Why Carbon Revenues Might Be More Important Than Carbon Prices

I recently had an interesting conversation on Twitter and via email that got to the heart of why I hold what many consider an unorthodox view on the role of a carbon price in climate policy.  

While most academic economists and conventional climate policy analysts hold that addressing climate change is first and foremost about putting "the right price" on carbon pollution via a carbon tax or emissions cap and trade program, I have long advocated a somewhat alternative view.  

Establishing a price on carbon is important and can greatly improve the economic efficiency of climate mitigation efforts. On this point, the economists are correct. 

But it is also important for climate policy makers to recognize the pervasive constraints placed on efforts to price carbon in any political economy the world over. In practice, these political economy constraints are binding long before we get close to what most economists consider "the right price" on carbon (and as we'll show below, there is considerable debate about what that "right price" is).

These political constraints should prompt policy makers, academic researchers (including energy economists!), and climate advocates to rethink the role of carbon pricing in climate mitigation. And given the failure of climate policy to date to drive the rapid decarbonization needed to confront climate change, it is long past time for some fresh thinking.

I hope recounting this conversation will shed light on my contention and prompt continued discussion here at TheEnergyCollective.com (TEC). This is a long essay, and I hope you'll stick with me. Feel free to start with the summary at the end of this post and then return for the full argument... 

Technological change at the heart of climate mitigation efforts

My interlocutors last week included Carleton University doctoral candidate Suzy Waldman, energy consultant and columnist Eli Hinckley, and carbon footprint and data visualization guru Lindsay Wilson.

Our conversation began when Suzy tweeted the following:

Suzy was quoting from a letter written by CU-Boulder political scientist Roger Pielke, Jr. in response to a Financial Times editorial urging China to adopt a carbon cap and trade program. Here is an excerpted version of Roger's letter:

Carbon emissions are the product of (a) GDP growth and (b) technologies of energy consumption and production. ... Thus, a “carbon cap” actually means that a government is committing to either a cessation of economic growth or to the systematic advancement of technological innovation in energy systems on a predictable schedule, such that economic growth is not constrained. Because halting economic growth is not an option, in China or anywhere else, and technological innovation does not occur via fiat, there is in practice no such thing as a “carbon cap.”

Where carbon caps have been attempted, clever legislators have used gimmicks such as carbon offsets or set caps unrealistically high so that negative effects on GDP do not result and the unpredictability of energy innovation does not become an issue.

It should thus not come as a surprise that carbon caps have not led to emissions reductions or even limitations anywhere. China will be no different. The sooner that we realize that advances in technology are what will reduce emissions, not arbitrary targets and timetables for reductions, the sooner we can focus our attention on the serious business of energy innovation.

That global climate mitigation at its core is about sparking a technological transformation in the way the world makes and uses energy should be non-controversial. 

What drives innovation?

Eli, always incisive, then cut to the core question: 

That's when I chimed into the conversation:

I then referenced three excellent books (and one report edited by myself) that each drive home this key bit of historical evidence:

Indeed, from railroads and interchangeable parts to nuclear, solar, and wind power and from microchips and the Internet to blockbuster drugs and agricultural innovations, the "invisible hand of government" is often just as important as the invisible hand of the market in driving game-changing innovation throughout American history.

(Note: I make this case at greater length in a video and presentation here.)

While getting price signals right can be an important spur for the adoption of available and cost-competitive technologies, looking back at the history of American innovation tells us that something besides price signals has been responsible for the kind of dramatic innovation and technological transformation needed to address climate change: a proactive public sector, most-often acting as both direct investor in early-stage research and as demanding customer of emerging technologies. 

Investing in innovation in an Age of Austerity

While granting my point, Eli once again cut straight to the heart of the matter: 

Indeed, in an era of budget cuts, deficits, and declining public investments in energy innovation, it is certainly fair to ask: can a public-sector led effort to catalyze energy innovation and advanced technology adoption really succeed?

I'll let the Twitter conversation play out for a minute here...

We'll come back to the solution that problem in a moment...

How binding are political constraints on carbon pricing?

First, I must note that I didn't choose 10 cents per gallon in that tweet above arbitrarily.

It would take a carbon price of about $10 per metric ton of CO2 to increase the price of gasoline at the pump by about 10 cents per gallon.

The same carbon price would increase the price of electricity produced by a coal-fired power plant by about $10 per megawatt-hour (MWh), or about 1 cent per kilowatt-hour (kWh), in round terms. The price of electricity from a natural gas-fired power plant would rise about $5 per MWh or half a cent per kWh. For comparison, average wholesale electricity prices in the United States range from about $30 to $50 per MWh and retail prices (the prices you and I pay on our monthly electricity bill) range from about 8 to 16 cents per kWh in the contiguous U.S. states.

It should be pretty clear that an increase of 10 cents per gallon of gasoline at the pump is unlikely to drive dramatic changes in consumer behavior or spur corporate investments in advanced vehicle technologies. Nor will a $5 to $10 per MWh increase in the cost of fossil fueled electricity transform the electric power sector. The value of the production tax credit (PTC) responsible for the growth of wind and geothermal energy in the United States, the most cost-competitive non-hydro renewable energy technologies available today, is now worth $23 per MWh for comparison.

Despite the modest impacts of a $10 per ton carbon price, given political opposition to raising energy prices here in the United States and indeed across the globe, a carbon price of that order of magnitude may be the best we can hope for.

Political scientists and observers of climate policy debates have long observed that while citizens of various nations express varying degrees of concern about climate change and a willingness to pay to reduce the risks of unchecked global warming, that willingness has its limits. While that limit may vary from nation to nation (and at different times within a given nation), these political economy constraints are an ever-present feature of climate policy debates the world-over. These constraints are so pervasive, Roger Pielke Jr. dubs their impact the "Iron Law of Climate Policy," in his book The Climate Fix

So how binding is this Iron Law?

Evan Johnson and Greg Nemet of the University of Wisconsin-Madison surveyed 27 studies estimating citizens' willingness to pay (WTP) for climate mitigation, measured as additional annual costs per household. Despite a diversity of methods in the underlying studies, after excluding outliers, Johnson and Nemet find that a majority of studies estimate a WTP that falls within the range of $51-$470 in household cost per year (in constant 2012 dollars), with a mean value of $180 and median of $145. (How much would you be willing to pay?)

In the midst of the national debate over climate change legislation in the United States in 2010, Ana Villar and John Krosnik also surveyed voters’ preferences in three politically diverse U.S. states: Florida, Maine, and Massachusetts. Their results perfectly illustrate the Iron Law in action: while majorities of voters in all three states supported mitigation policies if the policy cost less than $100 in extra taxes per household per year, support dropped by 9-10 percentage points to a slim majority in each state as this expected cost increased to $150 per household. Support fell further as the estimated cost increased to $200 per household, dropping below majority levels in Florida and Maine, while retaining a sim majority in Massachusetts (see table below).

Willingness to Pay for Climate Mitigation in Florida, Maine, and Massachusetts
(Villar and Krosnik, 2010)

Willingness to Pay for Climate Mitigation in Florida, Maine, and Massachusetts (Villar and Krosnik, 2010) 

Villar and Krosnik’s results are broadly consistent with the mean and median WTP estimates from Johnson and Nemet’s literature survey.

Both studies indicate that a majority of citizens are likely to oppose climate mitigation policies with expected annual household costs on the order of $150-200 per household.

So what does that mean for carbon pricing efforts?

Annual CO2 emissions per household vary in the United States depending on household wealth and location. For a forthcoming paper, I've estimated that average annual household CO2 emissions in the United States range from around 14 to 67 24 to 44 metric tons across income deciles and 26 to 39 25 to 48 metric tons for average households across each state and the District of Columbia. The mean U.S. household emits about 31 34 metric tons of CO2 per year. [Note: state-by-state household emissions figures updated July 31st to reflect revision in paper methodology to reflect varying average household sizes in each state. Related values updated throughout post with original values noted with strike-through.]

Assuming that citizens' willingness to pay is exhausted after an additional cost of about $150-200 per household per year, this implies that the Iron Law begins to bind and political opposition from citizens will mount quickly for policies imposing carbon taxes as they move upwards in the range of $2 to $15 $3 to $8 per metric ton of CO2.

After considering the Iron Law of Climate Policy, pushing for a $10 per ton carbon price in the United States may be the best we can hope for. A modest $5 per ton price may be more realistic.

Getting the right price on carbon, or the price we can get?

If the political economy constraints bind the feasible range of carbon prices in the neighborhood of $5 to $15 below $10 per ton, what are the implications for the efficiacy and efficiency of carbon pricing efforts?

If dumping CO2 into the atmosphere is free, then when businesses and consumers buy products and services that emit CO2, they don't have to pay any price associated with the climate-related damages caused by carbon emissions. These costs are thus said to be "external" to the market transactions, and that leads to distortions in our economic behavior. 

The aim of carbon pricing policies is to force businesses and consumers to "internalize" the costs of these climate-related damages by pricing them into the cost of a gallon of fuel or a ton of coal etc. That forces economic actors to consider the full costs of climate pollution when making economic decisions, and brings the economy back to its optimal efficiency. 

With the right carbon price in place, consumers and businesses would adjust their behavior accordingly, and they would only purchase goods or services that result in CO2 emissions when the economic benefits of those goods or services exceeds their full cost including the associated climate damages. Consumers would also have new incentives to pursue behaviors that emit less CO2 yet deliver the same benefits and businesses and entrepreneurs would be rewarded for inventing and providing such alternatives.

Economists are so faithful in the power of prices that many argue that if we can fully internalize the cost of carbon emissions, that would be pretty much all we'd need to do to confront climate change. For an archtypal example of this argument, listen to the NPR Planet Money team's recent chat with MIT economists Henry Jacoby and John Rielly as they outline their "one page plan to fix global warming." 

Yet for this kind of carbon pricing policy to work as designed, the price imposed by the carbon tax must exactly equal the estimated economic damages imposed by a ton of CO2 emissions, or the so-called "social cost of carbon."

Estimating this cost is notoriously tricky. It depends on estimating both the expected damages caused by climate change over time, as well as the contribution of a ton of CO2 emissions to those damages, both of which involve inherent uncertainties.

Additionally, as climate damages will occur over long periods of time -- centuries even -- and costs must be paid up front in the form of a carbon price, those damages must be discounted to arrive at a present value. This introduces another factor that can lead to dramatically different estimates of the cost of carbon: the social discount rate used to equate damages far off in the future with the price we should pay today. The rate we choose is essentially an answer to the question, How much should we value the welfare of future lives relative to those living today? Your choice of discount rate is thus ultimately a matter of irresolvable philosophical and ethical preference.

For my forthcoming paper, I surveyed a range of widely-cited estimates of the social cost of carbon. The graph below illustrate how the estimates differ depending on their choice of discount rate. For a mid-range value, the Obama Administration's latest estimate of the social cost of carbon using a 3 percent discount rate is $36.54 per metric ton of CO2 in 2010, rising to $78.62 by 2050 (in 2012 real dollars). 

Estimates of the Social Cost of Carbon (from Jesse Jenkins, forthcoming)
Estimates of the Social Cost of Carbon (from Jesse Jenkins, forthcoming)
Values in 2012 US$ per ton CO2

While estimates of the social cost of carbon range by an order of magnitude, one thing is clear: it is all but certain that the external damages caused by CO2 emissions oustrip citizens' willingness to pay for climate mitigation. 

While the Iron Law likely binds carbon prices in the U.S. to somewhere below $10 per ton CO2, estimates of the social cost of carbon range from $10-100 per ton CO2 today, and prices are supposed to rise steadily over time, to as high as $50-800 per ton by 2050.

For me, the implication is unshakeable: when policy makers try to implement a price on carbon via a carbon tax or cap and trade program, we don't end up with "the right price" on carbon that economists model when they argue for the efficiency of carbon pricing.

Either policy makers ambitions outstrip citizens' willingness to pay and policy proposals fail (as one could easily argue was the case for the 2009-2010 effort to enact a federal cap and trade policy in the United States), or policy makers work to make the carbon price comform with the Iron Law and the policy has a chance of succeeding. 

In short: when we try to price carbon, we don't get the right price on carbon, we get the price we can get.

When carbon pricing policies meet political constraints: implications for economic efficiency and environmental efficacy

When it comes to the economic efficiency and environmental efficacy of carbon pricing policies, the implications of a politically-constrained carbon price are pretty clear. 

A carbon price only eliminates the external damages associated with CO2 emissions when the price exactly equals the social cost of carbon. If the carbon tax falls well below this optimal level, much of the social cost of CO2 emissions will remain "externalized" or unpriced in market transactions. As a result, market distortions will persists and the economy will continue to emit far too much CO2.

The graphic below illustrates this dynamic. The optimal carbon tax T* below is sufficient to equate the marginal private cost (MPC) of a given good or service that emits CO2 with the marginal social cost (MSC) of that good or service, including the full climate-related social costs. The marginal cost of the good or service rises to the red dashed line. Looking where that line intersects with the marginal benefit of consuming that good or service, in light blue, and we see that the optimal tax will reduce consumption of that good from Q to Q*, fully eliminating the climate-related social costs. This is the textbook picture of carbon pricing policies. 

Yet if what we end up with is the constrained carbon tax TC, the price of the good or service will rise only to the blue line, far short of the full marginal social cost. As such, consumption of the good or service will only fall to QC, and as a result, a considerable amount of climate-related social costs remain (the shaded red area known to economists as a "deadweight loss").

Political Constrained Carbon Price and Remaining Climate-related Social Costs

Political Constrained Carbon Price and Remaining Climate-related Social Costs
(from Jesse Jenkins, forthcoming)

That's a rather long-winded way of saying: a politically-constrained carbon price won't work as advertised. If it falls short of the true social cost of carbon, it will fail to effectively reduce CO2 emissions or mitigate climate change.

Can carbon revenues succeed where carbon pricing falls short?

I hope by now you are beginning to understand why I urge climate policy makers and advocates to think creatively about the role of carbon pricing.

While it would be wonderful to snap our fingers and establish the perfect price on carbon, there's no forseeable way that's going to happen. If what most energy and environmental economists consider Plan A is off the table, then perhaps it's (long past?) time to consider some possible Plan Bs.

That brings us back to my Twitter conversation. As I urged Eli and Suzy, perhaps its time to start thinking not just about the price signals we can send by taxing carbon, but also about the revenues we can raise.

As I said earlier, a $10 per ton tax on carbon is unlikely to spur anything close to the kind of deep decarbonization of the economy needed to address climate risks. At the same time, it could raise quite a bit of money! As I put it on Twitter:

The United States currently emits about 5 billion metric tons of CO2 annually. Even a modest carbon price of $5 to $15 $10 per ton would therefore raise on the order of $25 to $75 $50 billion annually in new federal revenues. That would be more than enough to fund the scale of proactive public investment responsible for the blockbuster American innovations discussed previously.

In short, the carbon revenues may be more than enough to make up for the short-fall of a politically-constrained carbon price and get the climate job done.

In fact, generating much-needed revenues to fund a well-structured and well-funded effort to catalyze advanced energy-innovation may rightly be considered the primary purpose of a carbon pricing policy. The added economic efficiency and mitigation benefits that come with internalizing some portion of the climate-related social cost of carbon is an added benefit, as I argued in the following exchange...

Keep in mind that when I talk about "innovation investment," I am by no means referring solely to expanded R&D funding.

While a wide variety of energy and climate experts, from the President's Council of Advisors on Science and Technology to the industry titans behind the American Energy Innovation Council all advocate roughly tripling federal energy R&D funding over time, a goal I heartily support, that is only half the historic recipe for unlocking American innovation. The government must also help create demanding markets for emerging advanced energy technologies with well-structured incentives that drive market demand and continual innovation and improvement.

(For more on my vision for a suite of interconnected advanced energy innovation policies, see my May 22, 2012 testimony before the United States Senate Energy and Natural Resources Committee.)

Rethinking the carbon pricing "double dividend" 

At the end of the day, Eli, Suzy and I all agreed that it would obviously be ideal to establish a carbon price that fully internalized the social cost of carbon. But we also all recognized that such a policy was not forthcoming.

In the face of political economy constraints on carbon pricing -- the shackles of Pielke's "Iron Law" -- it is time for policy makers, academic researchers, and climate advocates to get creative about re-envisioning the role of carbon pricing in climate mitigation efforts.

Perhaps we will start viewing the revenues potentially raised by carbon pricing policies as just as important as the price signal they establish and think proactively about how best to structure the kinds of public investments that could make up for the short-falls of a constrained carbon price.

As I noted on Twitter, perhaps we may even come to view the use of carbon revenues for these purposes as an important "double dividend" in return for carbon pricing efforts.

After viewing our Twitter conversation, Lindsay Wilson extended the discussion via email. Responding to my use of the "double dividend" term to describe using carbon revenues to fund further energy innovation and climate mitigation efforts, Lindsay fairly noted that my use of the term was "pretty much completely backwards from the use of the term 'double dividend' in the economics literature."

As Lindsay explains, the economist's argument is as follows:

"By earmarking revenue for energy purposes you increase economic distortions, reducing the efficiency benefits of the environmental tax. That is the major reason most economists prefer a 'revenue neutral' carbon tax that recycles revenues by offsetting other distortionary taxes, like payroll or income taxes. You can then make a strong argument that such a carbon tax would be pro-growth, as the benefits of reducing income tax distortions are likely to exceed the tax interaction effect of the carbon tax."

I replied to Lindsay by explaining that my use of the term "double dividend" in this context was a conscious effort to co-opt to conventional use of the term.

The concept of a double dividend for carbon taxation is that you want to use the revenues of the tax to offset existing economic distortions (thus earning the double dividend) rather than to introduce new distortions into the market by say subsidizing your favorite economic sectors or constituents.
For an economist who (a) assumes we know the exact right price for carbon such to perfectly equate marginal private costs and marginal social costs and (b) who thinks we will fully internalized said price via a carbon tax/cap, their favored use of the tax revenues is to offset existing taxes that introduce distortions into the economy, e.g. taxes on labor, capital, etc. 
That said, IF you think that political constraints prevent pretty much any political economy from fully internalizing the environmental externalities associated with carbon pollution -- and I think the real world track record and evidence presented above supports that theory pretty well -- THEN the constrained carbon price we can actually get will not fully internalize the climate externality (as discussed above). Substantial external costs will remain due to fossil fuel combustion and CO2 emissions. These external costs are distortionary on the economy as well (e.g. cause deadweight loss). 
Thus, if political economy constraints bind carbon prices well below the optimal level, there may be another double dividend out there to earn. If you want to further improve the economic efficiency of the constrained carbon price (earning a double dividend) AND you want to increase the environmental efficacy of the policy (further reducing CO2 emissions), then I'd say you have two very legitimate uses for the revenues:
  1. Subsidize low/zero-carbon energy technologies that are currently too expensive to be deployed under the politically constrained carbon price; and

  2. Fund well-structured R&D, demonstration, and deployment policies that drive down the real, unsubsidized costs of low/zero-carbon energy technologies such that they can be deployed at the politically constrained carbon price.
Both efforts help reduce the remaining climate-related externality, improving the efficiency of the economy. Note also that most economists would agree that there are remaining distortionary market failures all throughout the innovation process, so the second use of funds may even earn you a triple dividend, helping address those innovation-related market failures at the same time.

Like I said: time to think creatively about the role of carbon pricing!
Why carbon revenues may be more important than carbon prices

In summary:
  • The history of American innovation points to the critical role of public (e.g. government) investment in advanced technologies, working both as direct investor in cutting-edge research and as demanding early customer for emerging technologies, from interchangeable parts and railroads to microchips, jet engines, nuclear, solar and wind power, and the Internet.

  • Unlocking the kind of advanced energy innovation needed to tackle climate change will almost certainly require a similar role for the government. Yet the current Age of Austerity (and, to be fair, a resurgance of anti-government conservativism) constrains the ability of government to perform this critical role, undermining climate mitigation efforts.

  • Most economists argue that putting the right price on carbon, through either a carbon tax or cap and trade program, can succeed at reducing CO2 emissions and mitigating climate risks without any direct government intervention. To work as advertised, however, the carbon price must internalize the full social cost of carbon emissions. 

  • Citizens are willing to pay to reduce climate risks, but that willingness is limited, leading to pervasive political economy constraints on carbon pricing efforts (what Roger Pielke Jr. calls the Iron Law of Climate Policy). In the U.S., the Iron Law may bind a politically viable carbon price to as low as $5-15 at or below $10 per ton of CO2. In contrast, estimates of the social cost of carbon range from $10 to $100 per ton CO2 today rising steadily over time to as high as $50-800 per ton by 2050.

  • A politically-constrained carbon price won't work as advertised. A $5-15 $5 to $10 per ton carbon price will fall far short of internalizing the full social cost of carbon, failing to sufficiently reduce CO2 emissions or mitigate climate change.

  • At the same time, establishing such a price could raise between $25 and $75 $50 billion annually in new carbon revenues. That would be more than enough to overcome the limits of the Age of Austery and fund the scale of proactive public investment responsible for the blockbuster American innovations discussed previously.

  • If political economy constraints bind carbon prices well below the optimal level, policy makers may be able to earn an economic and environmental "double dividend" by putting carbon revenues to use to reduce the deadweight losses associated with the remaining, un-internalized social cost of carbon. There are two very legitimate uses of these revenues: 1) Subsidizing low/zero-carbon energy technologies that are currently too expensive to be deployed under the politically constrained carbon price; and 2) Funding well-structured R&D, demonstration, and deployment policies that drive down the real, unsubsidized costs of low/zero-carbon energy technologies such that they can be deployed at the politically constrained carbon price.

  • At the end of the day, thinking creatively about how to design effective climate policies despite pervasive political economy constraints on carbon pricing may require us to view carbon revenue as just as important, if not more so, than the price signal established by a politically constrained carbon price.

Thanks for reading, and please share your thoughts in the comments below.