What carbon price would you like?

Client: "What's 1 + 1?"

Accountant: "What answer would you like?"

The joke could equally be about jobbing economists. Political economy was originally about challenging rulers' bad ideas, such as mercantilism. But over the twentieth century, mainstream economics evolved into a tool for governments. Economists increasingly calculated what should be discovered, and refined (rather than challenged) governments' mechanistic view of the economy.

Take carbon pricing. The essence of carbon pricing is to internalise the climate externality in economic decisions. If the solutions to climate change are determined by government, we don't need a carbon price. They can just pick their winners. The carbon price enables the decentralised coordination of the market to take the externality into account.

The climate doesn't differentiate between greenhouse-gas sources. A tonne of carbon-dioxide-equivalent (CO₂e) is a tonne of CO₂e, wherever it comes from (or however it is absorbed). The social cost of each tonne of CO₂e is the same, wherever it comes from.

A carbon price supposedly reflects that cost. When a government combines targeted support (whether direct or indirect) for certain technologies with carbon pricing (explicit or shadow), it effectively says that some CO₂e is more equal than others.

Let's say we have a carbon price of £50/tCO₂e. This equates to around:

But we introduce mechanisms that:

1. provide different levels of support for low-carbon electricity generation:
   • depending on technology and scale,
   • regardless of the emissions that will actually be displaced,
   • mostly way more than the carbon value based on grid-average emissions, but
   • some of it less than the carbon value on the same basis;
2. insulate some low-carbon generation technologies from the varying half-hourly values of their electricity in terms of:
   • wholesale value (supply/demand), and
   • the carbon intensity at the time of generation,
   so every MWh is guaranteed a price above the wholesale average, regardless of whether:
   • there is any demand for it, or
   • it displaces any carbon;
3. insulate some low-carbon electricity technologies (e.g. wind in northern Scotland) from some of the cost of distributing their power to customers;
4. insulate intermittent electricity technologies from the effect of their intermittency on their wholesale price, by instituting a capacity market whose cost is passed to customers;
5. time-limit support mechanisms (without even a residual carbon value, because excessive value is provided in many cases during the support period), so depreciated low-carbon assets that could offer cheap low-carbon energy are disadvantaged relative to new projects, and apply different time periods depending on technology, scale, end-use (electricity, heat, transport, etc);
6. provide tax incentives to encourage people to drive diesel rather than petrol cars;
7. use grants, taxes and congestion charges to encourage the purchase of electric vehicles regardless of the source of the electricity and how much petrol or diesel they will avoid (and then remove some of those incentives without replacing them with something more rational);
8. tax all road vehicles at a level well beyond the cost of (i) carbon and (ii) road spending, while giving other forms of travel tax exemptions (air) and/or subsidies (rail);
9. use cap-and-trade to create an alternative carbon price for some carbon sources (industry and electricity generation) but not others (heating and vehicle fuels), with a scheme designed to (i) generate an inadequate carbon price, but (ii) apply that price to native production but not to foreign competitors whose cheap embodied carbon undercuts our native producers;
10. ignore low-carbon heat for two decades while other sectors are being richly subsidised, then
    • institute a subsidy that provides radically different levels of support depending on sector (domestic or non-domestic), technology and scale, then
    • adjust those levels of support depending whether the technologies are delivering what the government thinks they should deliver, then
    • replace it with a sparse set of support mechanisms that provide different levels of support for certain of the favoured technologies ("green gas", and heat pumps, with a supporting role for biomass), while
    • hanging hopes of long-term heat decarbonisation on technologies that are unproven and whose viability is unknown, with no mechanism to relate its cost to its carbon value, and with mass roll-out kicked into the long grass again.
11. provide no material value for sequestration, other than a failed prize scheme;
12. etc. (CCL)
13. etc. (CRC)
14. etc. (EEC/CERT/CESP/ECO)
15. etc.................................................

What signal are we getting from the carbon price in this environment? It is barely a yardstick. It is not a realisable value for most projects. The values available are complex but bear almost no relation to the carbon value. Investments are made on the basis of projects' real expected costs and values. The impacts are experienced on the same basis.

Imagine (a) the combination of the above provides a value that enables a project to proceed, and (b) the project would not have gone ahead if there was a simple carbon price instead of all these mechanisms. The project is proceeding even though its cost to society exceeds its benefit to society. This is a net loss to society, aka a reduction in welfare.

Now imagine the opposite: (a) a project cannot proceed on the strength of the mechanisms actually in operation, but (b) would have gone ahead on the strength of a carbon price. A project is prevented whose benefits to society would have exceeded its costs to society. This is also a net loss to society.

The only way we can be sure that all/only those projects proceed whose social benefit exceeds their social costs, is if the net value of the various interventions available to any project with climate ramifications is a carbon price equal to the social cost. The only practical way to achieve that is to set a carbon price and wind-down other climate mechanisms.

Bureaucrats can do all the Impact Assessments they like, putting arbitrary numbers on the expected (i) costs, (ii) benefits including a (shadow) carbon price and (iii) scale of delivery, to justify the implementation or rejection of each mechanism. The fact that the carbon price was incorporated in each IA does not mean that the combination of mechanisms provides a rational internalisation of the externality, even in the unlikely event that the assumptions built into the IA are borne out by experience.

The use of a carbon price in this context (shadow pricing in Impact Assessments) is highly misleading. The carbon price's purpose is to internalise the externality in actual economic decisions, not to be a component in the justification for alternative mechanisms, which generate a complex matrix of technology-, scale-, and sector-dependent impacts on market costs and values.

Economists know this. That is why over 3,500 of them, including 27 Nobel Laureates, have backed the Carbon Dividend proposal. Normally, if you put three economists in a room, you get four opinions. This must be the most widely-backed proposal in the history of economics, because it is widely-recognised as the technically-correct, least-bad way to address climate change.

And yet when 34 governments or public-sector bodies, 200 corporations and 80 colleges and pressure groups commissioned a study on carbon pricing, they managed to find two economists who had not endorsed a Carbon Dividend, and appear not to accept the basic economics underlying it. And not just any economists: Joseph Stiglitz (a Nobel Laureate) and Nicholas Stern (Lord Stern of Brentford, author of the British government's 2006 Stern Review on the Economics of Climate Change).

In his earlier report, Stern had made the classic economic arguments about a consistent carbon price, and attempted to calculate the carbon price in the way most consistent with the economic principles: estimating the Net Present Value (NPV) of the future harm. Unfortunately, the report received mixed reviews in economic circles. Many were critical of Stern's use of a very-low discount rate to generate a relatively-high NPV. The recent award of the Nobel Prize to Bill Nordhaus could be seen as a victory for Stern's opponents. Nordhaus was one of Stern's fiercest critics and advocates a more traditional discount rate (and therefore a lower carbon price).

This may partly explain why the Stiglitz and Stern report adopts a radically-different approach to pricing carbon. It attempts to estimate not the social cost of the risk of harm from Anthropogenic Global Warming, but the level of support needed to deliver the technologies most suitable (in the opinions of some experts and interest groups) to achieve the level of emissions thought to be consistent with avoiding runaway climate change.

There are several problems with this, including:

1. It assumes a level of knowledge and foresight that is not available;
2. It discards the most fundamental aspect of the underlying welfare economics: current and future costs are a trade-off, so it makes no sense to fix rather than discover a trajectory for emissions as though we know where that unpredictable trade-off balances out.
3. Even if (1) and (2) were not a problem, the objective would still be for all sources and sinks of carbon to receive the same value in the round (i.e. carbon price plus any other mechanisms) for the social cost of carbon, so that the most efficient solutions were encouraged. Yet S&S's approach was to pick certain technology "winners" (not even covering all uses of energy, but straight-line extrapolating from the easy sectors they did cover to the difficult sectors they did not), assume various government interventions to reduce their cost of delivery, and then estimate what was needed on top to make them happen, and call that the "carbon price".

It is (3) that really rankles, and that this post was written to address. By all means, argue that (a) minimal or (b) drastic action is required. You are effectively arguing for a carbon price that is (a) low (even zero) or (b) high. You'll need to back that up with how you reach the conclusion that in the main (a) adaptation is better value than mitigation, or (b) vice versa, but at least we're having a rational debate. But don't argue that some solutions should be valued differently to others. It's just rent-seeking.

If you do it as a business (say 200 corporations and their representative bodies sponsoring a study), you are simply pursuing your self-interest rationally. It's why governments should ignore businesses that claim strategic benefits worthy of government support for solutions that they just happen to be invested in.

If you do it as a government (say 34 public bodies commissioning a study), you are predictably falling for the Fatal Conceit, and responding to your public-choice incentives. Voters want to be told you have the answers, and you want to believe that you do. You are wrong and harmful, but anyone who understands government should expect no better.

But if you do it as a senior economist, you deserve nothing but contempt. You know it's wrong and damaging, and you don't need to pander to the interest groups. Why do it? I am struggling to think of any reason other than you were given the gig because your answer to the question "can you estimate a carbon price?" was "what answer would you like?"

Postscript

Here's an example of the effect of this cynical, irrational methodology.

There has been a persistent problem with developed nations offshoring carbon through the application of social costs to their own industry that developing nations do not apply to theirs. There is no environmental benefit to this. In fact, it is probably net negative because the production facilities in the developing nations are probably less efficient (i.e. emit more carbon for each unit of output).

A solution gaining increasing prominence in the light of China's behaviour is a border carbon adjustment system (see the Carbon Dividend proposal again). This should be set at the same carbon price that applies within your economy (both for economic efficiency, and to have any chance of approval by the WTO).

If you have manipulated your carbon price downwards by assuming various other interventions to make the government's "winners" viable at an artificially-low carbon price, you must set your border adjustment at that price. But it reflects only a fraction of the true cost. So your foreign competitors still get to undercut your native producers, except for the privileged "winners". Your consumers get to pay for this convoluted way of privileging some and harming others of your native businesses. All for the dubious benefit of reducing the efficiency with which we address the risk of climate change.

Stiglitz and Stern: knaves or fools?