Impact Assessment of Africa RISING Seed & Fertilizer Technologies in Tanzania with TOA-MD Analysis

Ex-ante impact assessment of Africa RISING seed & fertilizer technologies in Tanzania with Trade-off Analysis Model for Multi-Dimensional Impact Assessment (TOA-MD) Lieven Claessens-IITA, Srabashi Ray-Oregon State Univ., Job Kihara-Allisnce Bioversity- CIAT, Anthony Kimaro-ICRAF, Julius Manda- IITA & John Antle-Oregon Univ. Africa RISING ESA Project partners update meeting 3 June,2021-Virtual

Context Over the years, many experiments/trials have been conducted on efficient/sustainable use of agronomic inputs (organic & inorganic fertilizer, ISFM, water ) in Africa RISING However, these experiments took place in geographically limited agro- ecological zones Taking validated technologies to scale is at the core of the project Ex ante impact assessment can provide a valuable contribution to more targeted scaling

Methodology: Tradeoff Analysis model for Multi- Dimensional impact assessment (TOA-MD) 1) Ex ante impact assessment tool ( bio-economic model ), compares alternative systems (can be now or in the future climate change e.g.) 2) Assesses technology adoption and associated economic, environmental and social outcomes (e.g., income, poverty rates, carbon sequestration, ) 3) Inputs: household survey data, simulation models (climate, crop, livestock), experimental data, 4) Heterogeneous population of farms: distributional outcomes 5) Population can be stratified to look at differential impacts (e.g., livestock ownership, farm or HH size, off farm income, gender, .)

Methodology: Tradeoff Analysis model for Multi-Dimensional impact assessment (TOA-MD)

Data sources Worldbank LSMS household survey (3,924 HH) quantities and prices of inputs (seeds, labor, fertilizer, manure, feed), outputs (crop yields, milk production and land areas), and farm management Africa RISING fertilizer trial data for Babati, Kongwa and Kiteto NP fertilizer recommended rates, hybrid maize seeds, optimal agronomic management, P with and without added micronutrients (Di- Ammonium Phospate DAP-Minjingu Mazao MM). Grouped agro-ecologies to tropic-cool/sub- humid and tropic-warm/sub-humid

Model parameterization System 1 = HH survey System 2 = AR technology Average values, variation important Optimistic scenario (project managed, inputs and extension provided .) vs realistic scenarios considered (60 and 80 %) Input and output price elasticity considered 6000 5348 5018 5000 4558 4233 4000 3000 2000 792 1000 748 689 657 0 System 1: Non-Hyrid users System 1: Hybrid users System 2DAP: Hyrbd seed with N and P (DAP) application Tropic-cool/sub-humid System 2MM: Hybrid seed with N and P (MM) application Tropic-warm/sub-humid

Model Parameterization System 1 System 2DAP System 2MM Warm region: Non-hybrid users Warm region: Hybrid users Cool region: Non-hybrid Cool region: Hybrid users Warm region Cool region Warm region Cool region A b users c xxxx avg(Yield) 123 245 4233 657 689 748 792 5348 5018 4558 sd(Yield) 528 615 530 552 2233 1952 2536 1796 CV(Yield) 80 89 71 70 53 36 51 39 yyyy avg(Variable cost) 241 424 576 20 68 20 56 622 486 508 zzzzz avg(Net Revenue) 234 234 440 138 97 160 134 719 662 585 CV(Net Returns) 129 181 118 127 129 129 118 118 avg(FEXP) 286 358 240 295 414 342 422 339 CV(FEXP) 105 93 124 116 105 105 124 105

TOA-MD: adoption rates DAP Economic adoption rates based on net returns 70 59 58 60 54 53 50 47 50 44 43 41 40 37 40 30 26 DAP adoption falls sharply with realistic yield levels in warm agro- ecology, MM more suitable (price, yield levels and variability) 30 17 20 8 10 0 Overall Strata 1: Warm region, non- hybrid users Strata 2: Warm regions, hybrid users Strata 3: Cool region, non- hybrid users Strata 4: Cool regions, hybrid users Potential Yield Scenario Y80 Scenario Y60 MM fertilizer estimated to have higher adoption across agro- ecologies (also confirmed by price elasticity scenarios) MM 56 60 53 50 49 49 48 46 46 44 50 44 43 36 40 34 32 29 30 20 10 Potential to disaggregate adoption rates according to farm typologies to inform scaling 0 Overall Strata 1: Warm region, non-hybrid users Strata 2: Warm regions, hybrid users Strata 3: Cool region, non-hybrid users Strata 4: Cool regions, hybrid users Potential Yield Scenario Y80 Scenario Y60

Results Overall, 47 50 percent maize farms in sub-humid conditions are likely to adopt the seed, fertilizer and management recommendations MM fertilizer has a marginally higher adoption rate DAP is relatively more suitable in Cool regions while MM is more suitable in Warm regions In the optimistic high yield scenario, the difference in predicted adoption rates for DAP and MM is small Under more realistic yield scenarios (60 and 80% of potential yield), DAP fertilizer adoption falls sharply in Warm regions (to 8 17 percent) relative to MM fertilizer (29 36 percent) In Cool regions the fall in adoption rate of MM fertilizer under realistic yield scenarios is relatively smaller Better adoption potential of MM fertilizer across agro-ecological conditions under realistic yield scenarios (policy/extension implication)

Cont. The improved technology recommendations are likely to increase the value of food expenditure per capita (as an indicator of food security) by 16 percent Under realistic yield conditions, this falls to 10 13 percent MM fertilizer has higher impact on food expenditure in Warm regions, while DAP has a higher impact on the food expenditure by farms in Cool regions

Main points for discussion Yield levels: experimental yields vs on farm yields worked with yield scenarios Generalization/grouping of agro-ecological zones limited observational data (in the process of cross checking some recent data (coordinates) from KK) Outcomes of the TOA-MD analysis are expressed as averages across the population (~4000 HHs), variation is important!

Thanks for your attention! Any questions?

Thank You Africa Research in Sustainable Intensification for the Next Generation africa-rising.net This presentation is licensed for use under the Creative Commons Attribution 4.0 International Licence.