Content uploaded by Barbara Haya
Author content
All content in this area was uploaded by Barbara Haya on Jun 28, 2020
Content may be subject to copyright.
1/7
POLICY BRIEF: The California Air Resources Board’s
U.S. Forest offset protocol underestimates leakage
May 7, 2019
Barbara Haya, PhD, Research Fellow, Center for Environmental Public Policy, University of California, Berkeley,
bhaya@berkeley.edu
SUMMARY
Analysis of projects generating 80% of total offset credits issued by the California Air Resources
Board’s (ARB) U.S. Forest offset protocol finds that 82% of these credits likely do not represent
true emissions reductions due to the protocol’s use of lenient leakage accounting methods. The U.S.
Forest protocol has generated 80% of the offset credits in California’s cap-and-trade program. The
total quantity of emissions allowed because of this over-crediting equals approximately 80 million
tons of CO2, which is one third of the total expected effect of California’s cap-and-trade program
during 2021 to 2030 (ARB 2017).
Leakage, in the context of the protocol, occurs when a reduction in timber harvesting at a project site
causes an increase in timber harvesting elsewhere to meet timber demand. The way ARB’s protocol
accounts for leakage when calculating the number of credits awarded has three serious problems.
First, the protocol uses a 20% leakage rate when a rate of 80% or higher is supported by published
studies of leakage rates from reduced timber harvesting in the United States (Gan & McCarl 2007,
Wear & Murray 2004). Using an unsupported low rate results in over-crediting.
Second and more importantly, there is an inconsistency between the timing of when increases in on-
site carbon storage and releases due to leakage are accounted for in the protocol’s methods. Most
improved forest management projects assume and credit a large reduction in timber harvesting in
the first year of the offset project, but deduct the associated leakage over 100 years. This outcome is
physically inconsistent, as it assumes the forest would be harvested in the first year for the purpose
of giving credit but assumes harvesting would be spread out over 100 years for the purpose of
reducing credits to account for leakage. As a result, most forest offset projects begin in greenhouse
gas debt; project landowners generate offset credits that allow emitters in California to emit more
than the state’s emissions cap today, in exchange for promises that their lands will continue to
increase their storage of carbon over 100 years.
Third, it is unclear whether the protocol requires forestland owners to increase carbon stocks to
cover leakage for 25 years or for 100 years. The ambiguity relates to whether forestland owners are
required to continue to maintain on-site growth to cover the impacts of leakage after the end of the
project’s 25-year crediting period. If forestland owners are only required to account for leakage for
25 years, participating projects could result in no net increase in carbon storage over 100 years
compared to the baseline scenario.
The below table presents the actual emissions reductions achieved by projects under the protocol
under different assumptions, reported as proportions of the credits already issued. For example, the
cell on the upper left (100%) represents the assumptions underlying current policy. If these
2/7
assumptions are accurate, then 100% of the credits issued represent true emissions reductions. On
the other hand, if these assumptions are inaccurate, the proportion of credits that represent actual
emissions reductions can be much lower. The cell on the lower right (18%) shows that if the true
leakage rate is 80% and ARB chose to only credit reductions already achieved, rather than reductions
expected in the future, then the real reductions achieved to date by the project add up to only 18%
of the credits issued.
This analysis was performed on all credits generated by 36 compliance forest offset projects through
March 23, 2019. Collectively, these projects generated offset credits equal to 97 million tons of CO2
reductions, which is 80% of the total credits that ARB has issued under its U.S. Forest protocol.
Actual emissions reductions by U.S. Forest offset projects
as percent of credits issued to date
Expected over 100 years
(ARB’s current approach)
Achieved to date
(Recommended approach)
If the true
leakage rate
is:
20%
100%
65%
40%
99%
49%
60%
97%
33%
80%
96%
18%
ARB can avoid the over-crediting discussed here with a few modifications to its protocol. ARB
should (1) apply a leakage rate that is 80% or higher; and (2) determine the net benefits of reduced
harvesting on an annual basis by accounting for both the increased carbon storage on site and the
decreased carbon storage elsewhere due to leakage at the same time. This solution is reflected in the
bottom right cell of the above table (18%).
These changes are needed for the protocol to be in accordance with current law and regulation.
First, given the uncertainty in true leakage rates from reduced timber harvesting within the United
States, using an 80% leakage rate or higher, as is supported by the academic literature, better fulfills
the conservativeness principle laid out in ARB’s cap-and-trade regulations.1 Using low rates that are
not reflected in published literature is unjustified and does not fulfill the conservativeness principle.
Second, generating credits today for expected net reductions over many decades into the future runs
contrary to the goals of California’s Global Warming Solutions Act (AB32), the 2006 law authorizing
California’s cap-and-trade and offsets programs. This law states that for any trade in credits using a
market-based compliance mechanism, the reductions credited should occur “over the same time
period” and be “equivalent in amount to any direct emission reduction required” under California’s
climate change law.2
1 “ ‘Conservative’ means, in the context of offsets, utilizing project baseline assumptions, emission factors,
and methodologies that are more likely than not to understate net GHG reductions or GHG removal
enhancements for an offset project to address uncertainties affecting the calculation or measurement of GHG
reductions or GHG removal enhancements.” California Code of Regulations, title 17, § 95802.
2 California Health & Safety Code § 38562(d)(3).
3/7
DETAILED DISCUSSION
How the U.S. Forest offset protocol works
The large majority of U.S. Forest offset projects credit forestland owners for holding more carbon
on site per acre than they would have in the business-as-usual baseline scenario. Landowners must
commit to maintaining those higher carbon levels for 100 years. Projects can be anywhere in the
United States, and to date, approximately 20% of credits generated have been from projects in
California, and 80% have been from projects elsewhere in the United States.
Most of these improved forest management projects define a business-as-usual baseline scenario
that involves aggressive timber harvesting that brings on-site carbon storage close to the average per
acre for forests in their region. The assumption is that these offset projects maintain higher on-site
carbon stocks by reducing timber harvesting.
In the first year of an improved forest management offset project, the landowner earns offset credits
for the amount of carbon on their land above the business-as-usual baseline scenario minus two
factors. First, estimates of carbon released due to leakage are deducted. Second, not all loss of on-
site carbon is released into the atmosphere. The protocol accounts for the portion of harvested
timber that remains long-term in wood products like in houses and furniture and buried in landfills,
which would be reduced if total timber harvesting is reduced by the project. Each subsequent year,
the landowner is credited for any incremental increase in carbon sequestration on the participating
lands as trees grow and sequester more carbon, minus the same two factors.
Leakage rate
ARB’s U.S. Forest offset protocol uses a 20% leakage rate. A 20% leakage rate means that 20% of
the reduction in timber harvesting caused an offset project is replaced by an increase in harvesting
on other forestlands. The other 80% of the reduction is assumed not to be replaced and simply
represents a decrease in timber use (i.e., fewer houses built, less paper produced, etc.)
Published literature suggests the leakage rate from reduced timber harvesting in the United States is
at least 80%. Using a computable general equilibrium model, Gan & McCarl (2007) estimate that if
timber production were reduced in the United States, 77% of that that timber harvesting would be
displaced to other countries. Wear & Murray (2004) use econometric modeling to trace the effects of
reductions in federal timber sales in the western United States in the late 1980s through the 1990s.
They estimate that 84% of the reduced timber production was displaced to elsewhere within North
America. Both articles underrepresent total leakage from conservation on U.S. forestlands. The
former only estimates international leakage, ignoring leakage that might occur among forestland
within the United States; the latter only estimates leakage in North America, ignoring leakage that
could occur elsewhere. The existing academic literature on leakage rates from reduced forest
harvesting does not support a 20% leakage rate. A conservative approach to addressing uncertainty
in the true leakage rate would apply a leakage rate that is at least 80%.
The Climate Action Reserve, which developed the original U.S. Forest offset protocol on which
ARB based its own protocol, revised its leakage rate from 20% to a sliding scale up to 80%,
4/7
depending on the amount of timber harvesting performed by the offset project itself. Under this
protocol, an 80% leakage rate is applied to offset projects that do not harvest at all.
The timing issue explained
As is typically done with offset projects, emissions reductions are estimated against a baseline
scenario representing what would likely have happened without the offset program. Almost all ARB
improved forest management offset projects define baseline scenarios that are well below their
actual carbon stocks in their first year. On average across all projects analyzed, these baselines equal
70% of current carbon stocks. This means that in the first year of a project, the land owner is issued
a quantity of credits equal to, on average, around 30% of the carbon stocks on their project lands,
adjusted downward to account for leakage and any reduction in carbon held long-term in harvested
wood products and landfills.
To create a baseline, the landowner models the carbon stocks and fluxes associated with a 100-year
timber harvest scenario that reflects the harvesting expected to take place without the financial
incentives from the offset program. The modeled scenario should be financially feasible and fulfill
all legal and contractual obligations. In order for most projects to earn credits under the protocol,
the calculated average carbon stocks in the baseline scenario over 100-years should be no less than
that of the average forestlands for the project’s region and forest type.
This modeled scenario is then abstracted into two key parameters used to calculate emissions
reduced and credits generated by the project. Baseline on-site carbon storage and harvesting rates are
assumed to equal the average values generated by the modeled scenario over 100 years. This
simplified baseline is treated as equivalent, in terms of carbon accounting, to the range of financially
feasible timber harvest scenarios that could have happened without the offset program. Flat average
baseline values have the advantage of not requiring the landowner to calculate year-to-year increases
in carbon storage against the harvest and growth cycles in one specific baseline management regime
for each of 100 years. But this approach has one important disadvantage—flat average baseline
values for carbon storage and harvest rates are internally contradictory and physically impossible.
The figure below presents an example of a modeled harvesting scenario used to define the baseline
for one large offset project – ACR360, a half million acre project in southern Alaska. The curved
dotted line is the modeled business-as-usual scenario for above-ground standing live carbon stocks.
The straight dotted line is the baseline used to generate credits, which is the average above-ground
standing live carbon stock in the 100-year modeled scenario. The solid line is the actual carbon
storage on the project lands at the start of the project.
This simplified baseline scenario suggests that, if the project were not earning offset credits, its lands
would be harvested to baseline levels in year 1 and maintained at those carbon stocking levels for
100 years. However, contradicting this assumption, the baseline also assumes that a constant
quantity of timber is harvested each year over the project life, equal to the average rate over the 100-
year modeled scenario. This second assumption is used to calculate leakage.
These two assumptions are contradictory because it is not possible for both carbon storage and
harvesting to simultaneously remain at their respective average values over the project life. Carbon
storage and harvesting rates are correlated with one another, and inextricably tied to the actual net
growth rate of the project forest. If carbon storage is assumed to drop to the baseline in year 1, that
5/7
would happen because of a large amount of timber harvesting. If the harvesting rate is assumed to
be constant over 100 years, however, then the carbon storage on the land will also decrease slowly,
rather than abruptly in year 1. By mixing these two assumptions into a physically impossible baseline
scenario, the protocol maximizes credits generated without reflecting the actual rate at which
emissions to the atmosphere are avoided. The protocol calculates gains in carbon against the
baseline using the first assumption, and losses in carbon from leakage using the second assumption.
As a result, credit generation is frontloaded, and landowners need to continue to increase net carbon
storage for decades to make up for the leakage effects associated the reduced harvesting credited at
the start of the projects.
Baseline carbon stocks for Finite Carbon – Ahtna Native Improved Forest Management
offset project
From: ACR360 “Finite Carbon – Ahtna Native Alaskan IFM” Version 1.3, Attachments G and H: Baseline
Carbon Stocks, Submittal Date: 1/19/2018
This over-crediting allows emitters in California to emit more than the state’s emissions cap today in
exchange for promises of forest carbon sequestration over 100 years to cover leakage from the start
of the project. This is problematic for several reasons. First, emissions today are not equivalent to
reductions decades from now given the urgency of climate change mitigation to avoid tipping
points. California is designing its cap-and-trade and offset programs as models for other
jurisdictions. If California exports a model that trades emissions today with reductions decades from
now, California would promote a form of climate policy that fails to reduce emissions in these
immediate critical years. Second, these promises can be difficult to keep since productivity slows in
ageing forests (Gray et al 2016) and as forests respond to a warming climate. On project lands with
less harvesting, fewer older trees will be replaced with younger trees, and the average tree age will
increase over the 100 years of the project.
ACR360 generated close to 15 million offset credits in its first year, equal to more than 60% of the
expected average annual effect of California’s cap-and-trade program on emissions during 2021-
2030.
6/7
The 25 year versus 100 year issue explained
If forestland owners are required to increase carbon to cover leakage for 100 years, then there would
be no over-crediting over 100 years of the project. Over-crediting in the early years of the project
would slowly be compensated as leakage is deducted each year for the project life.
However, it is unclear whether the protocol requires forestland owners to account for the emissions
from leakage for 25 or for 100 years. The crediting period of a U.S. Forest offset project is 25 years.
After the end of each 25-year crediting period, landowners can choose to renew their offset project
for another 25 years but are not required to do so. For each year of a crediting period, landowners
must report the net impact of the project on emissions taking into account any change in on-site
carbon storage, and any releases due to leakage or reductions in carbon held long-term in harvested
wood products and in landfills. If the net impact of the project in any year is negative, a reversal is
understood to have occurred. The carbon reductions that were previously credited and later released
must be replaced with additional procurement of allowance or offset credits.
How a reversal is defined after the last year of crediting is unclear in the protocol. Following the last
year of crediting, forestland owners are required to maintain the credited on-site carbon storage for
another 100 years. It is unclear if they are also required to ensure their forestland continues to grow
to cover off-site releases due to leakage and due to reductions in carbon held long-term in harvested
wood projects and landfills.
If forestland owners are only required to account for leakage for 25 years, crediting for reduced
harvesting in the first year of the project will be awarded in full, while potentially, as low as only 1%
of the leakage associated with that reduced harvest is deducted each year for only 25 years. It would
be possible for participating projects to result in a net decrease in carbon storage over 100 years
compared to the baseline.3
Methods
Landowners report how they calculate their requested credit issuance in Offset Project Data Reports
(OPDRs) based on instructions laid out in the protocol. These reports are made public through the
offset registries. We reproduce these calculations for all credits issued to 36 projects as of March 23,
2019. We use data provided by the landowner in their OPDRs and supplemental materials, and
adjust the projects’ assumptions for leakage and the timing of harvesting in the baseline to
investigate the quantity of over-crediting.
Adjusted leakage rate
Using data reported in the OPDRs, we reproduce the calculations of leakage (also called secondary
effects), carbon in harvested wood products and landfills (HWP&L), and total reductions achieved
using leakage rates of 40%, 60%, and 80% instead of 20%.
3 Please see public comments submitted to ARB on May 10, 2018, Comments on proposed cap-and-trade regulatory
amendments, for a more detailed discussion of this need to clarify and revise how the protocol defines a
reversal after the last year of credit issuance, found at http://bhaya.berkeley.edu.
7/7
Adjusted timing of baseline harvesting
We recalculate the credits that would have been generated if the protocol’s leakage calculations
matched its assumption that timber is harvested in year 1 of the baseline scenario to bring carbon
storage down to baseline levels, and continues to be harvested at smaller rates needed to maintain
the baseline carbon storage level for one hundred years.
We do this in the following manner:
First, the baseline harvesting level prior to delivery to the mill (PDM) in the first year of the project
is calculated as the difference between standing live carbon in the project compared to the baseline.
Second, we calculate the baseline carbon in trees harvested in years 2 to 100 so that the sum of the
baseline PDM over 100 years is the same as the sum using ARB’s current methods. We calculate the
baseline PDM in years 2 through 100 (99 years) as:
PDMannual after year 1 = (PDMtotal – PDMyear 1) / 99
Third, we recalculate the carbon in baseline HWP&L in a similar manner, by:
a) using the ratio of HWP&L to PDM in year 1 of the baseline in the OPDR to recalculate carbon
in HWP&L in year 1 of the baseline for the revised PDM value;
b) calculating carbon in HWP&L in years 2 through 100 using the same process as for timber
harvesting, so that the sum of carbon in HWP&L over 100 years of the baseline is the same in our
estimates as it is in ARB’s current estimates over the project life;
Fourth, we recalculate emissions reductions from the project using these revised leakage and carbon
in HWP&L figures, and otherwise following the methods defined by the protocol.
When baseline or project PDM figures are missing from any of the OPDRs, we calculate the missing
PDMs mathematically from other reported figures when possible, and apply the following
assumptions when needed:
§ The ratios of carbon in HWP&L to PDM remain the same across reporting periods.
§ When the first reporting period does not equal exactly one year, the PDM in the first year is a
prorated amount, reflecting what most projects with at least two reporting periods have done.
§ The ratio of carbon in HWP&L to PDM is the same in both the baseline and project scenarios.
Other than the changes and assumptions described above, we repeat the methods used in the
OPDRs to re-estimate emissions reduced and credits generated.
REFERENCES:
ARB. 2017. California Air Resources Board, California’s 2017 Climate Change Scoping Plan.
Sacramento.
Gan, J. & B.A. McCarl. 2007. Measuring transnational leakage of forest conservation. Ecological
Economics, 64(2), 423-432.
Gray A.N., Whittier T.R., Harmon M.E. 2016. Carbon stocks and accumulation rates in Pacific
Northwest forests: role of stand age, plant community, and productivity. Ecosphere, 7(1).
Wear, D.N. & B.C. Murray. 2004. Federal timber restrictions, interregional spillovers, and the impact
on US softwood markets. Journal of Environmental Economics and Management, 47(2), 307-330.
