Index insurance for agriculture in Ethiopia

Addis Ababa, Ethiopia, 9 December 2010

Design and Development of IBLI

for Southern Ethiopia

Andrew G. Mude

International Livestock Research Institute

Outline Developing IBLI Ethiopia Research Agenda

Objective

– To discover the viability and poverty reduction impacts of

index-insurance and establish whether, how and when these

impacts can be realized and sustained

Key Steps

• Design a livelihood-focused IBLI

• Identify suitable contract structure

• Identify suitable delivery mechanisms

• Address impacts of climate change and induce climate

change adaptation

PARIMA (2000-02)

• Quarterly household

survey (30 hhs/kebele)

• 5 kebeles in 4 woredas

Desta (1999)’s Herd

Recalls Data (1981-97)

• Annual household

herd recalls (~15

hhs/center)

• 35-km radius of 4

town centers

Existing Data Longitudinal household data

LIBEN

DIRE

MOYALE

AreroTELTELE

Yabelo

ODO

SHAKISO

BORE

URAGA

GELANA

ABAYA

ADOLANA

WADERA

Dillo

Negele

ARERO

Qorate

Moyale

Finchawa

HAGERE MARYAM

YEBELO

Mega

Wachille

KENYAKENYA

Dida Hara

Desta’s Town Center

PARIMA’s Kebele

Note on Studied Sites:

Borana and Guji Zones

Existing Data Longitudinal household data

Livestock Mortality rate (%)

0

1

2

3

4

5

6

7

8

LR 2000 LD 2000 SR 2000 SD 2001 LR 2001 LD 2001 SR 2001 SD 2002 LR 2002

Live

sto

ck lo

ss (

TLU

) p

er

ho

use

ho

ld

Drought

Disease

CCPP

Predator

Snake bite

Accident

Aging

Others

PARIMA (2000-02)

0%

10%

20%

30%

40%

50%

1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997

Livestock Mortality Recalls in Desta's Data (1981-1997)

Yabello

Mega

Negele

Arero

0%

10%

20%

30%

40%

50%

LR2000 LD2000 SR2000 SD2001 LR2001 LD2001 SR2001 SD2002 LR2002

Livestock Mortality experienced in PARIMA (2000-02)

Dida Hara

Dillo

Finchara

Qorate

Wachille

Drought 2000 Drought 2009 Good forage condition 2010

Satellite imagery NDVI (8km resolution, available every 10 days in real-time)

Liben

DireMoyale

Arero

Teltele

Yabelo

HegermariamOdo Shakiso

Bore

Uraga

Gelana Abaya

Adolana Wadera

Liben

DireMoyale

Arero

Teltele

Yabelo

HegermariamOdo Shakiso

Bore

Uraga

Gelana Abaya

Adolana Wadera

Liben

DireMoyale

Arero

Teltele

Yabelo

HegermariamOdo Shakiso

Bore

Uraga

Gelana Abaya

Adolana Wadera

-3-2-101234

0%

10%

20%

30%

40%

50%Drought 84-85 Drought 90-91 Drought 99-00 Drought 05-06 Drought 09

Area livestock loss (%) (Recalls from Desta 1981-97, PARIMA 1998-2002)

Standardized NDVI

Key Questions:

1. Other longitude household-level data? Verification of past data?

2. The need to collect recent herd recalls?

Existing Data NDVI and Livestock Mortality

Objectives

1. To introduce IBLI to local

communities/authorities

2. FGD to learn the local

opinions on IBLI

3. To meet with local

authorities, financial

institutions, NGOs

Key discussions

1. Contract design (Group-based, ex-ante payout)

2. Delivery mechanism

3. Climate change

Scoping Mission (2010)

Individual IBLI Group as delivery channel Group as client

+ Substitute for absence financial facilities in remote locations

+ Use group learning to enhance education and extension

+ Use group rules to reduce individual basis risks

+ Enhance prospect for linking credit access with insurance

-- Groups need to be well-established, participated by pastoralists

Contract design/delivery/regulation Group-based IBLI

i ii

G

Financial Institution

i ii

Financial Institution

i ii

G

Financial Institution

Ex-ante payout

for asset protection

Contract design/regulatory Ex ante/ Ex post payout

Observed lossStart of the contract

Ex-post payout

for asset replacement

IBLI payment to insured to

protect animals before drought

+ Could be cheaper to protect

animals than to replace them

- Rely on accuracy in predicting

livestock mortality ex-ante

Potential use:

• Supplementary feed and water

• Hired transport to access market

IBLI payment to insured to

replace animals after drought

+ Higher accuracy in predicting

livestock mortality

- May be more expensive to

replace animals

Key Questions:

1. Which product could provide more effective risk management?

2. What are regulatory pre-requisites of ex-ante product?

Reducing risk coverage for

commercial insurers for lower

premiums

Cap commercial provision of risk

to an intermediate risk layer.

Catastrophic zone can be explicitly

taken up by government or donors

who already offer some response

in times of catastrophic loss

Contract design/regulatory Risk Layering

Climate change more/worse droughts IBLI more

expensive … mitigation to limit premium rise, adaptation to

cushion against premium rise?

Evidence of change in NDVI

Climate change?

Incorporating Climate Change in IBLI modeling and

delivery:

Feeding climate change predictions into IBLI response functions and

pricing in expected climate change

Conditional Insurance Transfers

Climate change adaptation/mitigation

y = -6E-05x*** + 0.321

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

19

81

19

82

19

83

19

85

19

86

19

87

19

88

19

90

19

91

19

92

19

93

19

94

19

96

19

97

19

98

19

99

20

01

20

02

20

03

20

04

20

05

20

07

20

08

20

09

NDVI Trend in Dida Hara

(1981-2010)

• Follow a similar impact evaluation strategy as in Marsabit:

• Baseline of 920 households with 3 annual repeats

• Should allow us to rigorously establish impacts directly

attributable to IBLI

• Discount coupons and educational games to encourage

uptake

• Also plan for qualitative surveys to generally establish

perspective of target clientele.

Impact Evaluation

Thank you

For more information please visit:

www.ilri.org/ibli/