Punjab Economic Opportunities Programme

Preliminary Findings for Livestock and Dairy Interventions

Imran Rasul (UCL), Adnan Khan (IGC, LSE)( ) ( )and

Robin Burgess (LSE), Orazio Attanasio (UCL) g ( ), ( )and Oriana Bandiera (LSE)

1

861 PSUs

I. Background to PEOP: Livestock and Diary Interventions

• Goal is to improve the lives of households through:p g

Better access to marketable skills

Access to dairy yield enhancing practices, better animal health care, nutrition and breed

improvement to 100,000 dairy farmer

Better functioning livestock markets and linkages accessible to 50,000 livestock farmers

• Achieving this means understanding issues on both demand & supply sides• Household, Village and Supply Side Surveys

• Interventions (evaluated through RCTs):

Constraints on knowledge and practices - Information Campaigns (Farmer Days, mobile telecoms)

Constraints on raising productivity yields – Mobile Vets, etc

Asset transfer programs

Shortcomings in s ppl side factors linking farmers to markets Shortcomings in supply side factors linking farmers to markets

3

Livestock OwnershipLivestock Ownership

Full Sample Rural Only Urban Only

Non-owner HHs 48.23% 36.22% 79.57%

HHs owning at least one 51.78% 63.78% 20.43%animal 51.78% 63.78% 20.43%

Mean of HHs with animal,conditional on ≥1 5.72 5.63 6.47

Mean of Cattle + Buffaloes,conditional on ≥1 3.73 3.70 3.96

• Asset transfer programs targeted towards non-owners: benefit 48% of households• Livestock rearing practices programs: benefit 52% of households• Livestock rearing practices programs: benefit 52% of households• Market linkage programs: directly benefit 52% of households

Hi hl l d f li k h h ld i b

4

• Highly selected group of livestock households in urban areas

ProductivityMilk yield distributions by animal type

Animal Mean 25th Percentile Median 75th Percentile

Cow 2.29 1.5 2 3Buffalo 2.98 2 3 4

• The top 25% of farmers have yields more than double the bottom 25% of farmers

• This productivity variation remains when we take account fof:

• the number and types of animals owned, other h h ld hhousehold characteristics• the PSU in which the farmer lives

5

Implications for Livestock Policy

• Big differences in self-reported skills and practices among farmers in the 1st

and 4th quartile of the productivity distributionq p

• POLICY LESSON 1: If this productivity dispersion is because of basic kill d ti th 40% i t i i ld f fskills and practices, then 40% improvements in yields for some farmers are

achievable

• Farmer days: low cost, potentially high return intervention

/• NEXT QUESTION: do farmers express a desire for more skills/training?

6

Current Skills

Table Skills most suited for Males Table Skills most suited for Females

Current Skills

What tasks are you skills

most suited for?

District Code

BHN BHW LDH MZG Total

What tasks are you skills

most suited for?

District Code

BHN BHW LDH MZG Total

Table Skills most suited for - Males Table Skills most suited for - Females

Farm Maintenance 44.93 14.58 3.45 3.92 17.67

Farming 49.76 47.92 43.10 60.29 50.49

Farm Maintenance 50.00 14.81 5.26 6.25 9.65

Farming 50.00 44.44 5.26 12.50 20.18

Animal Breeding 1.45 34.38 53.45 15.69 25.14

Animal Health Care 2.90 1.82 0.00 5.88 2.74

P l 0 97 1 04 0 00 3 43 1 43

Animal Breeding 0.00 25.93 89.47 75.00 63.16

Animal Health Care 0.00 14.81 0.00 0.00 3.51

P l 0 00 0 00 0 00 3 13 1 75Poultry 0.97 1.04 0.00 3.43 1.43

Veterinary 0.00 0.26 0.00 10.78 2.52

Poultry 0.00 0.00 0.00 3.13 1.75

Veterinary 0.00 0.00 0.00 3.13 1.75

N 207 384 116 204 911 N 4 27 19 64 114

7

Desired Skills

T bl Skill lik i M l T bl Skill lik i F l

Desired Skills

What skill do youthink Name would like

t i ?District Code

What skill do youthink Name would like

t i ?District Code

Table Skills like to acquire - Males Table Skills like to acquire - Females

to acquire? BHN BHW LDH MZG Total

Farm Maintenance 50.53 14.81 4.31 2.87 18.25Farming 43.16 44.97 43.1 60.29 47.93

to acquire? BHN BHW LDH MZG Total

Farm Maintenance 33.33 8.33 0 6.35 6.36Farming 66.67 50 0 12.7 20

Animal Breeding 2.11 35.98 52.59 16.27 26.32Animal Health Care 3.16 2.65 0 6.22 3.25

Fish Farms 0 0.26 0 0 0.11Poultry 1.05 1.06 0 3.35 1.46

Animal Breeding 0 29.17 95 74.6 66.36

Animal Health Care 0 12.5 0 0 2.73Fish Farms 0 0 5 0 0.91

Poultry 0 0 0 3.17 1.82y 5 5Veterinary 0 0.26 0 11 3

N 190 378 116 209 893

yVeterinary 0 0 0 3.17 1.82

N 3 24 20 63 110

POLICY LESSON 2: DEMAND TO ACQUIRE MORE SKILLS AND TRAINING IN LIVESTOCKDemand predominantly for more of the same skills, not diversification

8

I t U A i l H lthInput Usage: Animal HealthFigure 4.2. Livestock Health Status by

Animal Type Figure 4.3. Percentage animals by

80

90

100

Animal Type

90.00

100.00

Figure 4.3. Percentage animals by animal type and cause of loss

60

70

80

cattle 60.00

70.00

80.00

30

40

50 buffalo

sheep

goat

chicken/duck30.00

40.00

50.00

10

20

30

0.00

10.00

20.00

Cow Cattle Buffalo Buffalo Sheep Goat Goat Chicken/ Chick/ 0

percent vaccinated

percent sick percent of sick that were treated

Young Young Duck Duckling

Died due to illness Died due to accident Theft

9

Input Usage: Animal Health p g

Table: Percentage of government treatment facilities availed by households by district

% of LivestockTotal Govt Vet

Reason for not availing as %age of Livestock owning HHs that did not avail

District% of Livestockowning HHsthat availed

facilities (Source: LDDD) Too far Poor Response Too Expensive

Bahawalnagar 40.29 184 83.74 13.82 2.44

Bahawalpur 53.95 173 88.57 8.57 2.86

Lodhran 18.75 37 73.08 22.31 4.62

Muzaffargarh 38.61 132 78.49 18.82 2.69

N 526 468 93 18

POLICY LESSON 3: need more service availability: mobile veterinary services

10BRING SERVICES TO THE FARMERS: FARMERS CANNOT TRAVEL TO THE SERVICES

Perceptions of Animal Health Service ProvidersPerceptions of Animal Health Service Providers

S ti f ti ith

Table: Quality of animal health services

Perception of quality(in percentages)

Satisfaction with outcome of dealing with animal health(in percentages) with animal health service provider

Private Govt Private Govt

Very Unsatisfactory 3.15 4.16 3.43 5.17U ti f t 10 67 18 2 11 52 18 55Unsatisfactory 10.67 18.2 11.52 18.55Neutral 33.93 38.03 32.81 36.37Satisfactory 49.61 37.08 49.44 37.38Very Satisfactory 2.64 2.53 2.81 2.53

11

V Outputs: Milk Sale and PurchaseV. Outputs: Milk, Sale and PurchaseFigure 5.2. Average milk selling price by animal type and time of day

Average cow milk price by type of individual and time

Average buffalo milk price by type of individual and time

25 0030.0035.0040.0045.00

25 0030.0035.0040.0045.00

0 005.00

10.0015.0020.0025.00

0 005.00

10.0015.0020.0025.00

0.00

Dho

dhi

Chill

er

Cons

umer

-Ru

ral

onsu

mer

-U

rban

Dho

dhi

Chill

er

Cons

umer

-Ru

ral

onsu

mer

-U

rban

0.00

Dho

dhi

Chill

er

onsu

mer

-Ru

ral

nsum

er -

Urb

an

Dho

dhi

Chill

er

onsu

mer

-Ru

ral

nsum

er -

Urb

an

POLICY LESSON 5: increasing market linkages to dodhis and chillers will

C Co C Co

Morning Evening

Co Co Co Co

Morning Evening

12

raise farmer incomes substantiallyRecall majority of low productivity farmers are consuming all milk

VI. Intermediate Markets ChillersVI. Intermediate Markets Chillers

Figure 6.4. Reasons why not use private chillers (by district)

35404550

20253035

51015

0Chiller Too Far

AwayExpensive They don't give

a fai Didn't Know

About ItDon't Want To Other (specify)

Bahawalnagar Bahawalpur Lodhran MuzaffargarhBahawalnagar Bahawalpur Lodhran Muzaffargarh

POLICY LESSON 6: LINK FARMERS TO THE SERVICES: FARMERS CANNOT TRAVEL TO THE SERVICES

13

TO THE SERVICES

FinanceFinance

Fi 8 1 P p f L

Credit Landlord

6%

Fig 8.2. LS Loan SourceFig. 8.1. Purpose of LoanConsumption expenditure (personal items

Investment expenditure, such a B ildi

Formal Inst11%

from shop11%

28%

Building

Business expenditure

To pay for unforeseen expenses

MFI/NGO11%

Neighbourhood/Friend

61%3%

50% To purchase animals, livestock, feed, r

Education

To Purchase seeds

Pawnshop0%

9%

4%

1%

0%4%

1%

Other

Total

1%

POLICY LESSON 7: Microfinance is not the most severe constraint to farmers

14

Practices, linkages, asset transfers will have larger impacts

Baseline Surveys-Phase I

• Results from Phase I presented strong evidence for the following potential interventions– The need for increased livestock asset ownership (1/3rd of rural

households did not own livestock) – Asset Transfer ProgramsA t b tt li t k d d i p ti F D /– Access to better livestock and dairy practices - Farmer Days / Mobile Telecom Services / Mobile Vet Services / Lady Livestock Workers

– Improved linkages and increased access to livestock markets• Supply side interventions• Collective bargaining power

Interventions For Which Current Evidenceis Less Supportive

• Less compelling evidence that lack of credit is a major impediment for achievement PEOP goals in livestockg– Farmers do NOT say they don’t report services because they are

too expensive– Farmers currently ARE able to take out credit loans, and they do

NOT use them for livestock related purposes

• Less evidence of need to obtain access to cattle markets (well integrated prices of animal sales and purchases)integrated prices of animal sales and purchases)– Well integrated markets for buying and selling livestock– NO strong evidence on need to establish central cattle marketsg– Focus should be on more localized service provision and delivery

16

Livestock Supply Side SurveysLivestock Supply Side Surveys

Surveys target the supply side of market:1. Veterinary Institutes2. Milk Collection Centres and Chillers3. Cattle Markets4. Microfinance Institutions5. Private Vets6. Informal Vets (quacks)7. Dodhis and milk traders7. Dodhis and milk traders8. Informal Moneylenders

17

Livestock Supply Side SurveysLivestock Supply Side Surveys

These surveys will collect information on:These surveys will collect information on: - Services offered, pricing strategy, groups served and client

demographics business networks constraints to expansiondemographics, business networks, constraints to expansion

WHERE ARE THE GAPS IN THE SUPPLY SIDE?

ARE SOME GROUPS BEING EXCLUDED FROM THESE SUPPLY SIDEARE SOME GROUPS BEING EXCLUDED FROM THESE SUPPLY SIDE SERVICES

RCT EVALUATION OF SUPPLY SIDE CHANGESRCT EVALUATION OF SUPPLY SIDE CHANGES can we work with the private sector and NGOs to evaluate impact ofinterventions like opening chillers?

18

Intervention -Farmers’ Day

Intervention Design

• Two day activity comprising registration, information and training sessions on Two d y c v y co p s g eg s o , o o d g sess o s othe first day and door to door vaccination on the second day

• Feedback regarding the quality of information session was obtained from tt d th fi t d O th d d f ll f fill dattendees on the first day. One the second day, follow-up forms were filled

out by the treatment households• RCT design was used: assignment of villages to treatment and control groups

19

Feedback

Usefulness of training in creating awareness regarding…(Scoring on a scale of 1=very poor & 5=very good)

4.12

4.1

4.15Prevailing Infections

Worms (Treatment nd C ntr l)

Milk value addition4.04

3.94

3 95

4

4.05

and Control)

B l d F diG t /Sh i4.044.09

3.85

3.9

3.95 Balanced Feeding Goat /Sheep rearing

4.094.09

Breed ImprovementMineral

supplementation

4.014.1

Fodder improvement

Calf Rearing

3.98

Dairy Farming business Training Component

20

Reasons for Previously Not Attending Farmers’ Day

• More than 50% of the farmers seemed to be unaware of such an activity or its purpose

Others 

activity or its purpose

• This coupled with the feedback

Do not need training

Venue of farmers day is

received from the participants makes clear the need for such an intervention given the huge

i i Venue of farmers day is too far away

Did not know about a farmer’s day

existing gaps

• Farmers are aware of the need for

0% 10% 20% 30% 40% 50% 60%

Farmers day will not be helpful

better animal health care but either do not have access to facilities or lack practical knowledge

21

Impact on Key Outcomes

OutcomeBaseline Measure

• Upon running tracker surveys on the PEOP participating h h ld i f Productivity Per Cow Per Day (litres) 2.50

Productivity Per Buffalo Per Day (litres) 2.49

How often is water provided to livestock

households, impact of intervention will be measured

How often is water provided to livestock (times/day) 2.33

Number of cows and buffaloes vaccinated 3.29

Ho sehold Income from Li estock Per Month

• Key baseline and endlineoutcomes of interest fall into three categories Household Income from Livestock Per Month

(Rs.) 16986.23

Food Consumption Per Capita Per Month (Rs.) 1374.89

three categories– Awareness (changes in

livestock healthcare) Skill E h ( h i– Skills Enhancement (changes in productivity, better feeding practices)H h ld Li i S d d– Household Living Standard (income and consumption)

22