Ex post impact assessment of NRM research in the arid and semiarid areas : the case of the Mashreq/Maghreb Project experience. The impact of Cactus in alley cropping in the Tuisian case study

(1)

Ex post impact assessment of NRM

research in the arid and semiarid areas: The

case of The Mashreq/Maghreb Project

experience

The impact of Cactus in alley cropping in the Tunisian

case study

Alary V., Nefzaoui A., Benjema M.

SPIA / ICARDA

(2)

Activity 1

Technology and site

characterization

(3)

Technology characterization

Plate 1.3. In rural areas cactus is planted close to houses and used as a fodder bank for livestock

(Central Tunisia, 200 mm average rainfall)

(4)

Brief history…

• XV century… Introduction in Andalusia on the return of Columbus’ first

expedition from America

• XVII to XIX: implementation in North Africa with the return of Moors to their

homeland in NA

• 1930… Extension of cacti with international project (FAO, WFP)

• 1975… Establishment of national incentives and subsidies by the tunisian

goverment

• 1990… National strategies for rangeland imporvement / MM project:

research on technology packages to rehabilitate degraded range land

• 1998: development of cactus in alley cropping in the Zoghmar community in

parternship with development agencies (OEP, CRDA…)

(5)

Multi functional Role of Cacti in Alley cropping

• Expected Environmental impacts

– Controlling erosion and runoff

– Enrichment in organic matter and nitrogen – Top soil structural stability Æ roots

– Prevention and control of top soil loss due to wind

– Conservation of biodiversity: food and shelter for many widlife secies – Combat desertification

– Water saving

• Expected socio-economic impacts:

– Testimony of land rights without no land registry

– Multi uses: forage, food, potential market, medicinal applications, – Low cost feed in drought years

• Expected agronomic impacts

– Cereal grain yield increase – Biomass increase

(6)

Site characterization

N Zoghmar community N N Zoghmar community

(7)

Annual and monthly rainfall frequency calculated

for 27-years period

mm/ans 450.0 400.0 350.0 300.0 250.0 200.0 150.0 100.0 10 8 6 4 2 0 Std. Dev = 88.04 Mean = 221.6 N = 27.00 1 2 1 6 7 8 2 26,12 30,28 18,81 11,67 22,37 17,18 26,75 24,77 20,35 13,90 3,56 10,05 0 10 20 30 40 50 60 70

Sep Oct Nov Dec Jan Fev Mar Avr Mai Jun Jut Aout

Mois m m

(8)

Farming systems characterization

Z01 Z02 Z04 Z16 Z23 Z41 6a Z03 Z10Z11 Z12 _Z20 Z25 Z28 Z29 Z30 Z40 6b Z06 Z13 Z24 _Z19 Z31 6c Z07 Z08 Z09 Z15 Z17 Z33 6d Z18 Z21 Z22 Z27 Z34 Z35 Z39 6e Z26 Z32 Z36 Z37 Z38 6f -2.8 4.4 -4.2 3.6 Class 3 (EI1):

Very diversified system with non agricultural activities

Aroung 30-40 ewe on 50 ha

Class 1 (EI2): Young farmers

Less than 10 ewe on 10-15 ha (with 1-2 ha in irrrigation)

Class 4A (EI3) :

Diversified livestock systems with sheep/goat/cow

Class 4B (EA3):

Old agro pastors oriented livestock (10-15 ewe) with less than 9 ha

Class 2B (EA2) :

Large agro pastors oriented to agriculture

Class 2A (EA1) :

Large pastors or agro-pastors (more than 60 ewe and 30 ha)

(9)

Feed price variability with climatic conditions

0.506 0300

0.230

0.151 Hay

0.170

0.170 Feed Block

0.028

0.022

0.02

0.022 Cactus

0.160

0.160 Bran

0.427

0.285

0.178

0.084 Straw

0.136

0.170

0.170 Barley grain

Dry year

Poor year

Medium

year

Good year

Feed

(10)

Diet composition (dry matter basis) for sheep

raised in the Community of Zoghmar, Sidi Bouzid

(Ben Salem, 2000)

Plate 1.4. Cactus pads are commonly chopped in slices by hand; this procedure is time-consuming 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

Autumn Winter Spring Summer

Barley W. bran Stubble Pasture Cactus Straw

(11)

Activity 2

Identification and Quantification of

Performance Indicators

(12)

Agronomic and environmental impacts

4 Tons/ha

Cactus in AC

on NR

Roots

Organic

matter

Nitrogen

Soil

Carbon soil

Biomass

Yield

3.3 Tons/ha

0.8 Tons./ha

2.2 Tons/ha

Cactus in AC

with barley

Barley

NR

Results

at

the

end of

(13)

Impact of technology adoption on livestock activity

As expected, there is a negative relationship between total cost

and cactus acreage. Indeed, a one ha increase in cactus

plantation reduces total cost of livestock activity by 0.133 %

while 1ha increase in pasture or cereal land reduce total cost of

livestock activity by 0.11%.

kt kt t CAC i kt kt L CAC kit kt i CAC kt CAC kt A kt tL i kit ti tt t kt y i kit kt Li kt L kjt i j kit ij i kit i kt t CAC L CAC w CAC CAC A L t w t t t y w L L w w w C ε α α α α α α α α α α α α α α α + + + + + + + + + + + + + + + =

∑

∑∑

∑

, , , 2 2 1 2 1 0 ln ln ln ln ln ln ln ln ln ln ln

(14)

Total Factor Productivity decomposition

(Mundlak, 95,98)

Cactus adoption has enhanced productivity growth by 1.5

percent during drought period.

This contribution is somewhat low but it is worth to

precise that till 2003, cactus in Alley cropping

plantation was still young and thus unexploited.

*Figures represent average annual rate

A: total area; TC: technology change; CE: cost efficiency

4.9 -21.3 11.8 1.1 0.2 10.4 2.7 2002-2003 -18.1 0.5 -16.4 1.5 -0.4 -4.0 1.0 1999-2002* TFP CE TC Cactus Labor A Scale

(15)

Efficiency analysis

.00 .05 .10 .15 .20 .25 .30 1 2 3 4 5 6 7 8 9 10 11 12 13 Cost inefficiency P roba bi lit y

Figure 1. Kernel Density function: Inefficiency level in 1999

.00 .05 .10 .15 .20 .25 .30 1 2 3 4 5 6 7 8 9 10 11 12 13 Cost inefficiency P roba bi lit y

.00 .05 .10 .15 .20 .25 .30 2 4 6 8 10 12 14 16 cost inefficiency P rob ab ili ty

0.0015 -3.270 -0.051 CAC 0.3826 0.877 0.286 L 0.4755 -0.716 -0.041 Qw 0.0000 6.585 0.403 Qf 0.2777 -1.091 -0.062 FEM 0.9328 -0.085 -0.011 IR 0.9327 0.085 0.007 INST 0.6353 0.476 0.016 F 0.7896 -0.267 -0.001 age 0.7750 0.287 0.005 REV 0.0157 -2.457 -2.587 Interc . p. value t-ratio coefficient

Table 2.6 : Determinants of inefficiency

• The inefficiency distribution in 2003

dominates both 2002 and 1999

• Indeed, an increase in cactus acreage

reduces degree of cost inefficiency

(16)

Social indicators

0.043 0.241 1.86 4.94 20.51 230 2002 0.117 0.245 4.28 11.01 20.00 221 1999 Gini pov Gini total Sen Indicator Poverty Gap ((y-p_m)/y Head count (poor/n) (in %) Poverty line

Poverty indicators and expenditure distribution

C once ntration curve : Total sample

0,0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1,0 0,0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1,0 1999 2002

C once ntration curve : Pove rty sample

0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1 0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1 1999 2002

(17)

Activity 3

Valuation of the benefit and Cost

of the Technology

(18)

Hypothesis

• Benefit cost analysis conducted for 1 ha

• Two alternative: cladode market (0.03 DT/kg);

no cladode market (equivalent energy)

• Two scenarios: With subsidies and without

subsidies

• Two beneficiaries:

– Farmers: subsidies are products

– Project: subsidies are costs

(19)

Cactus on marginal cereal land

3 88% 455.73 (2) 3 95% 575.08 (1) (S6) Farmers without OEP incentive

But real cost

1 *** 793.5 (2) 1 *** 1012.30 (1) (S5) Farmers with OEP incentive But real cost

5 29% 310.27 (2) 5 34% 429.62 (1) (S4) Farmers without OEP incentive 3 100% 648.04 (2) 3 111% 866.85 (1) (S3) Farmers with OEP incentive 5 28% 424.11 (2) 5 34% 642.92 (1)

(S2) Project without OEP incentive 7 20% 299.77 (2) 6 26% 518.58 (1)

(S1) Project with OEP incentives Recovery time (year) Internal rate of return (IRR) Net discounted value (DT) Hypothesis* Item

(20)

Cactus on degraded pasture land

8 13% 43.38 (2) 7 19% 167.72 (1)

(S6) Farmers without OEP Incentive

But real cost

5 52% 386.14 (2) 5 68% 604.95 (1) (S5) Farmers with OEP incentive But real cost

11 6% -97.08 (2) 9 11% 22.27 (1) (S4) Farmers without OEP incentive 7 22% 240.69 (2) 6 30% 459.5 (1) (S3) Farmers with OEP incentive 9 11% 16.76 (2) 8 16% 235.57 (1)

(S2) Project without OEP incentive 11 5% -186.47 (2) 9 13% 111.22 (1)

(S1) Project with OEP incentives Recovery time (year) Internal rate of return (IRR) Net discounted value (DT) Hypothesis* Item

(21)

Benefit Cost analysis

Pasture land

Marginal cereal land Types of land 14 17% 221 Without pad market 7 45% 1,320 With pad market 10 18% 217 Without pad market 7 46% 1,371 With pad market Recovery time (year) Internal rate of return (IRR) Net discounted value ( in thousand DT) Scenarios

Hypothesis:

- No taking into account the research cost

- The cost is 500 DT/ha with the OEP intervention

(22)

Activity 4

(23)

Adoption indicators

28.4%

26.9 Degree adoption for cactus in

alley cropping in Zoghmar***

??

Degree adoption for cactus in alley cropping in the project**

18.9

40.7

21.5 38.1

Degree adoption for cactus

533

1475

711 536

Potential area for cactus in alley cropping in Zoghmar

??

Potential area for cactus in alley cropping in the area of the project

419

122 cactus area in alley cropping

101

726

153 204

Total cactus area

45.0 65.0 30.6 35.9 37.5 Adoption rate (%) 18 26 97 14 15 Adopters 40 40 317 39 40

Total number of farms (sample)

2004 2002/03* 2002 2001/02 2000 Date E D C B A

(24)

Adoption indicators per farm types

0% 15% 30% 24% 31% Adoption rate (%) 419 317 Total 0 0% 0% 0 21% 12 0 ha 0.26 11% 5% 23 20% 87 1-5 ha 0.85 16% 17% 73 26% 85 5-10 ha 1.25 13% 20% 83 27% 66 10-15 ha 3.58 21% 57% 240 3% 67 > 15 ha Average cactus area (in ha) degree of adoption % cactus AC Cactus AC (in Ha) % farm Nb Farm Farm size 0.17 0.38 0.19 0.17 0.08 % farm 419 317 Total 13% 0.07 28 2% 55 0 13% 0.20 86 26% 120 < 15 18% 0.22 91 36% 61 15-25 16% 0.30 127 38% 55 25-50 17% 0.21 87 46% 26 > 50 Adoption degree (%) % cactus (ha) Cactus AC (in Ha) Adoption rate (%) Nb far m Small Ruminant s

(25)

Determinants of technology adoption

0.0019 3.1055 6.3850 19.829 IRrigation 0.0760 -1.7744 5.4707 -9.7073 INSTruction 0.0019 3.1050 0.2516 0.7813 AGE 0.0007 -3.3963 1.3472 -4.5758 Household size 0.2722 1.0980 0.6631 0.7282 Off farm 0.3492 -0.9361 0.0517 -0.0484 Livestock size 0.0894 -1.6983 3.1171 -5.2940 Labour 0.0007 3.3858 0.2307 0.7811 Area 0.9834 0.0208 16.475 0.3437 Intercept Prob. z-Statistic Std. Error Coefficient

Method: ML - Censored Normal (TOBIT) (Quadratic hill climbing) Included observations: 33

Left censoring (value) at zero

(26)

Effect of subsidies on technology

adoption-Method of contingency

.0 .1 .2 .3 .4 .5 0 2 4 6 8 10 12 14 16 Acereage Pro ba bil ity Figu re.4.1 K ernel Den sity of acereag e allowed to CA C

.00 .04 .08 .12 .16 .20 .24 0 2 4 6 8 10 12 14 16 C actu s acreag e Pr ob ab ili ty Figu re 4.2. K ernel D ensity of acereage allowe

.00 .04 .08 .12 .16 0 2 4 6 8 10 12 14 16 Cactus acreage P robabi lit y Figure.6

Kernel Density of acereage allowed to CAC with S2

.00 .02 .04 .06 .08 .10 .12 0 2 4 6 8 10 12 14 16 Cactus acreage P robabi lit y

(27)

Activity 6

Modelling Approach For Ex Post

Impact Assessment

(28)

NRM research project- Cactus in Alley Cropping

Environment

-Soil conservation

(Soit texture, soil

enrichment, control soil

loss)

-Controlling erosion

-Water management

(avoid runoff,permeability

and water storage)

-Main biodiversity

Economy

-Feed stock, low

cost feed

-Diversification (fruit,

cladode market)

- biomass increase

-Testimony land

rights

Reproduction

Ecology

Efficiency

(29)

Classical view: Ex ante & ex post assessment

T-5

T (2004)

T+10

Ex post Impact

assessment

Benefit Cost

Ex Ante Impact

assessment

(Community model)

Rev.

Farm n

Without adoption

(Control group)

With adoption

With Change

Without Change

(30)

Ex Post Impact Assessment of NRM project

Problems of this approach:

• No taking into account the dynamic of these systems

• Natural Resource management needs to integrate the

trade off between present and future

• Complexity of the interactions

(31)

Extrapolation

T-5

Reference

T (2004)

T+10

Rev.

Farm n

With adoption

Counterfactual:

Without technology

Understanding the processus

Without adoption

Benefit due to community change

Negative impact

Confront to reality

Assess the viability at the community

Level (biological and Economic model)

(32)

Bio-economic model

Climate,

soil

Set

of technics

Bio-physical

model

Expected

Production

Yields

Externalities

Erosion

Resource

constraints

Economic

model

input output input input

Decisions

Feedback

(33)

Limitation to approach the sustainability

Climat,

soil

Technic

Bio-physical

model

Production

Yields

Engineering

production

functions

Externalities

Erosion

Resource

constraints

Economic

model

input output input input

Empirical observations

Modelling

(34)

Community model

Agro

pastor

Mixed

farming

system

Exchange of labor, land

and feeds

Informal credit

Cropping and pastoral

System

Livestock System

Market

Manure, labor

Intra-consumption

(straw, grain, stubble, pasture

Sell of

products

Animal sell/purchase

Inputs’ supply

Complementation

Self-consumption

Institutions:

Credit

Subsidies

Regulation

price

Market

R&D: technology introduction

Farm

level

Community

level

National

level

Off farm

(35)

Cropping and grazing management

Pastoral land: “Snc”

With or without Cactus, Atriplex

Cactus AC

Arable land in dry: “ScSec”

Cereal : durum wheat, bread wheat, barley Fallow: worked or no,Vetch

Cactus, Olive trees, Cactus AC

Arable and irrigated land: “SI”

Cereal : durum wheat, bread wheat, barley Fallow: worked or no

Sorghum, oat vegetables Olive or fruit trees

Grazing

grain, straw,

hay, stubble,

Pad,

twig,

fruit,

green,

“grignon”

Assessment Of forage available (DM, UF, MAD, min. barley)

F(till, seed, fertilization, traitment, )

Yield, quality

_{Nutritive intake}

(36)

Demography of the herd

Reproductive

animals

Kid M/F

(Less than 3 months)

Lamb M/F

(less than 3 months)

3 to 6 months

6 to 9 months

12 to 18 months

9 to 12 months

3 to 6 months

6 to 9 months

12 to 18 months

9 to 12 months

purchasing

urchasing

sold

purchasing

(37)

Cropping

act

. (ha)

Input/ Output coefficients

Yield

, Nutritive value, etc.

Till,

seed

,

fertilizer

,

traitment

,

Need

in

hour

-

-work

work

Need

in

mechanization

Investment

(land

management,

planting

, etc.)

Animale

act

. (

specie

, age)

Need

in

hour

-

-work

work

Need

in

mechanization

Growth

grain,

yield

,

Complemention

, ration

Investment

(Building,

(38)

Constraints related to

factors

Land : ∑X(pc)

≤

Tdisp(pc-1)

+Tpurchase(pc) + Trent (pc) + Tassoc.

(pc) - Tsold (pc) – Tgrent

(pc)-Tgasso(pc) – Tplanted (pc)

Equipment :

NeedHT (pc)

≤

HT.disp(pc-1)

+ HT. purchased(pc) + HT. rent(pc)

-HTgrent(pc)

Labour :

NeedMO (pc)

≤

MO.SalPer (pc-1)

+ MO.f(pc-1) + MO. occa(pc) –

Mof_sal(pc)

Animal demography

Nutritive need:

NeedNUT (pc)

≤

NUT.

Disp (pc)

Ration :

Cactus

≤ 20

% DM need

Concentrates

≤

50% DM need

Constraints related

to livestock act.

Seasonal constraints

Technical

(39)

Â

Short terme credit:

Informal/community/Formal

Â

Emprunt à Long terme :

Ceiling: EMPCT (PC)

≤

CCTLim (500 DT)

Guarantees: REMBT (PC)

=

EMPCT (PC-1)

Â

Threshol or break even point for the cash flow :

CASH (PC)

>

CashLim (5000 DT)

Â

Risk :

minimize the deviation relative to threshold of

income (Z

0 ) defined in advance

Â

Etc.

Economic

(40)

01 Â

Purchase

Animals

Â

Purchase

equipments

(

mechanization

,

car, irrigation, etc.)

Â

Land

management

Â

Purchase

land

Â

Building

Own

capital

Long

terme

credits

Investments

(41)

01 Â

variables charges

(

fertilizer

,

seed

,

carburant

…

)

Â

Fixed

charges

Â

Financial

charges

Â

purchased

services

Â

land

rent

Â

Family

costs

Â

Reimbursement

Â

Saving

Â

Sold

Â

Subsidies

Â

Sold

of services

Â

Interests

on

investment

Â

land

incomes

Â

External

incomes

Â

Short

term

credits

Balance

(PC

-

1)

Receipts

(PC)

-

Expenditure

(PC)

_B

al

a

n

c

e

p

c

(p

c)

+

=

Cash

(42)

Individual

Economic

model

Individual

Economic

model

Exploitation 1

Exploitation n

Community

model

Community

Constraints

Community

parameters

Model to assess

Policy or

technical

change

Coupling

with

biological

Models

Æ

Assess

impact

Of

technics

on

the

viability

Assess

Policy

changes

or

technical

changes

(43)

Objective function under risk constraints

∑

=

+

T t ye ye ye ye

X

C

0

(

1 τ

)

;

0 0,

X

;

0 M

=

;

=

P

;

s

1,...,

=

r

;

0 X

C

T

r ye j r ye r s =1 r r ye j ye r j ye n j=1

.

≥

→

Ω

≤

Σ

−

Σ

−

λ

, , , , , ,

Max E(Z) =

+ K /

With : A X

_ye

≤ B

_ye

; B

_ye

= bX

_ye-1

; X

_ye

≥ 0

Under risk constraints (Target Motad):

)

1

(44)

Calibration/ Validation

0 5 10 15 20 25

EA1 EA1 EA2 EA2 EA3 EA3 EI1 EI1 EI2 EI2 EI3 EI3 Barley Wheat Barley Wheat Barley Wheat Barley Wheat Barley Wheat Barley Wheat

Figure 6.2: Comparison of cereal area in 1999 and 2004 for each farm type

Real 99

Model 99

Real 2004

(45)

Calibration/ Validation

1.83 -9.11 -25.00 -27.15 -3.98 EI3 -13.89 -1.39 2.65 6.88 6.02 EI2 6.88 11.28 -3.65 -0.76 -8.21 EI1 -4.56 -4.43 7.99 -6.33 -5.62 EA3 -28.28 -14.34 -1.67 -6.33 -2.48 EA2 8.41 9.78 -5.64 -8.07 -33.98 EA1 2003/04 2002/03 2001/02 2000/01 1999/00 -6.88 24400.8019 8 26205 2002 4.33 25106.7645 24064 1999 Average deviati on -54.71 1090.51 2408 2002 -5.07 1706.74 1798 1999 EI3 -4.26 3774.03 3942 2002 1.83 1830.89 1798 1999 EI2 1.29 12294.2619 8 12137 2002 4.40 8461.4075 8105 1999 EI1 31.08 1713.23 1307 2002 0.33 2266.51 2259 1999 EA3 16.62 1909.07 1637 2002 12.16 1931.447 1722 1999 EA2 11.72 3619.7 3240 2002 6.29 8909.77 8382 1999 EA1 Deviation Model Reference Year of validat ion Farm

Ewe stock

Cash Flow

(46)

Ex Post Impact Assesment

To evaluate the impact assessment of the technology “cactus in alley

cropping” :

• Scenario 1: The technology option doesn’t exist. The farmers have

only one alternative related to cactus is to plant whole area of

cactus. This situation could be considered as the counterfactual

situation and allows to estimate the all benefit of the technology at

the farm level

• Scenario 2: The technology exists but there is no funded project

to facilitate the adoption. We can compare the adoption level with

and without the subsidies.

• Scenario 3: The technology exists with the funded project. We will

compare the real adoption with the adoption in the model.

• Scenario 4: The technology exists with the funded project. There is

no restriction for the support. We could estimate the potential

(47)

Scenario 1: Ex post Impact assessment of the

technology & institutional environment

Figure 6.3 : Gaps for ew e stock w ith and w ithout the technology+institutional action (in %) -30 -10 10 30 50 70 90 110 130 150 1999/00 2000/01 2001/02 2002/03 2003/04 year in % EA1 EA2 EA3 EI1 EI2 EI3

The annual average ewe stock is 6% more than in the counterfactual situation.

So this confirms the role of cactus during drought years to avoid de-stocking.

(48)

Scenario 1: Ex post Impact assessment of the

technology & institutional environment

Figure 6.4 : Gaps for cash flow with and without the technology+institutional action -30 -20 -10 0 10 20 30 40 50 60 1999/00 2000/01 2001/02 2002/03 2003/04 year in % EA1 EA2 EA3 EI1 EI2 EI3

The Zoghmar community registers in average

an increase of 7 % of the annual cash flow. .

(49)

Scenario 1: Ex post Impact assessment of the

technology & institutional environment

Reduction of traditional cereal system (5%), responsible of erosion

Figure 6.5 : Gaps for cereal areas with and without the technology+institutional action -60 -40 -20 0 20 40 60 1999/00 2000/01 2001/02 2002/03 2003/04 year in % EA1 EA2 EA3 EI1 EI2 EI3

(50)

Scenario 1: Ex post Impact assessment of the

technology & institutional environment

No impact on poverty at the community level

But no distinction between cactus adopters and the others

0.29

0.26

0.31

0.27

0.28 Counterfactual

0.31

0.29

0.31

0.28

0.28 Reference with

project

2003/

04 2002/

03 2001/

02 2000/

01 1999/

00

(51)

Scenario 2 & 3: Ex post Impact assessment of the

institutional environment

1 5 11.23 5 10.75 0 EI3 0.5 0 3.85 2.67 3.85 2.67 EI2 30 45.6 30 29.17 10.21 EI1 2 2 3.34 3.34 3.34 0.29 EA3 2 1 2.93 1 2.93 0 EA2 8 5 16.53 5 5.78 0 EA1 Area with spine cactus Area of cactus in alley cropping Adoption level with limited OEP incentive And yield increase Adoption level with limited OEP incentive Adoption level without OEP incentive+ 30% yield Adoption level without OEP incentive Farm Survey S4 S3 S2 S1

(52)

Scenario 4 : Ex post Impact assessment of the

institutional environment – Extension of public support

5

14.25

10.71

10.71 EI3

0

5.5

5.5 EI2

16

50

29.17

29.17 EI1

2

11.4

3.34

3.34 EA3

1

2.93

2.93 EA2

5

16.53

5.78

5.78 EA1

reference

2003/04

2002/03

2001/02

2000/01

1999/00

Scenario1

All the farmers increase their area three fold

(53)

First Conclusions

When we compare the area allocated to the technology between the different scenarios,

we can tell that a good information about the yield expectation with the technology could give similar adoption degree than subsidies.

But it is true that the reality is more complex:

• The expected subsidies can be more crucial, especially considering that during dry years, the expected yield of cereal in alley cropping could be inferior to the subsidies • Why implement alone this technology if we could profit from subsidies and yield

increase in the same time? So some farmers are waiting…

• Good information at the community level is always difficult or even impossible • As with good information, the level of believe in the information intervenes.

(54)

Other simulations

Figure 6.9 : Gaps for ew e stock w ith the cereal and m eat m arket liberalization and w ithout the technology+institutional action (in %)

-100 -50 0 50 100 150 1999/00 2000/01 2001/02 2002/03 2003/04 year in % EA1 EA2 EA3 EI1 EI2 EI3

Figure 6.10 : Gaps for ew e stock w ith the cereal and m eat m arket liberalization and w ith the technology (in %)

-50 -40 -30 -20 -10 0 10 20 30 1999/00 2000/01 2001/02 2002/03 2003/04 year in % EA1 EA2 EA3 EI1 EI2 EI3