IPCC Fourth Assessment Report: Climate Change 2007
Climate Change 2007: Working Group III: Mitigation of Climate Change

9.6.6.1 Leakage

There is no indication that leakage effects are necessarily higher in forestry than in project activities in other sectors but they can be significant (Chomitz, 2002). Some studies distinguish between primary and secondary effects. A primary effect is defined as resulting from agents that perform land use activities reflected in the baseline. Populations previously active on the project area may shift their activities to other areas. In land protection projects, logging companies may shift operations or buy timber from outside the project area to compensate for reduced supply of the commodity (activity outsourcing). Secondary leakage is not linked to project participants or previous actors on the area. It is often a market effect, where a project increases (by forest plantation) or decreases (deforestation avoidance) wood supply. Quantitative estimates of leakage (Table 9.9) suggest that leakage varies by mitigation activity and region.

The order of magnitude and even the direction of leakage (negative versus positive), however, depend on the project design (Schwarze et al., 2003). Leakage risk is likely to be low if a whole country or sector is involved in the mitigation activity, or if project activities are for subsistence and do not affect timber or other product markets. There are also well-documented methods to minimize leakage of project-based activities. For example, afforestation projects can be combined with biomass energy plants, or they may promote the use of timber as construction material. Fostering agricultural intensification in parallel can minimize negative leakage from increased local land demand. Where a project reduces deforestation, it can also reduce pressure on forest lands, for example, by intensifying the availability of fuel wood from other sources for local communities. Projects can be designed to engage local people formerly responsible for deforestation in alternative income-generating activities (Sohngen and Brown, 2004).

Leakage appears to have a time dimension as well, due to the dynamics of the forest carbon cycle and management (for example, timing of harvest, planting and regrowth, or protection). Analysis in the USA indicates that national afforestation in response to a carbon price of 15 US$/tCO2 would have 39% leakage in the first two decades, but decline to 24% leakage over five to ten decades, due to forest management dynamics (US EPA, 2005).

Table 9.9: Forestry mitigation activity leakage estimates by activity, estimation method and region from the literature

Activity Region Leakage estimation method Estimated leakage rate (% of carbon mitigation) Source 
Afforestation: tropical region estimates 
Afforestation of degraded lands Kolar district, Karnataka, India hypothetical project Household wood demand survey 0.02 Ravindranath, et al., 2007 
Plantations, forest conservation, agro-forestry of degraded lands Magat watershed, Philippines hypothetical project Historical rates of technology adoption 19 – 41 Authors estimates based on Lasco et al., 2007 
Afforestation on small landowner parcels Scolel Té project, Chiapas, Mexico Household wood demand survey De Jong et al., 2007 
(some positive leakage) 
Afforestation degraded uplands Betalghat hypothetical project, Uttaranchal, India Household wood demand survey 10 Hooda et al., 2007 
from fuelwood, fodder 
Afforestation, farm forestry Bazpur hypothetical project, Uttaranchal, India Household wood demand survey 20 Hooda et al., 2007 
from fuelwood, poles 
Afforestation: global and temperate region estimates 
Afforestation (plantation establishment) Global PEM 0.4-15.6 Sedjo and Sohngen, 2000 
Afforestation USA-wide PEM 18-42 Murray et al., 2004 
Afforestation only USA-wide PEM 24 US EPA, 2005 
Afforestation and forest management jointly USA-wide PEM -2.8 a) US EPA, 2005 
Avoided deforestation: tropical region estimates 
Avoided deforestation Bolivia, Noel Kempff project and national PEM 2-38 discounted Sohngen and Brown, 2004 
5-42 undiscounted 
Avoided deforestation and biofuels: temperate region estimates 
Avoided deforestation Northeast USA PEM 41-43 US EPA, 2005 
Avoided deforestation Rest of USA PEM 0-92 US EPA, 2005 
Avoided deforestation Pacific Northwest USA PEM 8-16 US EPA, 2005 
Avoided deforestation (reduced timber sales) Pacific Northwest USA Econometric model 43 West region Wear and Murray, 2004 
58 Continental US 
84 US and Canada 
Biofuel production (short rotation woody crops) USA PEM 0.2 US EPA, 2005 

a) Negative leakage rate means positive leakage; PEM means partial equilibrium model of forest and/or agriculture sector(s).

Source: Sathaye and Andrasko, 2007