Secondary Cities and Agricultural Transformation: Evidence from Ethiopia World Bank – May 19, 2016 Joachim Vandercasteelen, Seneshaw Tamru, Bart Minten, and Johan Swinnen Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized closure Authorized
38
Embed
Secondary Cities and - World Bankpubdocs.worldbank.org/pubdocs/publicdoc/2016/5/... · Cities & Agricultural transformation in Ethiopia 28 • Impact of urban proximity on teff production
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Secondary Cities and Agricultural Transformation: Evidence from Ethiopia
World Bank – May 19, 2016
Joachim Vandercasteelen, Seneshaw Tamru, Bart Minten, and Johan Swinnen
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Urbanization
2
• Urbanization is quickly increasing in SSA o Growth and concentration in medium-sized citieso Economic development and poverty
• Composition of urbanizationo Agglomeration in mega citieso More dispersed pattern leads to more inclusive growth
• Higher urban demand• Agricultural production in rural hinterland
• Rural development o Small scale and rural farming
Cities & Agriculture in Ethiopia
4
• Ethiopia is urbanizing fasto Urban share will triple in next decade o People are concentrating in cities
• Addis Ababa • Secondary towns
• Teff is major staple cropo Produced by 6 million farmerso Consumed daily by 2/3 Ethiopianso Economically superior good
Cities are growing fast in Ethiopia
5
0
100
200
300
400
500
600
Mek
elle
0
1000
2000
3000
4000
5000
6000
Add
is A
baba
1950 1960 1970 1980 1990 2000 2010 2020 2030Year
Population in thousands
Market Access has improved
6
Conceptual framework
P (teff price)
µf (Transportation costto City 1)0
Agriculture and 1 City
Agriculture and 2 Cities
9
• Now: Urban proximity to 2 citieso Presence of city 2 (Secondary Town) at μo Same price and per unit transport costo , and , .
Teff price
Transportation cost to City 1
μ1 μfnμ2
0
Impact of City 1 Impact of City 2
μf1
Pf,1(μ) Pf,2(μ)
Prices and 2 cities
Teff price
Transportation cost to City 1
μ1 μfnμ2
0
Impact of City 1 Impact of City 2
μ1'μf1
Pf,1(μ) Pf,2(μ)
Prices and 2 cities
Teff price
Transportation cost to City 1
μ1 μfnμ2
0
Impact of City 1 Impact of City 2
μ1'μf1
Pf,1(μ) Pf,2(μ)
Prices and 2 cities
Teff price
μ2
0
Impact of City 1 Impact of City 2
Pf,1(μ) Pf,2(μ)
μ3
Impact of City 3
μfnμ11μf1 μ1
Pf,3(μ)
Prices and many cities
Transportation cost to City 1
Teff price
μ2
0
Impact of City 1 Impact of City 2
Pf,1(μ) Pf,2(μ)
Prices and many cities
μ3
Impact of City 3
Transportation cost to City 1
Data
Survey Data
16
• Teff value chain survey in 2012 (cross sectional)o 36 randomly selected villages in major teff producing areas o 720 randomly selected farmers
• Urban proximityo Transport cost o Cost of transporting 1 quintal of teff (ETB/quintal)
• Cities in survey areao Capital: Addis Ababao Secondary towns: Nazareth and Bahir Dar o Select city that maximizes net prices : ,
Teff Transaction Trip
17
Donkey Trip Truck Trip
Self-reported cost GIS Road Network
max , )?
μcf
Total Transport Cost T
mf, pf
Farmers and cities
18
ResultsIs there an effect of Secondary Towns?
Simulation of model
20
• Compare model simulation with empirical observation
• Teff price farmers receive in trader marketso Predicted: ,
o Observed: ,
• Truck cost from trader town to Addis Ababa
21
Predicted vs. Observed teff prices
400
600
800
1000
1200
1400
-100 -50 0 50 100Truck Cost to Addis Ababa
Predicted teff price net of transport cost
950
1000
1050
1100
1150
1200
1250
1300
-100 -50 0 50 100Truck Cost to Addis Ababa
Observed teff price
22
Other outcomes
0
50
100
150
200
250
-100 -50 0 50 100Truck Cost to Addis Ababa
Use of DAP (kg/ha)
0
10
20
30
40
50
60
-100 -50 0 50 100Truck Cost to Addis Ababa
Use of Improved Seeds (kg/ha)
800
1000
1200
1400
1600
1800
2000
2200
-100 -50 0 50 100Truck Cost to Addis Ababa
Teff Land Productivity (kg/ha)
5000
10000
15000
20000
-100 -50 0 50 100Truck Cost to Addis Ababa
Teff Profits (birr/ha)
ResultsSize of effect
Urban proximity & Agricultural Outcomes
24
• Effect of (total) transport cost To Y α β ∗ T X P Z ε ,o Y : improved seed usage, productivity and profitso X : controlso P : output and input prices o Z : zone fixed effects
• Effect β dependent on type of city?o Y α β ∗ T γ ∗ S π ∗ S ∗ T X P Z ε ,o S dummy for Secondary Townso β is effect for farmers shipping to Addis Ababao β π is effect for farmers shipping to Secondary Town
Endogeneity issues
25
• Heterogeneity between farmers in So Household and teff productiono Double Robust method
• Estimation of Propensity Score• Weighted regression
• Endogenous location of road networko Better infrastructure in areas with higher economic potentialo Instrumental Variable (IV) Approach
• ‘Natural path’ walking distance
Preliminary results
26
Secondary TownTransportation Cost to Addis
Ababa
Transportation Cost to
Secondary Town
Difference in effect
Teff price (ETB) –*** –*** –*** 0
Wages (ETB) 0 –*** –*** 0
Fertilizer (kg/ha) 0 –** –*** 0
Improved Seed (kg/ha) 0 0 –*** 0
Labor (days/ha) 0 –* –*** <0**
Yield (kg/ha) –*** –*** –*** >0**
Labor Productivity (kg/day) –** –*** 0 >0**
Teff Profits (ETB/ha) –*** –*** 0 >0**
Conclusions
Cities & Agricultural transformation in Ethiopia
28
• Impact of urban proximity on teff production decisions
• Difference between type of citieso Effect on prices is similaro Strong effect on input use in Secondary Townso Strong effect on labor productivity in Capital
• Way forwardo Improve empirical resultso Welfare implicationso Other mechanisms
Thank you!
Propensity Score
30
Secondary town (Yes=1)0.00
(0.00)0.03
(0.16)0.14
(0.08)-0.67***
(0.09)0.02
(0.04)0.05
(0.04)-0.10(0.11)
-0.55**(0.24)
-0.53***(0.11)
0.16***(0.05)
-0.29***(0.10)0.06
(0.07)-0.09***
(0.03)0.82
(0.70)
Age of head (years)
Educated head (years=1)
Head is from Oromia (yes=1)
Size of the household (number)
Children in the household (number)
Gender of head (male=1)
Constant
Household owns radio (yes=1)
Household owns tv (yes=1)
Household owns mobile phone (yes=1)
Wealth index (PCA of household assets)
Farm assets (ln of ETB)
Land owned (ha)
Farming Ability (.)
Balancedness of matching
31
.2 .4 .6 .8 1Propensity Score
Untreated: Off support Untreated: On supportTreated