Transaction Costs of Project-Based Carbon Sequestration:

Preliminary Evidence from Forest Sector

Camille Antinori and Jayant SathayeCamille Antinori and Jayant SathayeLawrence Berkeley National Laboratory, Berkeley CALawrence Berkeley National Laboratory, Berkeley CA

andand

Ken Andrasko (Ken Andrasko (Presenter)Presenter)

Office of Atmospheric ProgramsOffice of Atmospheric Programs

U.S. Environmental Protection Agency, Washington, DCU.S. Environmental Protection Agency, Washington, DC

USDA Symposium on GHGs & Sequestration in Agriculture & USDA Symposium on GHGs & Sequestration in Agriculture & Forestry. Baltimore, MD 3/23/05Forestry. Baltimore, MD 3/23/05

Improving Information & Solving Technical Issues Are Critical to Ag & Forestry Mitigation

U.S. climate debate: a) science of climate change, b) stimulating technologies; c) sectors and scale of action?; d) rules & guidance for voluntary reporting.

Global sinks mitigation potential large, but uncertain: IPCC TAR: afforestation potential 200 – 584 MMTC/y in 2010, & avoided deforestation potential 1,790 MMTC/y in 2010.

CGE US and global climate economic models need improved representation of sinks.

But: Unresolved sinks technical issues have limited eligibility and investment: baseline setting, additionality, leakage, duration of GHG benefits, costs.

Voluntary registries could stimulate investment, if reporting is credible.

Solving key issues determines how much of biophysical potential is realized, and by whom:– - bundling heterogeneous land parcels, baselines, leakage,

MMV– - costs, inc. transaction costs

Effects of economic, environmental and social-institutional factors on mitigation costs and

potential (based on Raupach et al. 2004)

Mitigation Potential = Carbon Sequestered or GHG emissions avoided, as a fraction of technical potential mitigation

Cos

t of c

arb

on

($/tC

eq)

Technical Potential

Baseline Potential

Environmental factors

Social and institutional factors

Economic factors

Sustainably Achievable Potential

0 1

Uptake proportion at given cost

Mitigation Potential = Carbon Sequestered or GHG emissions avoided, as a fraction of technical potential mitigation

Cos

t of c

arb

on

($/tC

eq)

Technical Potential

Baseline Potential

Environmental factors

Social and institutional factors

Economic factors

Sustainably Achievable Potential

0 1

Uptake proportion at given cost

Market Potential

Economic Potential

Socio-cultural Potential

Technical Potential

Transaction costs

Can We Estimate Competitive Potential for U.S. Ag & Forest Options: by Region, by C Price, by Reporting

Rules?

0

100

200

300

400

500

0 20 40 60 80 100 120 140 160

C p

rice

($/

tCeq

)

Soil carbon sequestration (mmtce)

BiophysicalPotential

EconomicPotential

Land UseCompetitivePotential

Adapted From: B. McCarl and U. Schneider, presented 2001 EPA-USDA Forestry and Agriculture Greenhouse Gas Modeling Forum. Biophysical estimate from Lal et al., 1998 (mid-point)

Example: U.S. Ag Soil Potential:

Transaction Transaction

costs ??costs ??

Scenario 1 ($5 + 5% /year) --Land Use Change by Activity for All Regions

(Cumulative to year reported)

-800

-600

-400

-200

0

200

400

600

2000

2010

2020

2030

2040

2050

2060

2070

2080

2090

2100M

ha Short Rotation

Forestation

Long RotationForestation

Deforestation

Net Forest AreaGained

Note: net forest area declines to 2080, then

increases

Few Ag & Forest Models include Transaction Costs: Few Ag & Forest Models include Transaction Costs:

e.g., GCOMAP, scenario for EMF 21 Exercisee.g., GCOMAP, scenario for EMF 21 Exercise

Do Transaction Costs Change Options??Do Transaction Costs Change Options??

NoneNoneNoneNonenDeforestation rate (Kha/yr)

918.96.59.9mMean annual increment (t biomass/ha/yr)

34144215lRotation Period (Yrs)

112525kBase-year Vegetation C-stock(t C/ha)

346.6

307.4

614814jPlanting rate in base year (Kha/yr)

16.815.220.527.2iArea planted so far (Mha)

11.610.227.135.9hMax area dedicated to SR & LR afforestation (Mha)

0.5%0.5%1%5.%g% of timber exported

915483450250fTimber international market price ($/m3)

421123200150eTimber domestic market price ($/m3)

40494060dHarvesting and Transport Cost($/m3)

22312cMonitoring Cost ($/ha/yr)

1919312bRecurrent Cost ($/ha/yr)

778340245245aInitial Cost (Land cost + Establishment Cost) ($/ha)

LRSRLRSR Variable Name

India China Notes 

Chinaa, b, c: Weighted average for NE, SE & SW regions (Deying Xu et al, 2001).d: 40% of the product price goes towards harvesting and transport (Xu & Zhang, 2002).e, f: FAO Forest Products Year book, 2000 and Xu et al, 2001.g: Average 1995-2000, FAO 2000, Forest Products Yearbook.h: Long term plans to add 117 mi ha to Chinas Forestry, of which 63 mi ha will be for timber production. (PEER – Long term nationwide environmental restoration. From I – IV NFRI planting schedules adjusted for survival rate, minus the 11.5 mi ha of non-timber forests e.g. bamboos and orchards.i: FAO, 2000, GFPOSj: Current stated planting rate adjusted for survival rate of 80% (Zang, 2001).k: Original COMAP runs for wastelands.l, m: Weighted average for individual spp, with LR/SR decided on the basis of rotation age. Indiaa: Investment Cost + Discounted value of costs in yrs 1 & 2 – Comap.b: Comap 500 Rs/ha/yr @42Rs/$, Source: Ravindranath, 2001.c: Comap 100 Rs/ha/yr @42Rs/$, 200 rate, Source: Ravindranath, 2001.d: 40% of millsite domestic log price for SR & 10% of LR price due to smaller size and price differential while cost of operations, machinery and fuel is still as high as that of LR.e: LR Average Teak, Sal & Shisan, girth 91-120 cm, SR Average Sal, Eucs and Populus in Agra, Chandagar and Ambala.f: LR Average for Teak, Sal and Shisan Girth 151cm, SR Shorea Robusta at Agra, Barely & Shahampur: Source FRI Market Report, 2000.g: Average RWE for 1995-2000. India exports 3% of WBP and 1% of Pulp output.h: National Action Plan propose to increase area under Forest Cover by 33% (21.8 mi ha) over 20 yrs.i: Sum of all areas under forest plantations up to 2000.j: Same target as NFAP of 1.09 mi ha/yr, adjusted by survival rate of 60% and existing LR: SR ratios.k: Mostly, shrub and grass wastelands (Ravindranath, 2001).l, m: Average for all major spp, weighted by area under each. : Proportions based on individual species Rotation ages in existing plantations.

GCOMAP MODEL: Key Data Inputs: Monetary and Biophysical Data

Transaction Cost Components: Organizational, non-production Project search costs: Identification and stakeholder consultation

- May be spread over many projects

Feasibility studies: engineering, economic, and environmental assessments- GHG Baseline estimation and establishing additionality

Negotiations: obtaining permits, negotiating and enforcing contracts for fuel supply, arranging financing

- Marketing GHG credits, carbon contracting and enforcement

Insurance – project risk insurance– GHG credit insurance (Difficult to get or too expensive today)

Regulatory approval (GHG): Project validation and government review – (May include both domestic and international validation costs)

Monitoring and verification (GHG) – During project implementation– Monitoring including equipment cost, verification and certification (Spread

over many years of project life)

Source: Based on OECD, 1997; Stratus Consulting, report to EPA, 2003; analysis work supported by Economic Analysis Branch, Climate Change Division, US EPA, 2004-05.

Transaction Costs: Affect Price and Quantity of Emissions Reductions

Demand

Supply

Supply with transaction costs

Cost

Emissions Reduction from Climate Mitigation Projects

Transaction cost range: Vary by:- Project size? - Type of project?- Nascent and mature markets? - Region?

This Study, in Progress : Methods: Collect Original Data on Forest Mitigation Projects Forestry Project Identification: Actual projects, if possible

– EPA/CCD cooperation w/ TNC via Stratus Consulting data collection

– EPA cooperation w/ Climate Trust – LBNL cooperation w/ NR Canada & Indian Inst. Sc.

Thanks to: TNC, Climate Trust, NR Canada, Indian Inst. Sc.

Data: 11 projects (9 actual, 2 feasibility studies):

– The Nature Conservancy: Bolivia, Belize, Brazil 5 projects

– Oregon Climate Trust: Pacific NW, Ecuador 3

– Natural Resources Canada: Chile 1

– Indian Institute of Science: India (feasibility) 2

Transaction cost analysis work supported by Climate Analysis Branch (TNC) and Economic Analysis Branch (rest), Climate Change Division, US EPA, 2004-05.

This Study: Methods: 2 Literature review to identify Transaction Cost components

Built spreadsheet model for comparability (Stratus; LBNL)

Defined 2 scenarios of project implementation & transaction costs, with and w/out insurance:

– Nascent Market Scenario: Project developers have minimal experience in

baselines, leakage assessment; Few GHG accounting tools; Limited stringency of or guidelines for govt. approval

– Mature Market Scenario: Project developers experienced in C market; GHG accounting, baseline etc. tools avail., inc.

regional baselines or benchmarks; Govt. guidance and requirements for MMV extensive

Transaction cost analysis work supported by Climate Analysis Branch (TNC) and Economic Analysis Branch (rest),

Climate Change Division, US EPA, 2004-05.

Descriptions of 5 TNC Belize, Bolivia and Brazil Projects

Project NameRio Bravo,

Belize

Noel Kempff Climate Action Project, Bolivia

Guaraquecaba Climate Action Project, Brazil

Atlantic Rainforest

Restoration Project, Brazil

Antonina Pilot

Reforestation Project,

Brazil

Project type (AIJ Category)

Forest preservation

Forest preservation Forest restoration Forest restoration Forest restoration

Host CountryBelize Bolivia Brazil Brazil Brazil

Project developer TNC, PfB*TNC, FAN,

GOB*TNC, SPVS* TNC, SPVS* TNC, SPVS*

Carbon credit buyersUS utility

companiesAEP, Pacificorp, BP

AmocoAEP, CCX GM Texaco

Carbon sellersPrivate

landownersBolivian Government GOB GOB GOB

Institutional status USIJI, AIJ USIJI, AIJ Voluntary Voluntary Voluntary

Project Size 50,328 ha 634,286 ha 8,100 ha 12,000 ha 1,000 ha

Non-GHG goals No No No No No

Project duration (years) 40 30 40 40 40

Initial Start Date 1996 1997 2000 2001 2001

Est. total carbon benefit — low tons (C)

2.4 million 6 million 339,600 660,000 106,000

Est. total carbon benefit — high (tons C) 2.4 million 8 million 566,000 1.1 million 162,000

Frequency of monitoring/ verification 3 years Annual 5 years 5 years 5 years

*Abbreviations: PfB: Programme for Belize FAN: Fundacion de Amigos de la NaturalezaSPVS: Sociedade de Pesquisa em Vida Selvagem e Educacao Ambiental (Society for Wildlife Research and Environmental Education) GOB: Government of Brazil Transaction cost analysis work supported by Economic Analysis Branch, Climate Change Division, US EPA.

Descriptions of India, Chile, Ecuador and US Projects

Project Name

Village Based Forest

Restoration, India

Integrated Sink Enhancement &

Biodiv. Conservatn

India

Seedling Inoculation,

Chile

Rainforest Reforestatio

n Project, Ecuador

Pacific Northwest

Forest Preservation

Project

Riparian Zone

Reforestation Project, Oregon

Project type (AIJ Category)

Forest restoration Forest restoration Afforestation Forest restorationForest

preservationForest restoration

Host Country India India Chile Ecuador USA USA

Project developerCanadian

Government

Karnataka Forest

Department

Mikro-Tek, Instituto Forestal (Chile)

Climate Trust

Climate Trust

Climate Trust

Carbon credit buyers Open market Open market Open marketClimate Trust

partnersClimate Trust

partnersClimate Trust

partners

Carbon sellers Local residents Local residents Local landowners

Jatun Sacha Foundation, Conservation International

Lummi Indian tribe Private landowners

Institutional status None NoneCanada’s Voluntary

Challenge & RegistryOregon CO2

programOregon CO2

programOregon CO2

program

Project Size 2000 ha 13,000 ha 5000 ha 275 ha 700 ha 830 haNon-GHG goals Yes No No No No No

Project duration (years) 10 5 12 99 100 52

Initial Start Date 2002 2002 1999 2002 2002 2001

Est. total carbon benefit — low tons (C)

100,000 358,800 600,000 16059 95446 63631

Est. total carbon benefit — high (tons C)

Same Same 1,440,000 16059 95446 63631

Frequency of monitoring/ verification

Annual Annual Annual

Transaction cost analysis work supported by Economic Analysis Branch, Climate Change Division, US EPA.

Transaction Cost - Forestry Projects

$-

$1.00

$2.00

$3.00

$4.00

$5.00

10,000 100,000 1,000,000 10,000,000

t C over project life (log scale)

$ / t

C

Source: Data collated by LBNL and EPA from 3 US and 8 international projects of varying sizes

1-Ecuador Reforestation (CT)

2-Riparian Reforestation (CT)

3- Northw est Forest Preservation (CT)

4 - Adilibad Village Forest Restoration (India)

5 - Antonina Pilot Reforestation (TNC)

6 - Guaraquecaba (TNC)

7-Integrated Carbon and Biodiversity Project (India)

8 - Seedling inoculation Chile (Microtek)

9 - Atlantic Rainforest Restoration (TNC)

10 -Rio Bravo Sequestration (TNC)

11 - Noel Kempff (TNC)

1

3

2

4

56

78

9

10 11

TNC Brazil

Transaction cost analysis work supported by Economic Analysis Branch, Climate Change Division, US EPA.

TNC large, early projects

Preliminary Results: Project Transaction Cost by Project Preliminary Results: Project Transaction Cost by Project Lifetime Carbon BenefitsLifetime Carbon Benefits

USUS

Preliminary Results:Preliminary Results: Trans. Costs for Study Projects are 6% - 18% of Total Costs (n=5 projects)

Transaction Costs for 5 Nature Conservancy Forest Projects, Latin America (Stratus Consulting, for EPA, 2003)

Project Rio

Bravo Noel

Kempff AEP GM Texaco

Hectares 50,328 634,286 8,100 12,000 1,000

Est. C benefit (000 tons) 2,400

6,000 – 8,000 340 - 566

660 – 1,110

106 – 162

Total TR Costs ($ million) $0.43 $2.1 $0.84 $1.35 $0.34

Discounted TR Costs as % of discounted project costs 8.8 % 17.6 % 10.1 % 7.7 % 6.0 %

Preliminary Findings: Transaction Cost by Component, Mature Market: Low & Hi C (with Insurance), 11 Projects

0

0.5

1

1.5

2

2.5

Mature,Low C

Mature,High C

Search

Feasibility

Negotiation

Insurance

Monitoring

Regulatory

Total w/ Insur.

Total No Insur.

TR Cost, $ / tCTR Cost, $ / tC

TR Cost ComponentsTR Cost Components

Transaction cost analysis work supported by Economic Analysis Branch, Climate Change Division, US EPA.

Preliminary Findings: Preliminary Findings: 1) Transaction Costs (average) Decline with Project Size in GHG 1) Transaction Costs (average) Decline with Project Size in GHG

Tonnes. Tonnes.

2) Mature Scenario Costs < Nascent Market. 2) Mature Scenario Costs < Nascent Market. Transaction Costs

(Using low or single carbon estimate)

y = 2515.4x-0.615

R2 = 0.5993

y = 890.62x-0.507

R2 = 0.6764

y = 709.34x-0.5289

R2 = 0.4793

0.00

2.00

4.00

6.00

8.00

10.00

12.00

10,000 100,000 1,000,000 10,000,000

t C over project life

$ p

er t

C

Nascent Market (With Insurance)

Nascent Market (Without Insurance)

Mature Market (Without Insurance)

supported by Economic Analysis Branch, Climate Change Division, US EPA

Projects > 1 million t C Projects > 1 million t C over lifetimeover lifetime

have TR costs < $0.50/ t Chave TR costs < $0.50/ t C

MatureMature

NascentNascent

Comparison of Estimated Transaction Costs from Other Studies to This StudyReport Project Type TC Estimates

(2002 US dollars)

Fichtner et al., 2003

AIJ energy efficiency, renewable energy, forestry, afforestation,

agriculture

$0.05-261/tC

(no average)

Michaelowa et al., 2003

JI energy: GasJI energy: Biomass

$2.4 - $5.6/tC*$3.2 - 8.8/tC*

Woerdman, 2001 AIJ /JI energy sectorAIJ/JI industrial sector

12 - 19 % of initial investment15 - 30 % of initial investment

Climate Trust, 2004

Forestry 7.6 - 34.2 %

This study

(note: cost varies by project size, &Scenario)

Forestry Nascent Mature

Low $2.58 /tC

$2.18 /tC

High $1.06 /tC $1.26 /tC

*Assumes Swedish krona exchange rate of 8 krona per US dollar

Average Average Values:Values:

Preliminary Findings (work in progress): 1 Methods:

– Minimal data available: N = 11, w/ 2 only feas. studies– Potential self-selection & reporting biases of early-

actor projects– Nascent market scenario ~ current conditions.– Mature market: only a scenario, using expert

judgment– Study offers Comparability w/in dataset: Used

consistent definitions of transaction cost categories: unique in literature. Possibly not comparable w/ other studies.

Scope of this study: Include tangible organization-type costs. - But valuation of unforeseeable or intangible costs not included: (eg, enforcement costs)

Preliminary Findings (work in progress): 2 Economies of scale:

– Find returns-to-scale phenomenon for transaction costs: Larger projects have lower transaction costs per ton of carbon.

– Consistent w/ Michaelowa et al. 2003 finding– Largely due to monitoring costs.

– Differ from Michaelowa: we find search, regulatory, validation and approval costs show nonlinear variation with increasing project size. » (Michaelowa found these fixed costs across project

size. )

Preliminary Findings (work in progress): 2 Plotted project size vs. $/tC of transaction costs, & fit

trendline of average total transaction costs per project. – Costs rise steeply for projects 100,000 TC size or smaller. – Pattern holds for nascent & mature markets.– Mature market: most transaction cost categories flatten

at 2.5 million TC scale. – Data inadequate for econometric approach.

Costs lower in mature vs. nascent market: – True for all costs on average, except insurance costs.– Standardization is expected to lower trans. costs:

- for calculating baseline and leakage, and carrying out monitoring and verification.

Insurance as Trans. Cost: – Significant cost if insurance not available & forestry

projects must self-insure (with-insurance scenarios)– Suggests development of insurance markets, or risk

reduction via portfolios, could lower transaction costs

Preliminary Findings (work in progress): 3 Forestry appears low-cost investment in meeting

GHG targets: $1.06 - $2.58 / tC (with insurance)– Forestry has low transaction costs vs. Micheolowa

gas and biomass projects ($2 to $8 per tC).

Trans. Costs likely to vary across GHG program requirements (eg, 1605b, CDM, CCX): – not assessed yet in literature

Potential Next Steps in Analysis:– Increase n & universe of study: add c. 20 projects

?»Add ag sector, biofuels, & more U.S. projects

– Sensitivity case: different GHG program requirements

– Compare to energy and other sector project trans. costs:»EPA study on this in development (LBNL)