Shutting Down the Fog Machine

1. CDM, Additionality, and the Additionality Incentive Fix

A brief explanation of CDM, additionality, and Clouds’s proposed CDM incentive fix is needed before moving on to the critique. CDM allows wealthy Kyoto signatories to meet their national caps on greenhouse gas emissions (GHG) by funding projects in non-signatory poor nations. To the extent these projects reduce emissions, signatory nations can emit more, allowing reductions in uncapped poor nations to make up for the increase. Rich nations pay poor nations to cut pollution on their behalf. The claim: the climate does not care where greenhouse gas emissions are cut; thus this offset mechanism reduces compliance costs for rich Kyoto signatories, since poor nations can cut emissions less expensively than rich nations can. This helps poor nations, because they receive project funding.

Please note that, since CDM offsets serve as pollution permits in rich nations, they will harm the climate unless they are 100 percent additional. An “additional” offset is one that would not have happened, in full or in part, without CDM funding. This can be a problem, even if the project approved is a good one. Suppose a wind farm receives CDM funding, and offsets from that wind farm are approved. If the decisions to fund that wind farm had already been made by investors before CDM funding was applied for, then that wind farm would have been built anyway, regardless of whether CDM funding was received. Projects, such as this wind farm, for which CDM funding is a non-essential bonus, are called “anyway” projects, and are not additional. If non-additional offsets from the wind farm are used as permission to burn coal in Scotland, those credits result in more pollution than if they had not been approved. Only if the wind farm requires CDM credits to be viable is that project additional. Thus, additionality is critical if CDM is to help rather than harm the climate.

Offsets have many problems, including an often predatory nature. Many CDM projects documentably contribute to displacing the poor or harming their health (Roht-Arriaza 2010: 594-602). Clouds tackles a different problem: the current CDM system has massive incentives for CDM project creators to overestimate emission reductions, or to seek credits for projects that would have happened anyway. As previously mentioned, since certified emission reductions (CER) generated by CDM are used in Kyoto signatory nations as permits to pollute, to the extent that these offset certificates are granted for reductions that would have happened anyway, or are based on inflated claims, net greenhouse gas pollution is higher than it would have been without CDM. What primarily concerns Hahnel about CDM is that so long as regulators approve credits, neither buyers nor sellers have an incentive to care whether CDM credits are bogus or not. Because regulatory approval is what gives CDM offset credits their value, buyers lose nothing if the credits they buy are bogus. How bad has this proven in practice? Clouds admits that the problem is real, but asserts that its scale has been exaggerated. The data, and much expert opinion, contradict the latter assertion.

Historically a majority of CERs granted final approval under CDM were based on reductions in high global warming potential (GWP) industrial gases (UNEP 2013) which in turn are widely acknowledged to have serious problems with additionality (Schneider 2011: 851-864). A growing trend is the granting of CERs for wind power and hydroelectric projects. Banks mostly finance projects of this type that are sufficiently profitable to be feasible without CDM money, meaning they are almost completely non-additional (Alexeew et al. 2010: 242). There is a widespread agreement that CDM tends to fail additionality for a variety of reasons. CERs are granted based upon a business as usual (BAU) baseline, the story told about what would have happened in the absence of CDM project funding. We never truly know what might have been.

The value of the projects to CER buyers and sellers lies in regulatory approval, not emissions reductions. As Hahnel notes, neither buyers nor sellers have an incentive to ensure CERs are real or additional, provided those CERs win regulatory approval. Project validity depends 100 percent upon regulation, which places an overwhelming burden on regulators.

Leading experts on CDM agree that CDM as it currently exists is flawed. For example, the World Bank describes project based CDM as: “. . . the worst of both worlds: high transaction cost with substantial nonadditionality. A growing consensus views determination of additionality as quixotic at the project level. . . ” (Chomitz et al. 2010: 72).

In a recent Delphi survey overwhelming majorities of experts agreed that: “. . . In many cases, carbon revenues are the icing on the cake, but are not decisive for the investment decision. . .,” and “. . . Many projects would also be implemented without registration under the CDM. . . ” (Cames et al. 2007: 98).

A recent European Commission CDM study concluded: “. . . verifications, particularly of additionality and baselines, are widely critiqued in terms of: inadequate rigour and transparency, conflicts of interest. . . ” (Ruthner et al. 2011: 15-T2).

Alexandre Kossoy, senior financial specialist at the World Bank’s Carbon Finance Unit, was interviewed in July of 2013 about approved CDM projects that have responded to falling CER prices by no longer registering for credits. In that interview Kossoy said that the projects no longer participating in CDM “will survive without it. . . . For most, getting (credits) is the cherry on top of the cake” (Szabo and Twidale 2013).

Please note that these are criticisms of and admissions about CDM as it currently exists. Most of these experts advocate fixing rather than eliminating CDM. Again, this is one good reason for focusing on the CDM portion of Hahnel’s proposal. It is probably the strongest and most fundamental fix to CDM ever proposed. If CDM remains problematic with the changes Hahnel suggests, that is a strong indication it is broken beyond repair.

Clouds’s proposed incentive fix: offsets would occur in nations with caps rather than in nations without. National rather than international regulators would approve projects. But each approved CER would lower the cap of the nation approving it by the amount of that CER. Thus, approval of bogus CERs would not increase emissions; nations who approved failed projects would be required to compensate for those failures through other reductions. Given their liability for regulatory failures, Hahnel claims national regulators are likely to ensure that approved CERs are not bogus, that most projects are real and additional.

Unfortunately, as the next section will show, Hahnel’s proposal would continue to ensure approval of large numbers of bogus projects. At best poor nations would need to compensate with new reductions that would cost them most CDM revenue. More likely, as a later section will argue, bogus projects would undermine climate targets.

2. The CDM Incentive Proposal Does Not Solve the Distributional Problems

Under Hahnel’s proposal, buyers and sellers would continue to lack incentives to create or buy real CERs. As Robin Hahnel himself says: “. . . even in the. . . treaty proposed here, there will continue to be powerful incentives for individual parties to sell bogus carbon credits, allowances, and offsets. Buyers cannot be relied on to prevent this, because buyers have no reason to care if an allowance is bogus. . . as long as it is accredited. . .” (Hahnel 2012b: 88-89). So this “improved” market still relies entirely upon regulatory effectiveness. Does Hahnel believe that project originators will not find fooling regulators both possible and less expensive than creating real emissions reductions? As the 2011 United States Budget Control Act (sequester) shows, government officials are not strongly motivated by penalizing the public for officials’ failures. Such penalties against the general treasury provide little incentive for regulators, and are a weak bulwark against the principal/agent conflict inherent in the CDM system.

Hahnel’s proposal fails to solve the problem of measuring additionality at the project level. Measuring national emissions reduction, which can be done somewhat accurately and precisely, does not reveal the contributions of approved projects. When a nation meets its lower target we do not know whether this is due to aggregate project successes or other causes. In case of failure, we do not know where the failure occurred. CERs are still the difference between what happened and a story about what might have been, just as with existing CDM. That the nation approving these CERs guarantees them in aggregate does not change the fact that separately their validity cannot be determined. Projects are still being approved and guaranteed at a more granular level than the one at which they can be measured.

Aggregate measurement does not change the fact that CDM occurs in an environment of massive uncertainty, and massive information asymmetry. Not only is project level additionality inherently highly uncertain, but project creators and consultants have better access to what information does exist than regulators.

Capital requirements aggravate both uncertainty and principal/agent conflicts. The difficulty of measuring project level emissions produces high transaction costs per project. Large CDM undertakings have economy of scale advantages over small ones, because they generate more emission reduction claims per project. Dividing high per-project costs among more CERs reduces transaction costs per CER. Large scale funding of large projects means that the rich in poor countries, and foreign investors and financial institutions, will be the main backers for these projects. Just as at present, financial institutions and venture capitalists will prefer projects which are viable without CDM funding, that is projects which are not additional. For example, the 2008 United States State Department cable from the consulate in Mumbai, India, released by Wikileaks (United States Consulate-Mumbai 2008: 08MUMBAI340) says:

. . . For this reason, Santonu Kashyap of Asia Carbon maintains that Indian projects can never fulfill the additionality requirement as no developer will risk investing in a project unless he is certain of a revenue stream independent of the CDM incentive. In a separate discussion with GAO analysts and ConGenoff, Jamshed Irani, Director of Tata Sons and the Chairman of the Tata group’s Steering Committee on Sustainability, agreed that no Indian company is brave enough to rely entirely on a CDM-driven revenue stream.

Powerful investors who back non-additional projects will wield as much influence as possible against regulators who threaten to disapprove too many projects. Buyers seeking CERs to lower their compliance costs will also lobby for generous approval standards. Most importantly, carbon traders, including the most powerful financial institutions in the world such as Goldman-Sachs, will push back against regulatory stringency.

Powerful lobbyists often ensure underfunding of public agencies such as tax collection or financial rule-making and enforcement. Full funding of such agencies repays its cost many times over, but at the expense of powerful interests. What are the odds that carbon traders and project developers will not push for weak regulation under Hahnel’s proposed CDM system?

Look at incentives for regulators. Project creation and management is more lucrative and exciting than regulation. From the field workers to top executives, regulators will be reluctant to offend possible future employers.

One piece of evidence: the financial crisis of 2008 showed that the prospect of major government losses from weak regulation provides insufficient incentive for strong regulation. Decision makers at all levels of government seldom face any prospect of sharing losses, and reap rich rewards upon leaving office for lax rule making and regulation. Clouds’s CDM proposal does not reduce incentives to create fraudulent projects, and thus maximizes the odds for regulatory capture. Besides, does Hahnel really want to trust bankers with the climate?

Both because the influence of the people involved in CDM and the difficulty (really impossibility) of determining performance at a project level, we cannot assume that failed projects will be held accountable. If national regulators try to recover losses from project owners, those project owners will deny responsibility, with excellent prospects for success in this denial.

3. Project-level Offsets Undermine Global Caps

The previous section showed how project based offsets under global caps in nations that have agreed to national caps are nearly certain to produce unjust distributional effects. But, Clouds also claims that national guarantees of offsets approved at least ensures that global caps will be met. That claim assumes nations honor their treaty obligations to compensate for bogus CERs. Many readers may suspect that a treaty which results in repeatedly, humiliatingly, and publicly cheating signatories is likely to be violated. The evidence supports this suspicion.

One incentive to violate such treaties when targets are not met: the powerful and influential people who sold bogus CERs will fear that the government assuming such costs will arouse public pressure to recover those costs, overcoming the push-back outlined previously. Similarly, government officials have an incentive to claim that apparent failure to meet emission targets are false accusations by powerful foreign interests.

Note that these pressures come specifically from the combination of project based offsets and constant changes in emission target paths. Without a project focus, fewer powerful private interests have incentives to create bogus reductions. Without CERs derived from projects, no powerful community of carbon traders needs to push back against acknowledging problems.

Further, lowering an existing cap faster than previously agreed differs significantly from agreeing to faster emission reductions to begin with. If a nation approves a project-level offset and lowers its national cap in proportion to projected savings, it will treat that transaction as assuming the obligation and meeting it at the same time. The guarantee is a contingent liability. Under financial rules, contingent liabilities must be accounted for. However, the public and most government officials outside government accounting offices will consider that contingent liability a technicality.

When nations fail to meet more stringent emissions targets due to project failures, the need to honor the guarantee will come as a surprise to the public, and will at least be treated as one by the government. In contrast, if a nation agrees to a lower cap to begin with, in return for payments, the obligation to lower emissions is clear. There is no illusion that simply agreeing to that reduction is the end of the obligation.

In neoclassical terms there is no difference between not receiving something and losing something. But outside the airy realms of free market fantasies, there is a well-known phenomenon called “loss aversion” where people are much more reluctant to accept a loss than to fail to realize a gain under some conditions (Kahneman and Tversky 1984: 341-350; Gill and Prowse 2012: 469-503; Abdellaoui, Bleichrodt, and Kammon 2013: 411-29). This applies to institutions as well as individuals.

Social psychology experiments show little difference between groups and institutions in this regard (Bone, Hey, and Suckling 1999: 63-81; Whyte 1993: 434-450). Empirical data from performance in the finance industry show that when financial institutions perform below financial industry standards, they will take increased risks to try and “make up” losses (Johnson 1994: 73-89; Chih and Shen 2005: 2,687-692; Gurevich, Kligerand, and Levy 2009: 1,221–229; Kliger and Levy 2009: 330-46; Alam and Tang 2012: 156-64). At least one early study extended this to a wide variety of manufacturing industries (Fiegenbaum 1990: 187-203). Anyone who has encountered losing gamblers doubling their bets in an attempt to come out ahead will recognize the phenomenon. Incidentally, most of the studies cited belong to the field of prospect theory. Although discussion of prospect theory is beyond the scope of this paper, a list of suggested reading on prospect theory follows references.

There are good reasons to expect loss aversion in national governments who incur liabilities for failed projects under the proposed CDM fix in Clouds. The previous section has already shown that uncertainty, information asymmetry, and the unexpected revelation of significant sunk costs, all possible contributors to loss aversion, are structurally inherent in the proposal. Other reasons include the powerful interests already mentioned and the injury to national pride in submitting to a very public penalty. That national pride provides a powerful incentive for push-back by public officials in order to save face.

Agreeing to lower caps in the form of guarantees for projects nearly certain to fail has implications for whether the caps are met or not. As previously mentioned, the new liability comes as an unexpected requirement to fulfill an obligation that was widely believed to already have been met. Imagine a public official obliged under a treaty to take unexpected expensive steps to reduce emissions to compensate for failed projects. A very plausible push-back is to refuse to accept responsibility for that expensive failure, to refuse to pay those costs. There is a very good chance of rallying public opinion around not meeting this obligation imposed by a treaty with foreigners. And, unlike conventional foreign debt, the political forces behind the climate movement are weak. The international banks and bond traders are not likely to impose sanctions for refusing to take this action. Powerful foreign states may respond with strong rhetoric, but are unlikely to risk military or trade alliances by taking disruptive actions. This is a chance for a government that normally sides with powerful foreign economic interests against its own people to be seen to publicly defy such interests, at a much lower risk than such defiance normally carries. The incentives are even stronger for governments whose economic policy has not been captured by outside financial and industrial interests.

China offers an example. China has submitted large numbers of wind farm and hydroelectric projects for CDM credits, many of which were rejected. Although China expressed anger over this, its reaction has not gone beyond harsh language (Fielding 2010). However China has also been granted many HFC and N2O credits over the years. In 2010 the EU came close to canceling all HFC and N2O credits, because of repeated scandals surrounding such projects. China pushed back hard, saying if existing credits were canceled, it would release huge existing stored stocks of HFC-23 into the atmosphere (Pearce 2010). In 2011 the European Commission backed down, banning only new CDM credits from high GWP industrial gases issued after June of 2013. New CERs from HFC and N2O and other such gases continued to be issued until that date (European Commission 2011).

China was not issuing CDM under a national cap. But in this case the penalty it refused to accept under the current CDM system was essentially the same one it would face under an “improved” CDM; whereas it accepted the rejection of wind and hydro CERs, because that rejection did not make existing, approved, unsold CERs worthless. The one example we have of this type of penalty reinforces the data and studies suggesting that loss aversion is likely to apply to national entities when asked to spend public money to make up for failed “improved” CDM projects.

The particular structure in Clouds’s version of CDM makes treaty violation much more likely than a system where a more stringent cap was agreed to in the first place. This is especially true since any time the guarantee is triggered, the people of the nation really were bilked when the project was approved. As noted at the beginning of this section, an agreement structured to create a high probability that the people of a nation will repeatedly and publicly be humiliatingly cheated maximizes the risk of that agreement being dishonored.

Neither the distributional losses mentioned in previous sections, nor the risk to global caps, are worth the gains Clouds claims “improved” offsetting would produce: lowering costs and transfer of revenue from the developed to the less developed nations. Besides, there is no reason to believe this new version of CDM would accomplish either. In the current system, less than a third of each CDM dollar is spent on project capital and operating expenses, around half of CDM money stays in the developed world, and much of the rest ends up in the hands of the rich in poor nations (Carbon Retirement 2009). Clouds’s proposal would rely on the same layers of financial middlemen and incur the same transaction costs. Better means exist to achieve climate goals efficiently and compensate poor nations. These means include public investment, command and control regulations, carbon fees, and a climate fund.

4. Net Emissions, Additionality, and Cap Stringency

In addition to the CDM fix, Clouds contains a proposal intended to repair flaws in the measurement of damage to the climate by degradation and destruction of vegetation-rich wild ecosystems, such as forests, swamps, and grasslands. The particular flaw addressed is treated as an additionality problem, albeit one caused by mismeasurement rather than by principal/agent conflict. This treatment overlooks some of the technical realities behind the flaws it addresses, and thus proposes a flawed solution.

Clouds points out that healthy ecosystems, among other services, provide critical sequestration of carbon. Trees not only store large amounts of carbon. As long as they live and are healthy, they continue to transfer carbon from the air into green, woody, and root matter, and (at least outside of tropical climates) into the soil. Hahnel’s suggestion: since we wish to both avoid releasing additional greenhouse gas pollution into the atmosphere and preserve ecosystems that remove greenhouse gas pollution from the air, we should measure greenhouse gas pollution on a NET basis: pollution released minus pollution removed. In Clouds, Hahnel says:

Cap net emissions. Net emissions are what matters with regard to climate change, and surprisingly, as already explained, measuring national annual net emissions is as straightforward as measuring only national annual emissions. Capping net emissions rather than capping only emissions would solve an important problem arising from projects selling offsets for sequestration increases. Under Kyoto if a project increases carbon sequestration it can receive CERs. So creating a tree plantation can qualify for CERs because it is easy to demonstrate that new trees planted are sequestering carbon that would not have been sequestered had the trees not been planted. But at this point Kyoto does not give credit for carbon stored and sequestered by existing forests that are conserved because it is difficult to know whether or not the forest would have been preserved in any case. This creates a perverse incentive to replace existing forests with tree plantations. (Hahnel 2012a: 151)

The perverse incentives Hahnel discusses are real and damaging. Counting some forms of sequestration but not others leads to perverse non-additional actions, such as Hahnel’s example of replacing existing forests with tree plantations.

Clouds’s proposed solution, however, would disrupt measurement standards that mostly avoid combining carbon emissions and biological carbon storage for good reasons. The results would be unpredictable. If net sequestration simply means including full net LUCF in all national emission calculations, then in most nations, especially most rich nations, they would result in weaker targets for a given nominal percentage reduction.

Consider the Obama proposal for a 17 percent reduction in U.S. greenhouse gas emissions compared to 2005, when that reduction is translated into millions of metric tons on both a gross and net basis.

Translating that percentage into millions of metric tons shows a smaller drop for net emissions compared to applying the same percentage to gross emissions. A 17 percent reduction from 2005 net emissions results in the need to lower pollution by fewer tons than a 17 percent reduction from 2005 on a gross basis. Further, the difference in stringency between net and gross emissions varies depending upon what base year the drop is calculated from. The ratio of net to gross emissions varies from year to year, so the effect of using net rather than gross emissions on translating percentages into tons also varies from year to year.

The United States is not an exception among rich nations in this regard. In most rich nations (and in many poor ones) net emissions are lower than gross emissions. Thus, lowering net emissions by the same percent formerly applied to gross emissions translates into a reduction of fewer metric tons. Rich nations mostly have some forest and wilderness protection in place, however inadequate. A net emissions based percent cap may prove more stringent than a gross emissions reduction in a small number of nations undergoing extremely heavy deforestation or wilderness destruction, depending upon cap structure. Of the forty or so nations with emissions greater than sequestration from land use change and forestry (LUCF), only five are rich countries: Australia, Germany, The Netherlands, Iceland, and Switzerland (World Bank 2013). The others are poor or developing nations, including Brazil, Tanzania, India, and Indonesia. The world’s remaining nations, including China, Russia, the UK, and the United States, ¹ have net emissions that are lower than their gross emissions. A net emissions-based percentage cap without some form of adjustment would result in less stringent targets in most nations, including many of the largest greenhouse gas polluters. Adjustment is needed to avoid less stringent targets for major fossil fuel polluters. A true net emissions measurement would require similar adjustments, without difference in sign, but with comparable national variance in scale.

No such adjustment is mentioned in Clouds. Any attempt at adjustment would require more stringent nominal targets in most nations to achieve the same results as targets based upon gross emissions. For example, if the right target for the United States is an 80 percent reduction in emissions (according to the way reductions are normally measured) than if negotiations were taking place using net emissions, the nominal target would be 88 percent-92 percent. Adapting a new standard of measurement that requires nominally more stringent targets by the most politically influential parties to obtain the same results as nominally weaker targets under current standards would provide powerful political ammunition to those seeking to weaken caps or who sought to derail any treaty. Further, the adjustment is not simple: the difference in nominal targets required would vary significantly from year to year and from nation to nation in ways that are not predictable in advance.

Thus a percentage cap based on net emissions would be a political obstacle to winning strong caps. The new accounting method, with adjustments, would be more complex, and make these targets appear even more stringent. Without adjustment, the same new accounting method would be a useful political tool for weakening target stringency for the most politically powerful nations.

This in itself is a strong argument against net emissions as a primary means of measurement. However, there is an even more compelling argument against this kind of standard. There is a high degree of uncertainty in the permanence of biological sequestration. Coal is seldom burnt accidentally. With rare exceptions, such as mine and well fires, coal, oil, and gas are extracted only deliberately and burned intentionally. If a choice is made to not extract or consume fossil fuel, so long as that decision does not change, the odds are that fossil fuel will stay in place.

OPEN IN VIEWER

Table 1.

GHG Emissions: Millions of Metric Tons Equivalent.

Year	2005	2011	Obama Emissions Goal (–17%)	Emissions Drop from 2005	Emissions Drop from 2011
Gross Emissions	7,195	6,702	5,972	1,223	730
Net Emissions	6,197	5,797	5,114	1,054	653

Environmental Protection Agency 2013 Inventory of Greenhouse Gas Emissions and Sinks: 1990-1911: Table ES2, p 7. Washington, DC: The Environmental Protection Agency. http://www.epa.gov/climatechange/Downloads/ghgemissions/US-GHG-Inventory-2013-Main-Text.pdf

That permanence does not apply to biological systems. The amount of carbon in a healthy forest or a grassland varies from month to month, from season to season, and from year to year. Much of this is outside of human control. For example, whether the year is a warm or cool one will affect how much carbon ecosystems absorb and release. Beyond this, various types of disruptions affect carbon storage in plants. Fires, storms, disease, pests, flood, and drought all affect how much carbon ecosystems absorb and release. So net sequestration and storage is affected by many factors not predictable in advance, and not under short-term human control. Even though, as Hahnel says, aggregate net emissions are comparatively easy to measure in a single year, the sequestration portion of that net number will vary widely from year to year. Permanence is difficult to determine on large and small scales alike.

Ecosystem preservation and emission reductions fight climate change in different ways. Not burning a lump of coal or a barrel of oil has a high probability of reducing emissions permanently. If we close a mine, the carbon in that mine will probably stay there. The question of permanence in this case is mostly a human decision, under human control. But closing mines does nothing to remove carbon from the air that has already been released.

In contrast, preserving healthy ecosystems and healing sick ones do not result in permanently storing carbon with a high degree of certainty. Much of that uncertainty is not under human control. But such ecosystem preservation and healing not only keep stored carbon from being released, at least temporarily; that preservation and healing also remove carbon from the air, something shutting down a coal mine cannot do.

Both means of fighting climate change are important. But they are complementary and should not be traded off against one another. Saving a tree should not be an excuse for burning a lump of coal. Shutting a coal mine should not be an excuse for clear-cutting an old growth forest. We need separate targets for both, not a single number that treats them as interchangeable. The flaw in Kyoto when it comes to vegetation-rich ecosystems is not that it does not use net numbers, but that it mixes certain types of tree plantation sequestration with fossil fuel emissions in the first place. When it comes to biological sequestration, Clouds tackles the wrong problem.

This is especially true because long-term carbon storage by ecosystems is difficult to measure, and not easily captured by short-term snapshots. The same dynamism that results in lack of permanence in biological carbon capture also limits the value of one-time or short-term measurement of ecosystem carbon to determine long-term sequestration and storage. Forests, grasslands, swamps, and other vegetation-rich ecosystems absorb and release carbon dioxide analogously to the way humans inhale and exhale oxygen. Only multiple measurements of ecosystem carbon, made repeatedly over multiple years, can determine whether the trend in an ecosystem is increasing or decreasing carbon storage, let alone the scale of such sequestration or emission. Even if the individual snapshots upon which the estimates are based are precise, the estimates of medium- or long-term sequestration, by their nature, will tend to be rough, approximate, and more qualitative than quantitative. It is at least possible that measuring the general health of a forest or wilderness ecosystem will prove a better predictor of long-term capability to sequester and store carbon than snapshots of constantly changing carbon content at particular moments. Regardless, snapshots from a single year have low value for measuring medium- or long-term sequestration trends in vegetative ecosystems, at any scale.

Forestry, wilderness preservation, and wilderness restoration, along with more sustainable agriculture, are key components of solving the climate crisis, as is phasing out fossil fuel. But living carbon in stored biological systems and dead carbon stored in fossil fuels are heterogeneous to one another in important ways. The two forms of carbon storage should not be treated as interchangeable. Carbon emissions reduction and biological carbon sequestration must be tackled separately.

5. Conclusion

Clouds attempts to tackle critical climate policy issues within political economy. Unfortunately it overlooks the dictum, usually attributed to Einstein, that everything should be made as simple as possible, but not simpler.

The attempts to solve the CDM additionality incentive issue overlooks that problems with CDM additionality are a matter of capability as well as of intent. It also ignores the fact that the project-based offsets structurally include strong uncertainty, asymmetrical information, powerful principal/agent conflicts, unexpectedly revealed sunk costs, and a powerful tendency toward regulatory capture. That environment means market and state actors’ incentives are highly path dependent. The initial conditions and frame of reference established under Clouds’s modified incentive structure create strong political incentives to violate national caps, mainly in response to the proposal’s failure to repair the existing offset system’s many distributional failures.

The proposal to replace gross emission caps with net emissions caps, in an attempt to fix what Clouds sees as an additionality problem based upon flawed emissions measurement, overlooks important technical issues. The issues include the fact that a net emissions standard would make nominal caps appear more stringent than they are, especially in most rich nations.

The net emission proposal also overlooks the extreme difficulty in measuring permanence of biological sequestration, even on an aggregate basis. It fails to consider that the scale of long-term sequestration, whether local or aggregate, can only be measured over a much longer period of time, and with much less precision and accuracy than emissions. Thus, Clouds ignores technical and scientific arguments against combining greenhouse gas emissions and carbon removal from the atmosphere by biological systems into a single number.

The flaws in Clouds appear to stem from failures to sufficiently consider social, institutional, political, technical, and scientific context. Abstraction and modeling are powerful intellectual and problem-solving tools. But when using these tools, it is critical not to ignore messy complicating factors whose omission can lead away from workable solutions and towards cloud-cuckoo land.