Transcript PPT

Excellence-based
Climate Change Research
Prepared for the African Green Revolution Workshop
Tokyo, Japan
Dec 7-8, 2008
An African Green Revolution
:What drives farm productivity: natural endowments,
technology and capital, knowledge, or policies
S. Niggol Seo, PhD
Research Professor
* Based on the draft proposal by Robert Mendelsohn and Niggol Seo .
Agriculture in Africa
 Agriculture in Sub-Saharan Africa accounts for
 70 percent of employment,
 40 percent of exports and
 about one-third of economic growth from 1990 to 2005.
 Besides, more than 70 per cent of the continent’s poor people live in
rural areas.
* World Development Report (2008)
3
Low agricultural productivity

Farming in Africa is less productive than in other regions (Sachs et al 2004)

Although agriculture in other parts of the world has benefited from productivity
growth owing to capital investments and the Green Revolution over the past
decades, productivity has stalled or declined in many African communities

On average, value added per agricultural worker now averages around 12%
below 1980 levels in Sub-Saharan Africa.

Investment rates for new technologies have declined in recent years and
technology adoption rates are low when compared to other regions.
4
Economic Theory


A profit maximizing farmer:
Max   PQQ( I , X , L)  PI I

where PQ are the prices of outputs, Q, PI are the prices of inputs, I, X is a vector of
exogenous factors, and L is the available land to the farmer. We define the
productivity of the farm in terms of π.

FOC:

Input demand functions

Net revenue function as the locus of profit maximizing choices given the exogenous
variables:
PQ (Q / I )  PI
I  I ( PQ , PI , X , L)
   ( PQ , PI , X , L)
5
Economic Theory

Empirical question is which set of exogenous variables are most important: prices or
availability of inputs, natural endowments, knowledge, or policies.

Define net revenue broadly to include the value of own consumption. Own consumption is
valued at market prices.

Wages for own labor: However, we do have observed hours of own labor so we can
examine it as an input.

Farmers are broadly defined into two categories: a farm with a green revolution variety or a
farm with a traditional variety. A famer will choose a green revolution variety if it is more
profitable for the farm.
6
Empirical Model

Ricardian functions: we will regress crop net revenue per hectare on a set of exogenous
variables that reflect each of the competing hypotheses.

Climate: seasonal temperature and precipitation using linear and quadratic variables.

Soils: a set of soil measures provided by FAO for Africa.



Input prices: hired wage rates.

Knowledge: education, experience, and extension services.

Unfortunately, we do not have a good set of policy measures. We could use rankings of
governance. We could also include country dummy variables.
Access to capital : distance to nearest city (population over 100,000)
7
Empirical Model
 Whether the coefficients for a specific hypothesis are statistically
significant.
 How much of the variance in productivity across the sample is explained
by each set of variables.
 Input demand functions for capital, irrigation, modern crop varieties,
hired labor, and household labor. For inputs that are continuous, we will
examine both a linear and loglinear model. For inputs that are discrete,
we will use a logit model.
8
Data

Household survey of over 10,000 farmers in 11 countries in Africa
- Conducted under the supervision of CEEPA at the University of Pretoria and
- Yale University with the help of researchers from 11 African countries
- Burkina Faso, Cameroon, Egypt, Ethiopia, Ghana, Kenya, Niger, Senegal, South Africa,
Zaire, and Zimbabwe
- Funded by the GEF and the World Bank
- Information about net revenues, farming practices, and technology choices.

Soils from FAO

Climate from the National Oceanic and Atmospheric Association’s Climate
Prediction Center

Hydrology from the University of Colorado
9
Remaining questions
 Availability of information about modern seed varieties.
 Institutional variables
 Policy variables
10
Experiences from Africa and Latin America
S. Niggol Seo
11
South American Data
 The World Bank project on the impact of climate change on Latin
American agriculture (Seo and Mendelsohn 2008)
 Household surveys from seven countries (N=2300)
• Southern Cone: Argentina, Brazil, Uruguay, Chile
• Andean: Ecuador, Colombia, Venezuela.
 Includes all the major agro-ecological zones.
 Climate Data: US Department of Defense satellites, WMO
 Soils: FAO
12
13
African Data

Household survey of over 10,000 farmers in 11 countries in Africa
- Conducted under the supervision of CEEPA at the University of Pretoria and
- Yale University with the help of researchers from 11 African countries
- Burkina Faso, Cameroon, Egypt, Ethiopia, Ghana, Kenya, Niger, Senegal, South Africa,
Zaire, and Zimbabwe
- Funded by the GEF and the World Bank
- Information about net revenues, farming practices, and technology choices.

Soils from FAO

Climate from the National Oceanic and Atmospheric Association’s Climate
Prediction Center

Hydrology from the University of Colorado
14
15
Species ownership in South America
Portfolio
Number of Farms
Percentage
189
25.1%
Beef Cattle and Chickens
23
3.1%
Beef Cattle and Dairy Cattle
52
7%
Beef Cattle and Sheep
93
12.5%
Beef Cattle and Pigs
64
8.6%
Chickens
65
8.7%
101
13.5%
Dairy Cattle and Pigs
52
7%
Pigs
22
2.9%
Sheep
29
3.9%
Pigs, Dairy Cattle and Breeding Bulls
26
3.5%
Pigs and Chickens
27
3.6%
Beef Cattle Only
Dairy Cattle
16
African Primary Species Choice
P( Xi = 1 )
1. 0
0. 9
0. 8
0. 7
0. 6
0. 5
0. 4
0. 3
0. 2
0. 1
0. 0
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
T _ ME A N_ Y R
P L OT
CA T T L E _ ME A T
S HE E P
CA T T L E _ MI L K
CHI CK E N
GOA T S
17
South American Species Choice
1. 0
Pr o b a b i l i t i e s
of
Own i n g
Ea c h
Sp e c i e s
Be e f
Sh e e p
0. 8
0. 6
0. 4
0. 2
0. 0
0
10
20
An n u a l
Me a n
T e mp e r a t u r e
30
( d e g C)
18
Thank you!
Photo Credit: World Bank
19