John Lanchbery (RSPB)

1.  Introduction

Time is now running out in which to conclude agreement on the outstanding issues left over from Kyoto. Two months before COP 6 in the Hague, where everything is meant to be sorted out, nearly all of the major problems remain unresolved. This paper examines the most important of these problems and proposes some solutions to them.

The first subject to be considered is that of sinks for carbon dioxide and how, or indeed whether, to include them in Clean Development Mechanism (CDM) and Joint Implementation (JI) projects. Also addressed are the questions of what, if any, “additional activities” should be included in Article 3.4 and how to include forest-related activities in Article 3.3. The answers to these questions are important because they will not only determine how parties set about achieving their targets but also what the targets actually are, in practice.

The second main subject is that of the flexibility mechanisms (emissions trading, the CDM and JI). It is especially important that progress is made on the CDM in Lyon, otherwise there is a danger that the mechanism may not become operational for some time, in spite of the fact that it should start this year.

The paper concludes with a discussion of the enforcement of compliance, and how to achieve it.

2.  Sinks

This section focuses on sequestration by forests but similar issues arise in other forms of sequestration based on biological activity, such as agricultural soils and oceans.


There are many reasons for being extremely cautious about using forests for carbon sequestration but the three main ones are

i) the uncertainty of measurement,

ii) the fact that the rate of, and capacity for, sequestration will change over time, and

iii) the potentially adverse impacts on biodiversity.

These subjects are examined in detail here, and are followed by a short discussion of definitions of forests and methods for measuring their carbon storage capacity.

Before considering these matters, however, it important to point out that the solution to human-induced climate change is not carbon sequestration but reducing emissions. The Earth’s forests, oceans and soils simply do not have the capacity to sequester the ever increasing volume of greenhouse gas emissions. Any solution to climate change based on sequestration is thus, at the very best, a partial solution.

2.1.  Permanence

Sequestration of carbon by green plants and soils is a temporary phenomenon.  Not only is climate change itself likely to change the sequestration capacity of forests (and soils) but they are also prone to being affected by a host of natural phenomena, notably fire but also diseases and pests. As recent events in the USA demonstrated, even the most highly developed countries can often do little or nothing to prevent fires in either managed or unmanaged forests. (Indeed, from a biodiversity standpoint one may not always want to prevent fires because they are often a vital part of natural life cycles of plants.)

The temporary nature of green plant-based sequestration poses fundamental problems for project-based activities, especially the CDM. One example of such problems is what happens if a forest sequestration project is conducted in a developing country but the forest is later destroyed? A “high quality” forest might be grown that is bulging with biodiversity and thronged with indigenous peoples then, when the project is over and sequestration credit taken for it, the forest burns down. The carbon has returned to the atmosphere as though it had never been sequestered (unlike, say, a renewable energy project where, if the facility ceases functioning, the carbon that it saved during its operation is, at least, not put back onto the atmosphere). Yet there is no obligation on a developing country host country to make up for the loss, because it has no emission limitation target.

Indeed, even if the host country is capped, as in the case of a JI project, the question of who is liable for the replacement of the forest still arises. To keep the integrity of the Annex I cap somebody has to make up for the loss, but does the donor pay again, and possibly again and again if the forest is destroyed and is replaced more than once? Or does the host (that was hoping to make money from the deal) pay?

This type of problem is made worse by the fact that loss or degradation of a forest may take place decades, centuries or even millennia after a project is complete and credit has been claimed for it. Could one retrospectively cancel sequestration credits awarded in a particular commitment period one or two hundred years after it was complete? What implications would this have for compliance?

There is also the question of liability for the causes of degradation or destruction. Who, for example, is liable if a forest is lost or degraded entirely or partly due to anthropogenically induced climate change? All of Annex B? All parties and non-parties? Pixies?

Because the liability issues concerning the CDM are essentially insoluble, forests should never be included in the CDM. Neither should they be included in JI, at least for the first commitment period, or ever unless the liability issues are satisfactorily resolved. It is easy to talk about matters such as guarantees of permanence and joint liability but it would be much more difficult to put them into effect.

In theory, lack of permanence is less of a problem for Articles 3.3 and 3.4 because these apply to capped countries, which should make up for any unpredictable loss of sequestration by other means, such as reducing emissions. In practice, however, making up for such losses, particularly if they are rapid and occur in the final year of a commitment period, is likely to be very hard indeed. So, whilst the Kyoto Protocol clearly states that forests can be included in Articles 3.3 and 3.4, it would be most unwise to do so.

Note that although the above argument has concerned forests, very similar arguments can be applied to agricultural soils. How does one ensure, for example, that a low tillage regime either continues forever or is the responsibility of a particular entity or entities forever?

2.2.  Uncertainty

Sequestration in biological systems is hard to assess accurately, with estimating uncertainties of 50% or more being common for the entire land use change and forestry sector. Whilst all emission source and removal estimates are subject to some degree of uncertainty, this level is unacceptable by any standards and in any of the Kyoto Protocol Articles.

There are ways of partially getting around this problem, notably by systems of reserving and discounting where the sequestration value credited takes into account the uncertainty. For example, if the uncertainty were 50% then only 50% of the estimated sequestration might be counted. The disadvantage of these sorts of system is that they rely on estimates of uncertainty being reasonably reliable and systematically arrived at, whereas they are frequently little more than guesses. A further disadvantage, to investors, is that they are only likely to obtain a fraction of the credit that they might otherwise obtain.

A better method might be to only allow sequestration where less than a given level of estimation uncertainty could be assured. Only counting sequestration which could be verifiably estimated to an uncertainty of less than 5 or 10% would probably be reasonable.

It is sometimes argued that, whilst such eligibility criteria might be reasonable for project-based activities (under the CDM and in JI) and as a restriction on Article 3.4, no such limit should be placed on direct, human-induced afforestation, reforestation and deforestation (ARD) activities under Article 3.3. But Article 3.3 refers to changes in ARD “measured as verifiable changes in carbon stocks” and reported in a verifiable manner.

The clear implication therefore is that if changes in ARD are not verifiable then they should not count at all. Yet a necessary component of verifiability is the ability to measure something with a reasonable degree of accuracy. It is hard to verify something when one is not reasonably confident of what it is and how much there is of it. It is thus not only desirable but essential to limit the degree of uncertainty of ARD-related estimates under Article 3.3.

2.3.  Impacts on Biodiversity


The basic problem with using forests as a mean of sequestering carbon is that the types of forestation that yield maximal sequestration are not optimal for biodiversity and vice versa. For maximum sequestration the imperative is to plant fast growing, densely packed forests, typically of monocultures, such as conifers, eucalyptus or acacia. This is precisely what is not needed to maximise and sustain biodiversity.

Another problem is that it is desirable from a biodiversity standpoint that some types of trees undergo regular burning cycles (such as banksia in Australia and giant sequoias in North America). Although such fires tend to mainly clear underbrush, regular burning of forests can conflict with a wish to maximise sequestration. Yet another biodiversity-related problem is that forests can abstract very large quantities of water, which can have disastrous consequences if they are grown in areas that are not naturally forested.

In theory, tensions between climate change mitigation and biodiversity strategies need not necessarily occur. In practice, however, it is difficult to see how they will not, at least in some countries. It will be extremely hard to write cast-iron guarantees about biodiversity conservation into an international agreement whose focus is climate change. Such guarantees have not been provided by the Biodiversity Convention and attempts to negotiate a global forests convention which included them have repeatedly failed. The chances of obtaining them in an agreement covering another subject thus seem slim.

Even if it were agreed that biodiversity provisions should be included in COP decision concerning the Protocol, they might be difficult to compose in such a way that they could be effectively implemented and monitored.

In practice, from a biodiversity perspective, it would be therefore be best not to include forests in Protocol at all. Too many things can, and probably will, go wrong.


Sequestration in oceans is sometimes mooted. Tinkering with the composition of the World’s oceans, upon which so much life depends, would be extremely rash given our current state of knowledge. Indeed, even with a better knowledge of possible impacts it is unlikely that all contingencies could be covered and nasty surprises would always be possible.

One can sequester carbon dioxide in the oceans either by injecting it directly or by encouraging natural processes by enhanced fertilisation. There are two main types of fertilisation: with iron or macro-nutrients.

In some ocean regions, phytoplankton utilisation of available nutrients is limited by lack of iron. Its addition should thus increase photosynthesis and hence carbon dioxide uptake. A problem with the iron fertilisation solution is that it may lead to a decrease in the oxygen concentrations in the water, which would most probably have adverse impacts on biodiversity. It might also lead to increased production of methane or nitrous oxide: both of which are greenhouse gases with higher Global Warming Potentials than carbon dioxide. It is also worth noting that iron fertilisation is only likely to have a significant effect in the Southern Ocean - the last of the relatively “untouched” oceans.

The alternative is to use macro-nutrients to stimulate biological activity. In this case adverse biodiversity impacts are more certain. Added macronutrients are clearly going to affect all phytoplankton and zooplankton directly, and hence higher animal life too. One obvious concern is that the enhanced growth of some species will predominate, resulting in the reduction or wipe-out of others. Far too little is known about plankton species to predict what effect this would have, either upon biodiversity or upon atmospheric carbon dioxide concentrations. As in the case of iron fertilisation, there is also a risk of creating oxygen deficiencies and of negative atmospheric feedbacks in terms of gases released.

Direct injection of carbon dioxide into the oceans could have severe impacts on wildlife, especially sequestration at the seabed. Changes in pH are alone likely to affect marine life adversely.

2.4.  Forest definitions and measurement

Some parties appear to have encountered difficulties in interpreting the recent IPCC report on land use change and forestry. In particular, some appear to have confused the definition of what a forest is with how to measure its carbon sequestration capacity. The two need have nothing to do with each other and, in this case, should have nothing to do with each other. Defining a forest in terms of canopy cover, as the FAO do, is perfectly reasonable but to base a measurement system for the Protocol on canopy cover would be silly.

Some parties have also interpreted the fact that the IPCC report mentions different measurement systems as meaning that all are equally valid and applicable to the Protocol. They are not. Some are much better than others in terms of comprehensiveness and accuracy. Whilst they are not perfect, the IPCC revised 1996 Reporting Guidelines and the new Best Practice Guidelines will provide the best way of assessing sinks in the context of the Protocol.

A few parties have expressed concerns about how to take account of forest harvest cycles, implying that the IPCC Guidelines do not take cutting and regrowth into account. But the current Guidelines do just that. A party is penalised (in terms of emissions) when a tree is cut down and rewarded if a new one is grown. It is hard to think of a fairer system. Of course, some countries will do better than others out of the setting of a common base year and a common period to a target year but this will always be the case.

3.  Flexibility mechanisms

This section is based upon the assumption that agreement on all three mechanisms mustshould be reached in the Hague, if only as the quid pro quo for the JUSCANNZ countries ratifying the Protocol. An agreement without JUSCANNZ, which comprises most of the World’s largest emitters of greenhouse gases, would clearly not be worthwhile although, equally clearly, if the mechanisms are eventually structured so as to deliver no environmental benefits, one must consider whether any agreement, however bad, is better than no agreement at all.

The section focuses on ways in which the mechanisms can be made to function so as to yield maximum environmental benefit. However, it should be noted that the framework for emission transfers established by the Protocol is far from ideal, and any attempt to make the mechanisms work well is, in many ways, trying to make the best of a bad job. To help in understanding why certain proposals are made here, a brief description is therefore first given of what an ideal emissions transfer system would comprise.

Criteria for effective mechanisms

The basis for the emissions transfer system established by the Protocol is the cap and trade regime for Annex I parties. (Emissions trading should really be called “cap and trade” or “capping and trading” because this term better describes how such systems work, i.e. that capping emissions is the key feature and that trading is a means whereby the participants can jointly attain the cap.)

Cap and trade systems work best if the cap (target) is challenging and applied uniformly to the participants. A challenging cap is needed both to ensure environmental benefits and to drive the trading system by setting up a situation where some participants find it hard to reach the target and others find it comparatively easy. In the Kyoto Protocol the cap is environmentally inadequate and differentiated - generally with countries which think that it will be hard to reduce emissions having smaller caps and with those that think it will be easier having larger caps.

Other features that a cap and trade system needs to work well are a strong, enforceable compliance system and a cap with no leaks - that is, no participation by uncapped players. Yet the nature of international agreements mitigates against enforcement, as is discussed at length later, and the CDM constitutes a massive leak.

With an inadequate and differentiated cap, a compliance system which is likely to be without effective enforcement, and a potentially huge leak from the cap, it is questionable from an environmental standpoint whether the emission transfer systems established by the Protocol will be worth using. When the mechanisms are in place parties might therefore do well to recall that just because a mechanism exists it does not mean that one has to use it.

On the other hand, there are clear environmental benefits to be had from developed countries promoting the use of climate-friendly technologies in developing countries via the CDM. There are also environmental benefits in promoting the introduction of cleaner technologies in Central and Eastern Europe, via joint implementation (JI).

The aim should thus be to ensure that the mechanisms work effectively in spite of the deficiencies in the Protocol.

3.1.  The CDM

There are two key questions concerning the CDM. The first is how to ensure that projects deliver real and verifiable emission reductions over and above what would have happened anyway (the so-called “additionality” question). Because developing countries have neither reliable baseline emission estimates nor reliable emission projections, it is not possible to know, or even to have a reasonable guess at, what would have happened anyway. It is therefore impossible to assess CDM project additionality in terms of overall developing country emissions.

Furthermore, even if developing countries did have reliable baselines and projections, they do not have Kyoto Protocol targets (i.e. they are not capped). They can thus increase their overall emissions whilst also participating in CDM projects aimed at reducing emissions.

To avoid the additionality problem and the perverse situation where a project might reduce local emissions whilst overall country emissions rose steeply, it is necessary to have project eligibility criteria which ensure that CDM projects either necessarily result in absolute emission reductions or, at least, do not result in any net emissions. Only renewable energy projects and energy efficiency retrofit projects should therefore qualify for the CDM, sequestration being ruled out for the reasons given in the previous section on sinks.

This solution does not, however, solve the problem of how to quantify renewables or energy efficiency projects. Whilst measuring the energy generated by a renewables project is, for example, feasible, one needs to know what was being replaced in order establish the value of a project in terms of emissions. So, the second question concerning the CDM is that of how to establish baselines.

Because many developing countries do not have reliable baselines, it will be necessary to establish generalised ones based on the technologies in place in particular countries, localities or regions. One might, for example, use the average emissions from all generating plant in a country as a baseline. However, whilst this type of baseline might be easy to establish in a quite highly advanced developing country, such as China, Argentina or Egypt, it could be difficult in least developed countries and, in addition, the cost of acquiring such information could be high.

To minimise costs for poorer countries with inadequate information about their energy systems, there is thus a strong case for having a general, global baseline for small projects of, say, under 1 to 5 MW. This would help least developed countries acquire the types of technology that they need to help prevent climate change whilst ensuring that the costs involved were minimised. To further reduce costs a degree of automaticity might be introduced so that, for example, a small renewables project in a least developed country would be sanctioned with minimum red tape.

3.2.  JI

Although additionality criteria should apply to JI projects, absolute additionality is not essential because JI projects will be undertaken in countries that are capped. The perverse situation in which a particular project cuts emissions whilst the rest of a countries emissions rise out of control should not therefore arise.

Also, most developed countries have relatively reliable baseline emission estimates and emission projections against which to judge additionality, at least approximately. Therefore, subject to their fulfilling project additionality criteria of the type outlined by the Protocol, there is no reason why, in theory, JI projects should not include renewables, a very broad range of energy efficiency measures and perhaps, after the first commitment period, carbon sequestration.

However, in practice many of the countries in which JI projects are likely to take place do not have reliable baselines and baseline projections and it can often be hard to improve them because, for example, 1990 marked the beginning of an economic and social transition in Central and Eastern European countries and data from that time can be hard to obtain and can be inherently unreliable. For the first commitment period it would thus be wise to restrict projects only to renewables and energy efficiency retrofits, as in the case of the CDM.

3.3.  Emissions trading

3.3.1.  “Hot Air”

Although many countries have reservations about emissions trading, it will probably be agreed as a quid pro quo for JUSCANNZ, and especially US, participation. In this case, a key question will be how to eliminate or minimise so-called hot air, which arises from countries of the former Soviet Union, notably Russia, having taken on a target that is considerably in excess of their current emission levels.

A number of solutions have been mooted to the hot air problem, notably the EU’s “concrete ceiling” concept. This does not, however, eliminate hot air, although it does limit it. (To be fair, the EU’s ceiling is aimed mainly at encouraging domestic action rather than restricting hot air.) Indeed, few solutions have so far been proposed which do solve the problem. The situation is aggravated by the fact thatHowever JUSCANNZ consider that trading, complete with hot air, is part of the deal done in Kyoto.

In fact, it would probably be best for themselves if Russia and the Ukraine participated in JI projects rather than trade. This is because it is very likely that any market based on trading between entities (firms) will mark down former Soviet Union permits or credits because of the inherent uncertainty in their baselines, the fact that they might oversell and other reasons. It is also likely that most of the money from trading will not go towards the establishment of more advanced, greener industry in those countries.

JI projects should ensure that cash goes where it needs to go so as to maximise the benefit both to the environment and to their industry. If Russia and the Ukraine opt for using JI rather than trading, hot air will be minimised.

3.3.2. Developing country participation in trading

A potentially far larger problem than Russian and Ukrainian hot air is so-called “tropical hot air”. This will arise if developing countries are allowed to take on “voluntary” commitments under the Protocol and are also allowed to participate in emissions trading.

The problem stems mainly from a US Senate Decision (the Byrd-Hagel committee decision of 1997) which stipulates “meaningful participation by developing countries” as a pre-condition for the Senate ratifying the Kyoto Protocol. The US Administration has interpreted this in a number of ways, one of which is to press for voluntary commitments by developing countries to reduce their emissions below what they would have been otherwise.

However, because developing countries are not obliged to limit emissions they can take on whatever level of commitment they like. They could adopt huge emissions increases. Although this is not what the USA wants, it may well be what it inadvertently encourages, especially because, as is discussed shortly, there are perverse incentives for taking on an unrealistically large target.

Moreover, even if a country was really trying to reduce its emissions below a level that would have been reached anyway, it is not possible to assess what would have happened because developed country emission estimates and emission projections are unreliable at present. Therefore, a country taking on a voluntary commitment with the very best of intentions could not reliably know what its target really was and could not adequately monitor progress towards it.

A situation is thus likely to arise where, deliberately or accidentally, a country takes on a voluntary emission increase which it will undershoot, possibly by a large margin. If that country is allowed to trade it will have available fictitious emissions reductions equivalent to the difference between the target and the “undershoot” level of emission. The cumulative effects of many developing countries trading this hot air could seriously undermine the integrity of the Protocol.

One could argue that developing countries should not be allowed to trade, and certainly the Protocol makes no mention of their doing so. However, if they cannot trade there is no incentive for them to “formally” take on voluntary commitments. Trading would earn them hard cash but the commitment, by itself, would earn them nothing in financial terms. If they are simply keen to limit their emissions they can do so anyway. There is no need to formalise such an arrangement.

Voluntary commitments coupled with trading should be opposed. Its proponents will find it hard to continue to advocate a fundamentally irrational position that will lead to problems that they do not want and for which they have no solutions.

3.3.3.  Standards for trading

It is essential to have certain minimum standards for trading. Two things are absolutely essential: sound eligibility criteria and some form of mixed buyer/seller liability (which might include modified versions of the “Swiss proposal”). Eligibility criteria are especially important because it will be extremely hard to control a trading regime once it gets going.

To be eligible to trade, parties to the Protocol must

·      have in place agreed national systems for inventory compilation,

·      have reported on their baseline emissions using their national systems, and

�      have had their reports reviewed by the international system prior to the first commitment period.

These criteria are simply common sense and it is thus encouraging that the EU and USA appear to be converging towards agreement on having them. Without such criteria, countries� emission baselines will not be known and agreed and hence they will have no agreed assigned amounts from which to trade.

A key question concerning eligibility is what degree of accuracy in baseline emission estimates is acceptable to the international community. Given the high degree of uncertainty inherent in some countries� baselines it may well be that they will not be deemed eligible to trade, or that trade with them will be restricted. A solution to the uncertainty issue would be to discount or �reserve� a portion of a countries� emissions transfers in proportion to the uncertainty on the inventory (as mentioned previously in the discussion on sinks). A 30% uncertainty might, for example, attract a 30% discount.

4.� Compliance and enforcement

Compliance regimes in international agreements suffer from the fundamental disadvantage that if countries really do not like an agreement, or a penalty imposed on them by that agreement, they can simply leave it. There is, fundamentally, no means of enforcing international agreements. Indeed, penalties for non-compliance can aggravate non-compliance. If a penalty is set too low then governments will simply �buy� themselves into compliance. If a penalty is set too high then there is a risk of double non-compliance: once for not meeting an obligation and once for not paying the penalty for it.

As a consequence, effective compliance systems are invariably based on the withdrawal of rights or privileges conferred by membership of a treaty, rather than upon externally imposed penalties. In the Kyoto Protocol the only rights and privileges (other than doing good for the environment) are participation in the flexibility mechanisms and it therefore, ostensibly, makes sense to base any penalties on the withdrawal of these rights.

However, not only may some countries not participate in the mechanisms but it also makes no sense to prevent countries from buying emission reductions to bring themselves into compliance - that is, after all, the whole point of the mechanisms. Also, it would be grossly unfair to the participants in CDM and JI projects to cancel them once underway and, moreover, if firms considered that cancellation was a serious possibility they would probably never participate in such projects.

This leaves the possibility of preventing sellers of emission reductions from selling if it is demonstrated that they were not complying with their commitments. Whilst this makes sense, it makes for a rather limited form of compliance mechanism.

The number of alternative viable options is limited. This is because one really needs a compliance and enforcement system in which parties have no choice but to pay a penalty. This means that the system either needs to accrue penalty payments as some sort of levy on transactions (with the levy being repaid to compliant parties at the end of the compliance period but not repaid to those that are non-compliant) or the next commitment period needs to be used to fund a current one.

The USA favours the latter course of action (borrowing emissions from the next compliance period) but there are a number of serious concerns about this approach. For example, there might be a tendency for parties that think that they will be non-compliant to artificially inflate their assigned amount for the following commitment period.

The alternative is to have a compliance reserve in which a levy (either in emissions or cash) is imposed on transactions made by participants in the mechanisms. A financial levy would be better because the accumulated cash could then be placed in a compliance fund of the type proposed by CIEL and WWF, to pay for projects that would repair the damage done to the atmosphere by non-compliance. However, to finance ministries, such a levy might well look like an international tax, and they tend to instinctively oppose taxes out of their direct control. But it would be environmentally effective - which is what the present negotiations should be concerned with.


�   Carbon sequestration should not be allowed in the CDM.

�   Carbon sequestration should not be allowed in JI projects during the first commitment period.

�   No �additional activities� should be allowed under Article 3.4 during the first commitment period.

�   Direct, human-induced afforestation, reforestation and deforestation activities should only be included under Article 3.3 where it can be verifiably demonstrated that the sources and sinks involved can be reliably estimated to within a given level of uncertainty.

�   Only renewable energy projects and energy efficiency retrofit projects should qualify for the CDM.

�   There should be a global, technology baseline for small CDM projects, say under 1 or 5 MW, in order to facilitate the introduction of renewables technologies to least developed countries.

�   For the first commitment period, it would be wise to restrict JI projects only to renewables and energy efficiency.

�   Developing countries should not be allowed to trade, irrespective of whether they take on voluntary commitments.

�   To be eligible to trade, Annex B parties to the Protocol should:

�               have in place agreed national systems for inventory compilation,

�               have reported on their baseline emissions using their national systems, and

�               have had their reports reviewed by the international system prior to the first commitment period.

�   There should be a compliance reserve in which a levy (either in emissions or cash) is imposed on transactions made by participants in the mechanisms. Payments would be returned to compliant parties and retained in the case of non-compliance.


Published by:
The RSPB, The Lodge, Sandy, Bedfordshire SG19 2DL, UK.

telephone� +44 1767 680551
fax� +44 1767 691178
email� John.Lanchbery@rspb.org.uk

During the meeting John Lanchbery will be staying at Association Phebus, 114 bd du 11 novembre 1918, Villeurbanne.
Mobile: 07720 599417