Excellence-based Climate Change Research Prepared for the African Green Revolution Workshop Tokyo, Japan Dec 7-8, 2008.

Excellence-based Climate Change Research

Prepared for the African Green Revolution WorkshopTokyo, JapanDec 7-8, 2008

An African Green Revolution

:What drives farm productivity: natural endowments, technology and capital, knowledge, or policies

S. Niggol Seo, PhDResearch 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

5

A profit maximizing farmer:

  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:

IPLXIQPMax IQ ),,(

IQ PIQP )/(

),,,( LXPPII IQ

),,,( LXPP IQ

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. Access to capital : distance to nearest city (population over 100,000)

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.

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

16

Portfolio Number of Farms Percentage

Beef Cattle Only 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%

Dairy Cattle 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%

African Primary Species Choice

17

PLOT CATTLE_ MEAT CATTLE_ MI LK GOATSSHEEP CHI CKEN

P( Xi =1)

0. 0

0. 1

0. 2

0. 3

0. 4

0. 5

0. 6

0. 7

0. 8

0. 9

1. 0

T_ MEAN_ YR

12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

South American Species Choice

18

Pr obabi l i t i es of Owni ng Each Speci es BeefSheep

0. 0

0. 2

0. 4

0. 6

0. 8

1. 0

Annual Mean Temper at ur e ( degC)

0 10 20 30

Thank you!

Photo Credit: World Bank

19