By Reimund Schwarze, John O. Niles, and Jacob Olander

Activities  that increase forest cover or decrease deforestation can help reduce atmospheric carbon dioxide levels.  Concerns have been expressed, however, that land use, land-use change and forestry (LULUCF) projects may only produce greenhouse gas benefits that are  illusory due to a phenomenon known as “leakage”.  Leakage is the unanticipated decrease or increase in greenhouse gas benefits outside of a project’s accounting boundary resulting from the project’s activities. 
 
Leakage can potentially be significant compared to the scale of planned GHG changes in mitigation projects. Thus, leakage constitutes a key challenge to sound climate change policy formulation. In this paper we review the literature on leakage and pay special attention to LULUCF projects in developing countries.
 
Leakage is complex and poorly understood Leakage is an intricate and diverse phenomenon. Market impacts, people moving from place to place, ecological feedbacks and product life cycle changes are some of the ways leakage can be manifested.

To illustrate: an avoided deforestation project may cause activity-shifting leakage (people leaving a project area to go cut trees elsewhere) or market leakage (less timber available due to the project, more pressure to cut elsewhere). These two types of leakage are the most commonly cited, and are often perceived as  negative (resulting in more emissions or less sequestration  – that is, more atmospheric greenhouse gases). Both of these processes are intricate, difficult to monitor and complicated by many outside influences. 

At the same time this hypothetical project could have unintentional consequences that lead to more greenhouse gas abatement (positive leakage). One protected forest may help adjacent forests stay healthy (an example of ecological leakage) or industries may re-gear production methods to be less polluting as a result of the forest project ( life-cycle leakage).  For different projects the relative magnitude of these types of leakage, both positive and negative, will vary.
 
Leakage is not restricted to LULUCF projects
Leakage can arise from specific projects as well as policies, it can be positive or negative and it can occur in any type of mitigation activity. While some modeling evidence shows that LULUCF activities are at a greater risk for leakage than other sectors of the economy, results are speculative at this point. For instance, fossil fuel leakage has been estimated in the range of 4-40% of the original project carbon offsets, a range similar for LULUCF leakage (~0-100%). However, no study has shown that real -world leakage is more pronounced in certain projects types than others.  Most modeling studies of LULUCF leakage have focused on market leakage and none are based on realistic implementation rates of projects in developing countries.  
 
Leakage in LULUCF projects
Two broad categories of LULUCF activities are considered: conservation projects and reforestation/aforestation projects. 
 
Conservation projects 
In projects that avoid deforestation or modify forest management practices, activity-shifting leakage risk will depend on project design and local conditions.  If a conservation project’s design does not address underlying drivers of deforestation, activities may shift outside project boundaries. This may be particularly true in areas where accessible forests are nearby and where the activity curtailed by a project is mobile. Conversely, well-designed projects and certain circumstances (e.g., all other nearby forests have been removed or the area is highly inaccessible) seem to be i ntrinsically less prone to activity-shifting leakage.  Protecting native forests will often likely generate positive ecological leakage.
 
Reforestation/Aforestation projects
The magnitude of activity shifting from reforestation or aforestation projects will depend (again) primarily on local conditions and the project design. If carried out on lands with other productive uses, tree-planting projects could be especially prone to activity-shifting leakage. Meanwhile, projects that plant trees on degraded lands with little or no other productive use will likely produce less leakage.  Market leakage from reforestation/aforestation projects might conceivably be either positive or negative. For example, large-scale timber plantations could depress timber prices. This in turn could reduce the incentive to establish new timber plantations elsewhere. Equally plausible, timber plantations might reduce timber prices and lessen harvesting on other natural forests. Whether market
forces will lead to a decline in forest harvesting (positive leakage) or the abandonment of plantations (negative leakage) will depend on local market conditions and circumstances.
In general, small-scale reforestation/afforestation projects focused on environmental restoration and/or local community needs should have more positive leakage profiles than large-scale commercial plantations, mainly because they aim to minimize the drivers and scale of potential leakage. 
 
Options for Responding to Leakage
At the project level, tools exist for preventing leakage, detecting leakage, and adjusting for leakage. Options for responding to leakage at the project level include:  site selection, project design, leakage contracts, monitoring. Good project design and careful site selections will minimize negative leakage and maximize positive leakage. Projects that integrate activities (forest conservation, forest restoration, community development, plantations, etc.) will probably be more successful in
reducing leakage (and overall). Leakage contracts can specify actions to deter activity shifting and/or to encourage desired actions. Monitoring can be used to estimate leakage, and once leakage is detected, project carbon offsets should be adjusted accordingly. Other broader policy tools can be used to address leakage. These include:  discounting, projecteligibility criteria, and the use of  “aggregate baselines”. Baselines, the development of national, regional or sector ‘standards’ have been one proposed way to address leakage. This tool is hampered by poor background information for many developing countries as well as political opposition.  Project-eligibility criteria and discounting (or adjustment coefficients) have also been proposed for grouping project types together and making broad generalizations and adjustments. 
 
At the broadest level, policy makers also have several other options including  a cap on certain types of mitigation and a balanced portfolio of mitigation types. Negotiators to the Kyoto Protocol have agreed to limit the volume of carbon offsets that can be achieved through LULUCF and/or in developing countries. Although these restrictions were not set primarily to address the risk of leakage, limiting the scope and number of projects may effectively keep in check the impact of
projects on local and/or global timber markets respectively, and thus limit market leakage.  Nevertheless, given the level of the cap, it is not clear that it will actually constrain projects, nor market leakage. Combining a “correct” number of avoided deforestation projects with plantations in a balanced portfolio could result in counter-balancing amounts of timber supply restrictions and expansions. However, in practice a balanced portfolio approach would be difficult to accurately implement and would also need to compensate for differences in carbon flux timings between project types and other factors. 
 
Leakage is a significant, but by no means insurmountable, risk for climate change mitigation.  Both project-level and macro-level approaches can effectively manage leakage. Many pilot LULUCF projects are already using many of these tools on the ground, although, not for long enough periods to make robust conclusions. 
 
There is a way to combine project-based calculations and general principles in a way that rewards positive actions in terms of leakage. A decision-tree framework (as proposed by Aukland et al, 2002) for appraising leakage would be able to combine many of the positive aspects of most of the possible tools for addressing leakage. Policy makers can abbreviate leakage management based on first-order principles, project appraisal, monitoring, discounting and other techniques. A decisiontree framework could allow certain types of projects easier certification while placing higher standards on projects that raise ‘red flags”. Doing so would appropriately combine general principles that can be inferred about leakage risks with project-specific evaluations.

More ►
 

June 2012:

Accounting for carbon leakage from REDD+ are current quantification methods suitable?

Focali Brief 2012:01 by Sabine Henders

Carbon leakage refers to the displacement of greenhouse gas emissions from one place to another due to emission reduction activities. It is caused by a direct or indirect shift of emission-intense activities from within to outside an emissions accounting system.