ETHIOPIAN DEVELOPMENT RESEARCH INSTITUTE The rapid expansion of herbicide use in smallholder agriculture in Ethiopia Seneshaw Tamru, Bart Minten, Fantu Bachewe, and Dawit Alemu 14 th International Conference on the Ethiopian Economy Ethiopian Economics Association July, 2016 Addis Ababa 1
20
Embed
The rapid expansion of herbicide use in smallholder agriculture in Ethiopia
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
ETHIOPIAN DEVELOPMENT RESEARCH INSTITUTE
The rapid expansion of herbicide use in smallholder agriculture in
EthiopiaSeneshaw Tamru, Bart Minten, Fantu Bachewe, and Dawit Alemu
14th International Conference on the Ethiopian EconomyEthiopian Economics AssociationJuly, 2016Addis Ababa
1
2
1. Introduction Fast transformation of agriculture in the
last decade
Modern inputs have played an important part- Fertilizer and improved seeds
But, very little attention towards agro-chemicals
They however play a crucial role in modernizing farming systems in developing countries
3
Why study this? Most of the research on the
environmental and health implications Little evidence on the economic
determinants of adoption of agrochemicals
Therefore, Does herbicide adoption lead to better
productivity? What determines the adoption?
We study the case of Ethiopia and teff
4
2. Data Two types of sources: A. Qualitative information-interviews with:
EIAR, Adami Tulu, private traders, etc
B. Quantitative1. Customs/Comtrade data (imports)2. AGP- 8,000 HHs in high potential areas3. ESSPβs 2012 teff survey-5 major zones:
1200 HHs
5
2. Methodology-1 - Production Function and Labor Productivity
Where: πΏπππ πs the logarithm (quintal)/labor productivity of producer i on a plot j. ππππΎ are other factors that influence teff production/labor productivity including soil type, plot slope, seed type, logarithms of altitude, fertilizer, and seed. ππ is individual specific effect, ππ plot specific effect, πππ a stochastic error term.
4.(a). Labor productivity effects -Weeding and herbicide use in teff
4.(a). Released Labor 10 152 015
29.52 994 844
Non Adopters AdoptersFemale weeding labor days/ ha 20 8Male weeding labor days/ ha 27 17
Non Adopters [millions]
Adopters [millions]
Saving on labor (persondays) [millions]
Average rural wage rate=birr 50/ day [millions]
Average herbicide= birr 68/ ha [millions]
Total women days for weeding 59.3 24.3 35.0 1 750.6 Total men days for weeding 79.7 49.9 29.8 1 492.0 Total labor 139.1 74.2 65 3 243 204 3 039
Cereal Area in 2014/ 15 % herbicide applied area herbicide area
4.(a). Labor productivity effects -Higher labor productivity-especially of
weeding labor Weeding labor only All labor
Weeding labor only All labor
Coeff. Coeff. Coeff. Coeff.Herbicide use (Dummy) yes=1 0.10** 0.04*** 0.07** 0.00Area of plot log (hectare +1) 0.66*** 0.15*** 0.67*** 0.18***DAP use log(kg +1) 0.15* 0.07*** 0.14** 0.08**UREA use log(kg +1) 0.06 0.04* 0.15*** 0.03Interaction (fertilizer and herbicide use) -0.03* -0.01* -0.03** -0.00Cost to addis ln(birr/qt) -0.04* -0.01*Rural wage ln(birr) 0.19*** 0.06***Land rent ln(birr) 0.11*** 0.02***Distance to nearest cooperative ln(minutes +1) -0.04*** -0.01***
Household characteristicsSlope of the plotOther plot charachteristics (soil easy to plow, teff last year)Application of manure/organic fertilizerColor of seedSeed varietyRain and pest patterns Color of soil
Seed use ln(kg +1) 0.21 6.34*** 0.18 7.63***DAP use ln(kg +1) 0.33 4.68*** 0.32 3.87***UREA use ln(kg +1) 0.27 3.77*** 0.30 4.29***Interaction (Fertilizer and herbicide use) -0.05 -3.17*** -0.05 -3.70***Labor use ln(person hours) 0.24 7.45*** 0.17 6.84***Gender of head 0.09 1.76*Education of head (1=edu, 0=no ) 0.05 2.28**Household is a model farmer yes=1 0.05 2.22**Distance to nearest cooperative ln(minutes +1) -0.03 -2.19**Land owned ln(hectares +1) -0.08 2.66***
Other household characteristicsSoil easy plow and whether planted teff last yearRain and pest patternColor of soilSlope of the plot No. of weeding, app.maure/organic fertilizerColor of seed
5. Drivers 2. The increased market access by Ethiopian farmers
-Travel time to a city of at least 50,000 people (1996/97 and 2010/11)
5. Drivers. Factors determining Adoption
Transport cost to Addis log(birr/qt) -0.578*** -3.534*** -2.040***Wage rate (village level) log(birr/day/person) 2.982*** 9.161*** 5.287***Distance to allweather road log(minutes +1) -0.103*** -0.526 -0.303Altitude log(meters +1) -2.584*** -5.886 -3.397Land size log(hectares +1) 1.549*** -8.876*** -12.67***Household size log(number) 0.412*** -0.275 -0.159Share of children log(share +1) -0.310 4.884* 2.819Total land owned log(hectares +1) 0.977*** 2.381** 1.374**Asset log(birr) 0.057*** 0.143 0.082
Other household characteristics Yes YesSlope of plot, color of seed, seed variety Yes YesApplied manure/organic fertilizer Yes YesSoil easy to plow, planted teff last year Yes YesRain and pest pattern Yes YesDistances to market and agric.coop Yes Yes
Dependent variable=Herbicide Application (yes/no dummy or amount)
Unit
Additional controls for:
Determinants of decision to use (yes/no) herbicide
Factor determining amount (value) of herbicide use
Average (APE)
6. Conclusions β’ Rising use of herbicides in the country, mostly on commercial cereals.
β’ 2-4-D has been most widely used
β’ Herbicide application leads to improved production, labor and land productivity.
β’ Transport cost to a big city, rural wage rate levels mainly influence the adoption (and amount) of herbicide.
7. Policy Implications β’ First, nationally, about 65 million person-days would be saved.
β’ If herbicides will be further applied, this will lead to significant further release of labor in rural areas, with significant impacts on labor markets, and urbanization.
β’ Second, there are reported issues with quality that should be monitored.
β’ Third, application of herbicides is often used without proper protection. This might lead to health implications for those farmers that do so.