Who bears the cost of forest conservation under REDD+? Mahesh Poudyal, James Gibbons, Sarobidy Rakotonarivo, Neal Hockley, Julia Jones (Bangor University)

Bruno Ramamonjisoa, Alexandra Rasoamanana, Rina Mandimbiniaina (ESSA, University of Antananarivo)

ESPA Annual Science Meeting, 27 November 2014

Ultimate aim is to explore how

international payment for

global ecosystem services (focusing on REDD+)

can best contribute to poverty alleviation

2

Corridor Ankeniheny-Zahamena (CAZ), Madagascar

3

Carbon

Biodiversity

Water

Corridor Ankeniheny-Zahamena (CAZ), Madagascar

4

People

Livelihoods

Poverty

WP 1: Developing sampling platform & scenarios

WP 6: local welfare impacts of alternative PES approaches

WP 5: on biodiversity

WP 4: on carbon storage/sequestration

WP 3: on wild harvested products

WP 2: on hydrological benefits

WP 11: Linkages to local, national and global stakeholders

WP 12: Project coordination & interdisciplinary integration

WP 8: Institutional analysis

WP 10: Synthesis

WP 9: Spatial and temporal trade offs

WP 7: Evaluation of realised LUCs from existing scheme

P4GES Project

IMPACTS OF ALTERNATIVE PES APPROACHES

P4GES Work Package 6: To estimate the

magnitude and distribution of net

welfare impacts of alternative PES

approaches at local scales (including evaluation of impacts of the existing Coridor

Ankeniheny-Zahamena ‘CAZ’ REDD+ scheme)

6

Corridor Ankeniheny-Zahamena (CAZ), Madagascar

• 371,000 ha

• 25 communes

• Population ~ 65000

• REDD+ Pilot Project

• Environmental &

Social Safeguards

Assessment to identify

Persons Affected by

the Project (PAPs)

7

▪World Bank has had social safeguards (to

identify and manage social risks) in place for

about 20 years

▪Currently undergoing consultation on how

these can be improved and strengthened

http://go.worldbank.org/WTA1ODE7T0

9

Environmental & Social Standard 5: Land Acquisition,

Restrictions on Land Use and Involuntary Resettlement

“When land acquisition or restrictions on land use cannot

be avoided, the Borrower will offer affected persons

compensation at replacement cost, and other assistance

as may be necessary to help them improve or at least

restore their standards of living or livelihoods”

10

CAZ REDD+ project aims to generate carbon credits by reducing deforestation – main driver of which is tavy (slash-and-burn agriculture)

Therefore project success depends on economic displacement of people from livelihoods based on tavy

11

Plan to ensure environmental & social safeguards are

met in CAZ project was published in 2012

The criteria used for identification of Persons Affected

by the Project were:

1) Live around the proposed protected area

2) Directly use natural resources

3) Use natural resources within the ‘core’ of the

protected area

Corridor Ankeniheny-Zahamena (CAZ), Madagascar: Safeguards Assessment

March-May 2010

▪ 114 fokontany assessed –

463 villages

▪ 12000+ households

surveyed

▪ Public consultations

household surveys

12

Corridor Ankeniheny-Zahamena (CAZ), Madagascar: Safeguards Assessment

• Persons Affected by

Project (PAPs) identified

in 33 fokontany

• 2500 households initially

identified as PAPs

• Around 1800

households being

compensated

• PAPs receiving micro-

projects at HH level

13

The identification of ‘Persons Affected by Project’ (PAPs) in the CAZ pilot REDD+ Project

1. Identification of fokontany (& villages) for

safeguards

2. Identification of households for safeguards

14

What are the characteristics of households identified as eligible to receive social safeguards?

Early results based on field work in one of the

safeguards fokontany

15

Study Site: Ampahitra fokontany with Safeguards

16

Source: Google Maps

• Very high deforestation from 2005-2010

• 77 households identified as PAPs

Conducting Household Surveys in Ampahitra

▪ Large fokontany with villages and households

spread over difficult terrain

▪ Good sampling frame essential to be able to get

representative random sample of households

17

Challenges:

▪ Poor quality of data (e.g. many villages not on the map)

▪ Many scattered households (shift with season)

▪ Many households are not registered in any village

(‘unofficial’ households)

18

Developing the sampling frame…approximately 33%

of total time for survey

19

Fokontany Level

Village Level

Hamlets level

Collect local available information on villages (sketch map)

Collect information on households and hamlets (sketch map & GPS)

Visited hamlets in person to cross check information (GPS), and map HHs location (sketch map & GPS)

20

Household Surveys in Ampahitra

▪ Mapped 468 households, interviewed 203 (stratified

by village, randomly sampled)

▪ 40 of these surveyed households had been

identified as PAPs, with most having received

safeguards project during July/August 2014.

21

Comparing PAP & non-PAP households

▪ Expectation: HHs identified as PAPs would own more

tavy land, be more dependent on wild-harvested

products, be more recently established, live further

from the fokontany centre

▪ We also included variables such as education of HH

head, livestock holding, food security, membership in

COBA to check whether wealth, social & political

capital increases chances of being identified as PAP

▪ We used a binomial GLM to explore which variables

predict whether a HH is identified as a PAP

22

23

Which factors influence the likelihood of being identified as a PAP?

Factors make HH more likely to be identified as PAP

Factors with no significant effect on PAP identification

24

Expectation We found

PAPs would live further from the

fokontany centre (i.e., closer to the forest) [+]

No significant effect (but PAP likely to

live closer to fokontany centre) [-]

PAPs would be more recently

established HHs [-]

Longer established HHs more likely to

be PAPs [+]

PAPs would own more tavy land

[+]

No significant effect (but –ve association) [-]

PAPs would be more dependent

on forest products [+]

No effect [-]

PAPs would be among the poorest (less food secure, low

livestock holding) [-]

No effect in terms of wealth (livestock holding) BUT more food

secure HHs likely to be PAPs [+]

PAP HH heads likely to be literate

[+]

No effect [+]

PAP HHs likely to be associated

with COBA (social & political

capital, access to information) [+]

COBA associated HHs significantly

more likely to be identified as PAPs

[+]

Key Findings

• There is some evidence of

local elite capture in the

identification of the PAPs.

• Poorest households likely to

bear the cost of forest use

restrictions, while potentially

being left out from safeguards

compensation

• Self-identification through

rapid assessment ≠ genuine

PAP identification

25

Next steps

▪ Qualitative work to complement quantitative findings

▪ Continue field work in 3 other ‘in-depth’ sites will allow us to estimate the opportunity cost of conservation restrictions more widely around CAZ

▪ Desk-based work on the transaction costs of different approaches to distributing benefits form carbon payments

▪ Field work on the benefits to local population from various conservation-linked projects

26

Thank you!

27

▪ CI-Madagascar and World Bank-Madagascar for sharing information on safeguards

▪ MEEF, CI and local leaders for permission to carry out the research

▪ The many people who took part in the research

▪ Our funders