Tool for updating yield growth parameters used in quantitative foresight models
Brief data tool summary and overview
The tool we present here provides insight into historic yields and the yields that are calculated by the IMPACT model based on the current available intrinsic productivity growth rates (IPRs). This is done for different countries and a variety of crop and livestock commodities. It serves as a starting point for discussing plausible and probable future yields
- In the rest of the page one finds crucial background information, including:
- * Definition of an IPR in the context of quantitative foresight models
- * Instructions on how to use the tool
- * Information about the underlying data sources
- * The IPR update methodology for generating improved intrinsic productivity growth rates needed to project future yields.
- * Some examples of applications
- * Some Dos and Don’ts
- * Annotated references and further reading
- * How to get involved in the process for updating and improving the IPRs
Interactive tool
What is an IPR
In the context of agriculture, an intrinsic productivity growth rate (IPR) refers to the rate at which the productivity of a farm or agricultural system increases as a result of improvements in technology, management practices, and other factors that are internal to the farm or system itself, rather than external factors such as changes in weather or market conditions. This growth rate can be a key indicator of the efficiency and competitiveness of an agricultural operation.
In quantitative foresight models, the intrinsic productivity growth rates (IPR) refer to the increases in productivity that are not influenced by model variables.
There is a relationship between IPRs and productivity (yield) growth rates but they are certainly not the same thing. The first step is identifying the drivers of productivity growth, i.e. understanding the factors that drive productivity growth in different agricultural systems, such as technology adoption, investment in research and development, changes in government policies, and access to markets and the related behavioral change such as application of fertilizers, irrigation, and improved seeds, or the application of improved agronomic practices.
The second step is determining which of these factors are part of the analytical system set-up. If a driver of productivity change is linked to a variable or scenario parameter, the the associated relationship between that variable or scenario parameter and yields needs to be identified. The change in productivity that cannot be explained by changes in variables or scenarios is considered as the intrinsic productivity growth rate.
IPRs are essential for quantitative foresight models such as IMPACT.
Fig 1. yields and intrinsic productivity growth rates of IMPACT model version 3.
As models evolve and other aspects are incorporated and as the world changes and scientific insights change, intrinsic productivity growth rates need to be revisited every now and then.
How to use the tool
The tool Home page section serves as the tool’s entry point, providing an overview of its objective. It introduces the user to the tool’s purpose of analyzing and updating exogenous yield growth parameters to 2050 and provides key references and citing information.
read more
Key Elements:
- Introduction (About): A brief explanation of the tool’s purpose, focusing on understanding the role of exogenous yield growth parameters in IMPACT yield projections.
- Navigation Buttons: On the left side of the page, four buttons allow users to navigate between the tool’s primary pages: Crops FAO, Crops Projections, Livestock FAO, and Livestock Projections. Clicking these buttons will take the user to the corresponding page, enabling quick and easy access to the desired data.
Navigation Guide:
Returning to the Home page: The user can return to the home page at any point by using the “Home” button in the left-side menu.
How to Navigate: The user can click on any of the four buttons on the left side of the home page to move to the corresponding page. Each button is labeled according to the page it leads to, and this simple navigation system allows users to move fluidly between different parts of the tool.
The Crops FAO section presents historical data on crop production, yield, and area harvested from 1961 to 2022 at global and country-specific level based on FAO data. It enables users to explore trends over time and compare production performance among countries.
read more
Key Elements:
Graphs
- Yield (ton/ha) by Year Graph: Located in the center, this scatter plot displays the yield trends over time, measured in tons per hectare, for the selected crop. It shows the evolution of yields from 1961 to 2022.
- Production (ton) x Countries Bar Graph: Positioned on the right side of the page, this bar graph ranks countries by their total production for the last three years (2020-2022). It displays the data in descending order, allowing users to identify the top producers.
- Harvested Area (ha) and Production (ton) by Year Graph: This combined line graph, found at the bottom right part of the page, shows the relationship between the area harvested (in hectares) and the production output (in tons) for the selected crop over time.
Functionalities
A. Time Filter (2020, 2021, 2022): At the top of the section, users can toggle between 2020, 2021, and 2022 to filter the bar graph and update the data displayed in the production rankings.
B. Countries Prioritization Button: This button, located above the Yield graph, provides additional functionality for users to view each country’s contribution to global production and cumulative participation.
C. Crop Filter: A crop filter dropdown menu can be found below the “Countries Prioritization” option. This filter allows users to select a specific crop from a predefined list. Once a crop is selected, all the graphs (Yield by Year, Harvested Area and Production, and Production by Country) will automatically update to display data related to the chosen crop.
D. Interactive Filtering: By default, all statistics displayed on the page correspond to global averages (yield) or aggregates (areas and production) for the selected crop. By clicking on any country in the Production x Countries bar graph, the Yield by Year, and Harvested Area and Production graphs are filtered to display data specifically for that country. To return to the global data view, users need to deselect the country.
How to Use:
Click the “Countries Prioritization” button above the Yield graph to further analyze country-specific contribution to global production for the selected commodity. For more information, please refer to section 2.6.
Select a crop from the dropdown menu to update all visualizations with data for that crop.
Use the bar graph on the right to explore which countries are the top producers of the selected crop in the most recent three years. Click on a country to filter the other graphs by that country’s data.
The line graphs below and in the center will automatically adjust to reflect global or country-specific trends in yield, production, and harvested area.
The Crop projections section is designed to show not only historical data but also projected trends for crop yield, harvested area, and production from 2020 to 2050. It allows users to explore future scenarios based on the Impact model projections, helping to visualize the potential effects of various factors, such as climate change and technological advances, on agricultural production.
read more
The layout and structure of the Crops Projections section are similar to the Crops FAO section, featuring the same elements such as:
- A Yield (ton/ha) by Year graph in the center.
- A Production (ton) x Countries bar graph on the right.
- A Harvested Area (ha) and Production (ton) by Year graph at the bottom.
What Changes:
- Projections: The primary difference is that this section includes projections for the years 2020 through 2050, based on the Impact model. Both the Yield and Harvested Area and Production graphs display these projections alongside the historical data, allowing users to analyze future trends.
The Livestock FAO section presents historical data on livestock production, yield, and the number of producing animals from 1961 to 2022 at a global and country-specific level, based on FAO data. It enables users to explore trends in livestock productivity and compare production across different countries.
read more
Key Elements:
Graphs
A. Producing Animals (Thousands) by Year Graph: Located in the center, this line graph displays the trend in the number of producing animals over time, measured in thousands, for the selected livestock product. The graph shows data from 1961 to 2022.
B. Production (ton) x Countries Bar Graph: Positioned on the right side, this bar graph ranks countries by their total livestock production for the last three years (2020-2022). It displays the data in descending order, allowing users to identify the top-producing countries.
C. Yield (Kg/Animal) by Year Graph: This line graph, found at the bottom, presents the yield trends for the selected livestock product over time, measured in kilograms per animal. It helps users assess livestock productivity by showing how yield per animal has evolved since 1961.
Functionalities
A. Time Filter (2020, 2021, 2022): At the top of the section, users can toggle between 2020, 2021, and 2022 to filter the bar graph and update the data displayed in the production rankings.
B. Livestock Filter: Below the “Countries Prioritization” button, there is a livestock filter dropdown menu. This allows users to select a specific type of livestock (e.g., bovine, poultry, etc.), and all the graphs will update accordingly with data specific to that type.
C. Interactive Filtering: By clicking on any country in the Production x Countries bar graph, the Producing Animals and Yield graphs are filtered to display data specifically for that country. To return to the aggregated data, users need to deselect the country.
D. Interactive Filtering: By clicking on any country in the Production x Countries bar graph, the Producing Animals and Yield graphs are filtered to display data specifically for that country. To return to the aggregated data, users need to deselect the country.
E. Countries Prioritization Button: Located above the Producing Animals graph, this button lets users view each country’s contribution to global livestock production and cumulative participation.
How to Use:
1. Using the Livestock Filter, select a specific livestock product from the dropdown menu. This will update all visualizations with data for that livestock.
2. Use the bar graph on the right to explore which countries are the top producers of the selected livestock in the most recent three years. Click on a country to filter the other graphs by that country’s data.
3. The line graphs below and in the center will automatically adjust to reflect global or country-specific trends in producing animals and yield.
4. Click the “Countries Prioritization” button above the Producing Animals graph to further analyze country-specific participation.
The Livestock Projections section presents both historical data and projected trends for livestock production, yield, and the number of producing animals from 2020 to 2050. It allows users to explore future scenarios using the Impact model projections, providing insights into how factors like climate change and technological advancements might influence future livestock productivity.
read more
This section shares a similar layout to the Livestock FAO section, featuring:
- A Producing Animals by Year graph in the center.
- A Production (ton) x Countries bar graph on the right.
- A Yield (Kg/Animal) by Year graph at the bottom.
What Changes:
- Projections: This section’s key difference is the inclusion of projections for the years 2020 to 2050. These projections are integrated into the Producing Animals and Yield graphs, allowing users to visualize future trends based on the Impact model .
Country prioritization is also possible in the tool.
Read more
Key Elements:
Graphs
- Table of country production statistics: It is located on the left part of the page. It shows the share of global production in each country for the selected commodity and the cumulative distribution of production across countries.
- Country Treemap: It is located on the right part of the page. It visualizes the contribution of each country to the global production of the selected commodity. This is done via rectangles, whose size represents the global production share of each country.
Dos and don’ts
coming soon
About underlying data sources
FAO
The data used in this tool comes from the Crops and Livestock Products database of FAOSTAT, the statistical division of the Food and Agriculture Organization (FAO). The information is sourced primarily from the elements related to crop and livestock production, covering historical data from 1961 to 2022.
read more
For crops, the specific elements used include:
5312 – “Area harvested” (hectares)
5410 – “Yield” (ton/ha)
5510 – “Production” (tons)[1]
For livestock, the following elements were grouped under the category of Producing Animals (measured in thousands of animals):
- 5320 – “Producing animals/slaughtered”
- 5318 – “Milk animals”
- 5313 – “Laying”
Livestock Production (element 5510) is measured in tons, consistent with the data used for crops. Yield for livestock is calculated by standardizing all production data into kilograms per animal, using the number of producing animals and the total production.
Categories of Crops and Livestock:
The crops and livestock products analyzed in this tool are categorized as follows. In the case of livestock, multiple FAO IDs are grouped together to represent a broader product category, such as Bovine Meat, Poultry Meat, or Eggs:
| FAO_ID | Crop Name | FAO_ID | Livestock Name |
| 486 | Bananas | 867, 947 | Bovine Meat |
| 44 | Barley | 1035 | Pig Meat |
| 176 | Beans, dry | 1091, 1062 | Eggs |
| 125 | Cassava | 977, 1017 | Mutton & Goat Meat |
| 191 | Chickpeas | 1058, 1069, 1073, 1083, 1080 | Poultry Meat |
| 661 | Cocoa, beans | 1058, 1069, 1073, 1083, 1080 | Milk |
| 656 | Coffee, green | ||
| 195 | Cowpeas, dry | ||
| 201 | Lentils | ||
| 56 | Maize | ||
| 79 | Millet | ||
| 197 | Pigeon peas | ||
| 489 | Plantains and others | ||
| 116 | Potatoes | ||
| 27 | Rice, paddy | ||
| 83 | Sorghum | ||
| 236 | Soybeans | ||
| 157 | Sugar beet | ||
| 156 | Sugar cane | ||
| 267 | Sunflower seed | ||
| 122 | Sweet potatoes | ||
| 15 | Wheat | ||
| 137 | Yams |
[1] All meat activities are expressed in terms of carcass weight.
IMPACT
The data used in this tool comes from the International Model for Policy Analysis of Agricultural Commodities and Trade (IMPACT).
Methodology overview: feathering
The IMPACT team relies on agricultural and economic experts to provide information onb how fast yields are likely to be growing 20 or more years into the future. But in the shorter term, the team believes that recent past yield growth is an important predictor of near-term yield growth, yet not the only important predictor of yield growth.
In 2017, during the last IR update of the IMPACT model, experts predicted growth over the 2005 to 2050 period. But using a Bayesian-inspired update method, in this IPR update, the IMPACT team blended FAOSTAT yield growth rates computed by regression for 2009-2018 for each country and commodity together with IPRs established by experts during the last update. The term “feathering” is inspired by use in painting in which two different colors transition from one to another, first with one color dominating, then with a more even blend, then the other dominating.
In the feathering methodology used for IMPACT, for the 2019-2024 period, the team multiplies 0.4 times the FAO regression rate and adds 0.6 times the expert-recommended IPRs from the previous IMPACT version. For 2025-2029, the team adds 0.3 times the FAO regression rate plus 0.7 times the expert-recommended IPRs. For 2030-2034, the team adds 0.2 times the FAO regression rate plus 0.8 times the expert-recommended IPRs. For 2035-2039, the team adds 0.1 times the FAO regression rate plus 0.9 times the expert-recommended IPRs. Then for 2040-2050, the team keeps using the expert-recommended IPRs.
In the previous version of the IMPACT model, in some cases experts established different growth rates for rainfed and irrigated for the same country and commodity. The FAO data does distinguish rainfed yields from irrigated yields, so we imposed the same growth rate for rainfed and irrigated on the FAO input to the updating process, but also used the different rainfed and irrigated rates for the expert-recommended IPRs. This allows for continuing differentiation between rainfed and irrigated yield growth, though in the feathered IPRs, the updating methodology forces the growth rates to be more similar than in the previous version of IMPACT.
Applications
The tool for updating intrinsic productivity growth rates was developed for updating the intrinsic productivity growth rates of the IMPACT model.
It was used in stakeholder engagement sessions to identify the possible and probable future yields under various scenarios.
Fig 2. Getting engagement from crop specialists
Getting involved
Open discussion on plausible and probable future yields continues on the discussion forum of the foresight community of practice. This is the way to get involved.
Annotated references
General documentation
The guidelines for updating the IPRs will be available online soon. We expect the links to be available before December 31, 2024: come back soon to find the links here.
Petsakos, A., Andrade, R., Alena, A., Pede, V., Hareau, G. 2024. Review and updating of the exogenous yield growth parameters of the IMPACT model: Guidelines to conduct the expert consultations. CGIAR Foresight Initiative report. Rome: Alliance Bioversity International and CIAT. 17 pages.
Rivera, T., Orozco, E., Petsakos, A., Hareau, G., Andrade, R. 2024. Manual for Exploring Yield Growth to 2050: Updating Exogenous Yield Growth Parameters in IMPACT. CGIAR Foresight Initiative report. Palimira, Colombia: Alliance Bioversity International and CIAT. 14 pages
Hareau, G. 2024. REVIEW AND UPDATING OF THE EXOGENOUS YIELD GROWTH PARAMETERS OF THE IMPACT MODEL. SLIDEDECK TO CONDUCT THE EXPERT CONSULTATIONS. CGIAR Foresight Initiative report. Lima, Peru: International Potatoe Center. 9 slides
Hareau, G. 2024. Análisis de tendencias de producción y rendimiento y proyección del sector XXXX en XXX hacia 2050: Actualización de los parámetros de crecimiento exógeno del modelo IMPACT. CGIAR Foresight Initiative report. Lima, Peru: International Potatoe Center. 18 slides
Methods
coming soon: <scientific publications and reports on the model structure, mathematics, mechanisms or code>
Model applications
coming soon: <list of publications and reports on model application>
Acknowledgements
the process for updating the IPRs was supported by the CGIAR Foresight Initiative in the period 2022-2024.
