clean energy projects and helps encourage deployment.  Although CO2 tax payments may discourage dirty energy production, the tax does not directly encourage clean energy production.
The output allowance system imposes no direct costs on the economy.  All payments from a dirty energy producers go to clean energy producers.  Either approach will redistribute wealth, but the allowance system transfers the dirty energy payment directly to clean energy projects.
The dynamic effect will reduce delivered energy costs.  Entrepreneurs will lay awake nights trying to deploy projects that profitably increase delivered fossil efficiency.
The output system lets the market select the technology of cleaner energy instead of a government such as Denmark selecting the favored technology.
Most importantantly, output is the correct metric for CO2 control.  The economy gains no direct benefit from burning fossil fuel, but that burning causes pollution.  Useful energy services have no direct impact on the environment, except to the extent their production caused pollution. The economy depends on the availability and cost of useful energy services.  Heat and power are the closest thing to useful energy services that are measured. (No one measures the light in the room, which is the truly useful 2% to 4% of the fuel energy that was spent to genrate and deliver electricity and then convert that power to light).  Success of any system depends upon measuring and rewarding the right metric. Output is the right metric to induce profitable greenhouse gas reductions.

Problems:
We have not thought of a way to use output allowances for transportation, since output is not recorded.  Government could set a tax on transportation fuel equal to the 'market discovered price' of a ton of CO2 from heat and power.  Be aware that 69% of US CO2 emissions are from production of heat and power (42% form power, 27% from heat).  Then use the revenues to promote cleaner transportation, perhaps funding mass transit.

Ditto the 17% or so of GHG from non-CO2.  Perhaps tax the same way as transportation.

Circuit Breaker? The government could have unlimited allowances to sell at a starting price deemed acceptable by the pols, say $10 per ton of CO2.  Then increase the price by 10% per year comppound or higher value.  This eases the transition problems from inefficient dirty production to cleaner and more efficient production and soon drives the price of government allowances out of reach.

Tom Casten
Chair, Recycled Energy Development

Tom Casten, Chair, Recycled Energy Development LLC