The trick is, how to achieve an equitable outcome that’s consistent with the physics of climate? Consider this reaction to ideas like climate debt:

Obama’s Global Tax

By INVESTOR’S BUSINESS DAILY | Posted Tuesday, July 29, 2008 4:20 PM PT

Election ‘08: A plan by Barack Obama to redistribute American wealth on a global level is moving forward in the Senate. It follows Marxist theology – from each according to his ability, to each according to his need.

…

Obama would give them all a fish without teaching them how to fish. Pledging to cut global poverty in half on the backs of U.S. taxpayers is a ridiculous and impossible goal.

…

We already transfer too much national wealth to the United Nations and its busybody agencies. …

…

If you’re worried abut gasoline and heating oil prices now, think what they’ll be like when the U.S. is subjected in an Obama administration to global energy consumption and production taxes. Obama’s Global Poverty Act is the “international community’s” foot in the door.

…

Obama has called on the U.S. to “lead by example” on global warming and probably would submit to a Kyoto-like agreement that would sock Americans with literally trillions of dollars in costs over the next half century for little or no benefit.

“We can’t drive our SUVs and eat as much as we want and keep our homes on 72 degrees at all times . . . and then just expect that other countries are going to say OK,” Obama has said. “That’s not leadership. That’s not going to happen.”

Oh, really? Who’s to say we can’t load up our SUV and head out in search of bacon double cheeseburgers at the mall? China? India? Bangladesh? The U.N.?

I suspect that these sentiments are quite prevalent, at least in the US. I’m even sympathetic in at least one respect: transfers from the global rich to poor are beneficial in principle, but difficult to execute. Transfers from country to country are susceptible to capture by elites. Direct transfers among individuals could be facilitated by a global carbon market with allowances allocated to individuals (one of the few good arguments for emissions trading in my mind), but would undemocratic regimes permit their citizens to participate?

I don’t see agreement on this front any time soon. I could see things going a different way: the US, EU and a few other developed nations move to reduce, then goad developing nations along with a mixture of carrot (offset projects and other transfers) and stick (border carbon adjustments).

BONN, Germany (Reuters) – China, India and other developing nations joined forces on Wednesday to urge rich countries to make far deeper cuts in greenhouse gas emissions than planned by 2020 to slow global warming.

I’m sure that the mental model behind this runs something like, “the developed world created most of the problem up to this point, and they’re rich, so they should get busy making deep cuts, while we grow a little more to catch up.” Regardless of fairness considerations, that approach ignores the physics of the situation. If developing countries continue to increase emissions, it hardly matters how deep cuts are in the rich world. Either everyone plays along, or mitigation doesn’t work.

I fired up C-ROADS and ran a few scenarios to illustrate:

The top blue line is the AIFI business-as-usual, with rapid emissions growth. If rich nations stabilize emissions as of today, you get the red line – still much more than 2x CO2 at the end of the century. Whether the rich start cutting emissions a little (1%/yr, green) or a lot (5%/yr, green) after that makes relatively little difference, because emissions from the rich world quickly become a small share of the total. Getting everyone to merely stabilize emissions (at 2009 levels for the rich, 2020 for developing countries, black) makes a substantially bigger difference than deep cuts by the rich alone. Stabilizing CO2 in the atmosphere at a low level requires deep cuts by everyone (here 4%/year, brown).

If we’re serious about stabilization, it doesn’t make sense for the rich to decarbonize faster, so that the developing world can construct more carbon-dependent capital that will ultimately have to be deconstructed. It may sound “fair” in carbon-per-capita terms, but I don’t think that’s a very good measure of human welfare, and it’s unlikely to end up with a fair distribution of damages.

If the developing countries are really concerned about climate impacts (as they should be), they should be looking to the rich world for help getting onto a low-carbon path today, not in 20 years. They should also be willing to impose a carbon price on themselves. It won’t collapse their economies any more than it will ours. Without a price on carbon, rebound effects and leakage will eat up most gains, as the private sector responds to the real signal: “go green (but the price of carbon is zero, wink wink nudge nudge).” Their request to the rich should be about the transfers, property rights, and other changes it takes to get the job done with some measure of distributional fairness (a topic that won’t be popular in some circles).

The bottom line:

In this study, we model the distribution of BC’s carbon tax and recycling measures. Our results conirm that BC’s carbon tax, in and of itself, is regressive. However, the overall carbon tax and recycling framework is modestly progressive in 2008/09 — that is, low-income families get back more in credits, on average, than they pay in carbon taxes. If the low-income credit is not expanded, however, the regime will shift to become regressive by 2010/11. It is important for policy makers to rectify this situation in the 2009 and future budgets by minimally ensuring that the credit grows in line with the carbon tax.

A related problem:

A second concern with the carbon tax regime is that tax cuts undermine a progressive outcome at the top of the income scale. In 2008/09, personal and corporate income tax cuts lead to an average net gain for the top 20% of households that is larger in dollar terms than for the bottom 40%.

I plotted the results in the report’s tables to show some of these effects. In 2009, the lowest income groups (quintiles 1-3) come out a little ahead, but the 4th quintile faces a net loss, while the top income group is overcompensated by the corporate tax cut:

As a share of income, things look a little different. It seems hard to get excited about effects this small, but of course the tax will rise by a factor of 3 (more, eventually). If you believe in diminishing returns to the marginal utility of consumption, then it’s clear that a focus on effects on low-income groups really is appropriate:

The report explores several compensation mechanisms, including a flat per-household rebate and an expanded low-income credit. However, the experiments retain a big chunk of the tax revenue for carbon-related projects. That might be sensible in some cases, but I think it’s not as likely to be popular, so I tried yet another alternative: a full flat rebate, which is much like an allocation of allowances on a per-capita basis in a cap & trade system:

The report left me wondering about one thing: what happens when you consider greater granularity. What happens, for example, when you consider household size diversity within income groups? The most salient impact of the carbon tax on wallets is through fuel prices. Since hauling kids is a frequent rationale for inefficient vehicles, I could easily see pressure developing to make the tax more “family friendly.” I’m not entirely sympathetic; I have kids, but don’t regard them as an automatic claim on global environmental property rights. But nevertheless aspects of distribution, like this, might pose important practical and political considerations for the design of instruments.

Update (afterthought): Revenue recycling to reduce taxes is not generally what carbon-intensive firms and their shareholders and employees want. They’d rather be compensated for stranded assets and other dislocations, hence the push for grandfathered emissions allocations in cap & trade systems. The corporate tax relief provided by the BC carbon tax is too coarse to accomplish this; it compensates low-carbon industries as much as high-carbon ones. Better targeting of corporate tax relief might be a way to reduce the overcompensation of the 5th quintile (see figures above) and to make the transition to the tax more palettable to key sectors. I’d also like to see a bigger focus on employees; shareholders are (or should be) diversified and hedged; employees not so much.

Our premeeting tackled Ethics, Values, and the Next Generation of Energy and Climate Modeling. I presented a primer on discounting and welfare in integrated assessment modeling, based on a document I wrote for last year’s meeting, translating some of the issues raised by the Stern Review and critiques into plainer language. Along the way, I kept a running list of assumptions in models and modeling processes that have ethical/equity implications.

There are three broad insights:

1. Technical choices in models have ethical implications. For example, choices about the representation of technology and resource constraints determine whether a model explores a parameter space where “growing to help the poor” is a good idea or not.
2. Modelers’ prescriptive and descriptive uses of discounting and other explicit choices with ethical implications are often not clearly distinguished.
3. Decision makers have no clue how the items above influence model outcomes, and do not in any case operate at that level of description.

My list of ethical issues is long and somewhat overlapping. Perhaps in part that is due to the fact that I compiled it with no clear definition of “ethics” in mind. However, I think it’s also due to the fact that there are inevitably large gray areas in practice, accentuated by the fact that the issue doesn’t receive much formal attention. Here goes:

• What do you recognize in the system boundary?
  • … in the model boundary?
• What constitutes welfare?
  • Do you value non-market services? Health, environment, …
  • Do you monetize the value? (A price implies possibility of substitution)
  • By what method – willingness to pay, willingness to accept, replacement cost? (Results may be skewed by endowment effects or implicit allocation of property rights)
  • Does happiness = consumption?
  • Consumption flows vs. stocks (long-lived durables)?
  • Does temporal distribution matter?
  • Do interpersonal comparisons matter?
  • Leisure time?
  • Does stability or gradual/orderly change have intrinsic value (e.g., because of mortgages)?
  • Does freedom count?
  • Does planning have a cost or benefit? (Or, is it valuable to be able to be impulsive?)
  • Is it intrinsically valuable to be part of the solution vs. part of the problem, help the poor, etc.?
• Are preferences stable?
  • Do they evolve socially, e.g. within networks?
  • Do they interact (vehicle size, phone networks, work time)?
  • Is some consumption an arms race?
• What attitude is taken toward risk?
  • Neutrality, aversion, …?
  • Are risks on earth diversified by risks on other planets?
• Are diverse stakeholders represented?
  • Do winning stakeholders compensate losers?
  • Is inequality within regions/groups considered?
• (How) do you discount? (A discount rate is a price, and implies substitutability)
  • Does discount vary with time, space, or income?
  • Is there heterogeneity in agent discount rates?
  • Do rates evolve socially?
  • Is there pure time preference > 0?
  • How much inequality aversion?
  • Regional/income adjustments (Negishi weights)?
• What is the model horizon?
  • What value is placed on results beyond the horizon?
    • Are there transversality conditions?
  • Is anything meaningful known about deep time?
  • Is there extinction risk?
• What is the basis for the discount rate?
  • Ethical?
    • Principle?
  • Empirical?
    • Market interest rates (revealed preference)?
      • Are rates adjusted for negative social externalities that make social returns < market returns, and possibly negative?
      • Do markets correctly transmit preferences?
    • Future potential for technology/productivity growth?
    • Experiment?
• Are there limits to growth?
  • Infinite technology?
  • Thermodynamic limits?
  • Material resource limits?
  • Environmental degradation?
  • Do they cause overshoot?
  • How will people respond?
• What is substitutable?
  • Can technology bring back extinct species, if we want them?
  • Material consumption replaces environmental services?
• Is the model anthropocentric?
  • Do nonhuman species have intrinsic value or rights?
• Is there rule of law?
• Who is responsible for action? Government? Firms? Individuals?
  • Who is sovereign?
• How do policy levers work?
  • Are they efficient?
  • Are they coercive?
  • Are transfers a “leaky bucket”?
  • Will intercountry transfers be defeated, squandered, captured by elites?
• How do you represent behavior?
  • What is the adaptive capacity of society?
  • How rational are agents?
    • What are the implications of agents’ misperceptions of feedback, information limitations, etc. for their fate?
• Are market failures, institutional barriers, agency problems, etc. represented?
  • Are there tradeoffs between local and global issues?
• What do you recognize and analyze as uncertain?
  • Have extremes been explored? (Reality isn’t plausible, but accepted models have to be plausible).
• Are demographic distinctions made (age, gender, …)?
  • Are children intrinsically valued?
  • Are family size decisions endogenous?
  • Does population enter into the calculation of welfare?
    • Is it better to have lots of poor people or a few rich ones?
• Does trade involve asymmetric power relationships?
• Are distinct successive generations recognized?
  • Can the current generation enforce investment plans on the future?
  • Will the actions of the current generation influence future generations by example?
  • Do generations value bequests to the future?
    • To succeeding generation as a whole, or only to individuals’ direct descendants?
• Are outcomes reduced to a single metric?
  • … or are policy makers (model users) presented with multidimensional output and allowed to construct their preferences for outcomes?
• Are equity and efficiency treated as separable?
• Is equity treatment motivated by principle, or a search for negotiation leverage?
• Do decision makers understand the implications of ethical assumptions in the model?
• Is it possible to identify the ethical implications of policy instruments that have not been explicitly modeled?
• Is there debt or responsibility for past actions (e.g., carbon debt)?
  • Does some time constant apply (do you retire debt as generations die, or natural processes offset damages)?

Backing up a bit, our role in the climate war game was to provide whatever decision support we could within the time frame of the negotiations. That included generating model runs evaluating the CO2 concentration and temperature consequences of proposed emissions trajectories. It also turned out to be helpful for us to generate a variety of metrics combining population, gdp, and emissions, to support arguments for various equity and burden-sharing allocations. For example, using the game scenario data,

That’s fairly predictable stuff, until you start allocating emissions reductions on the basis of population (per capita equity) or other metrics. For example,

The population-based reductions above are fair in the sense that every nation winds up with the same per capita emissions. So why does it appear that China, at 86%, is working nearly as hard as the US, at 96%? The answer is that in one sense it’s not. What really matters is not how much you have to cut, but what you get to keep. When emissions are allocated on a uniform per capita basis, the US gets to keep 4% of business-as-usual. China keeps 14%, and India keeps 61% – 3.5 and 15 times as much, respectively.

The big difference in retained emissions translates to a big difference in cost. While early emissions reductions are cheap or even negative-cost, the cost of deep cuts rises steeply, at least until changes in technology and preferences open up new possibilities. If, for example, costs follow Nordhaus’ parameterized function from the original DICE model, they are roughly cubic and it would take 7% of GDP to eliminate all emissions. (The actual equation is cost = 0.0686*GDP*reduction^2.887.) Then using the population-based cuts above, the US would be spending 6.1% of GDP, China 4.4%, and India 0.5%. The US would be spending almost 150 times as much as India on a per-capita basis, but also better able to afford it. Each of these different characterizations of cuts has a quite different “feel”. (Obviously there are many nuances that I’ve ignored here. For example, one can make a plausible argument that uniform per capita emissions are not equitable enough, when one considers cumulative carbon emissions over history. Economies have different sectoral compositions and climatic conditions. Strict population-based cuts would lead to large differences in regional carbon abatement costs, so they ought to be regarded as rights subject to redistribution by trade rather than as physical allocations.)

The perceptual difference between reductions and “retentions” is further complicated by the fact that any discussion of future reductions is inevitably judged against an arbitrary baseline, either a current benchmark subject to historical accident and measurement error, or a future business-as-usual trajectory subject to large forecast error. This leads me to wonder whether a conversation about actions and effort might be more productive than a conversation about targets. The trick is to make those actions enforceable and sufficient.

Tol concludes:

This paper presents an update of an earlier meta-analysis (Tol, 2005) of the social cost of carbon. Besides more data and more advanced statistical analysis, this paper offers four results. Firstly, there is a downward trend in the estimates of the social cost of carbon – even if the IPCC (Schneider et al., 2007) would like to believe the opposite. Secondly, the Stern Review (Stern et al., 2006) is an outlier – and its impact estimates are pessimistic even when compared to other studies in the gray literature and other estimates that use low discount rates. Thirdly, the uncertainty about the social cost of carbon is so large that the tails of the distribution may dominate the conclusions (Weitzman, 2007) – even though many of the high estimates have not been peer-reviewed and use unacceptably low discount rates. Fourthly, if everyone were to pay a carbon tax equal to the social cost of carbon (but not reduce emissions), there is a fair chance that annual taxes would exceed annual income for many people.

There are three implications. Firstly, greenhouse gas emission reduction today is justified. The median of the Fisher-Tippett kernel density for peer-reviewed estimates with a 3% pure rate of time preference and without equity weights, is $20/tC. This compares to a future price of carbon permits of $8/tC in the European Union (and a spot price of ¢3/tC). The case for intensification of climate policy can be made with conservative assumptions. One does not have to rely on dodgy analysis as in Schneider et al. (2007) and Stern et al. (2006). Secondly, the uncertainty is so large that a considerable risk premium is warranted. With the conservative assumptions above, the mean equals $23/tC and the certainty-equivalent $25/tC. More importantly, there is a 1% probability that the social cost of carbon is greater than $78/tC. This number rapidly increases if we use a lower discount rate – as may well be appropriate for a problem with such a long time horizon – and if we allow for the possibility that there is some truth in the scare-mongering of the gray literature. Thirdly, more research is needed into the economic impacts of climate change – to eliminate that part of the uncertainty that is due to lack of study, and to separate the truly scary impacts from the scare-mongering. Papers often conclude with a call for more research, and often this is a call for funding for the authors or a justification for further papers by the authors. In this case, however, quality research by newcomers in the field would be particularly welcome.

This survey is sometimes cited as evidence that the social cost of carbon is low, and thus not really worth worrying about. However, that conclusion conflates costs (the stream of material climate damages) with a value judgment (the weight assigned to damages in different times and places). The conflation is necessary if one wants to reduce a stream of damages across time and space to a single number, but the value judgment is a major component of the outcome. For a study with a century time scale, the 3% pure rate of time preference (PRTP) above equates to a 95% discount, i.e. that one unit of welfare today would be adequate compensation for the loss of twenty units at the end of the century. Offsetting choices make the ultimate difference to SCC less dramatic (unless one considers scenarios with low or no GDP growth), but still the SCC for studies with 0% PRTP is more than 5x that for studies with 3% PRTP (which you can demonstrate for yourself in the Exhibit).

The usual defense of discounting welfare involves revealed preference: if we observe that people make decisions as if they don’t care about future generations, that must be an expression of their desires, and who are we to contradict the people? Of course, if we routinely make policy in this way, the system becomes neatly circular: we act as if the future doesn’t matter because we observe action as if the future doesn’t matter. It’s actually worse than that, because observations generally involve aggregate outcomes not individual people; to rely on such data as a guide to behavior one must also believe that markets faithfully express individual desires, and that individuals can reliably translate their existential needs into correct actions.

Tol offers some additional caveats:

Although there are now over 200 estimates of the SCC, research in this area is still less developed than one would wish. The 200 estimates of the marginal costs of climate change are based on a dozen of estimates of the total costs of climate change (…). The total cost estimates omit some impacts of climate change; they tend to ignore interactions between different impacts, and neglect higher order effects on the economy and population; they rely on extrapolation from a few detailed case studies; they often impose a changing climate on a static society; they use simplistic models of adaptation to climate change; they often ignore uncertainties; and they use controversial valuation methods and benefit transfers. Unfortunately, this list of caveats has not changed much since Fankhauser and Tol (1996). The proximate reason is that few people work in this area, and none full time, as funding is difficult to get. The ultimate reasons are, firstly, that the issues are complex and uncertain, and require broad multidisciplinary knowledge and, secondly, that the results are unpopular with climate policy makers.

Given the uncertainty in this kind of work, it’s surprising that so much more of the energy of the climate debate focuses on climate science instead.