Syndicate content

Add new comment

Our food system depends on the right information—how can we deliver?

Diego Arias's picture
Also available in: Español | Portuguese, International
Photo: CIF Action/Flickr
For most of us, watching the weather forecast on TV is an ordinary, risk-free and occasionally entertaining activity. The weatherman even makes jokes! But when your income depends on the rain or the temperature, the weather forecast is more than just an informative or entertaining diversion. Information can make or break a farmer’s prospects. Farmers get a sense of the risks they face down the road and plan their planting, harvest, use of inputs like fertilizers and pesticides, crop and livestock activities and market sales around weather reports and other information—on prices, local pests and diseases, changes in credit terms and availability, and changes in regulations, among other things.

The availability and quality of such agriculture risk information is hugely important for farmers, and the potential impact of bad information can be quite costly, leading the farmer to make wrong decisions and eventually lose revenue. Information systems that have unreliable sources and/or poor data processing protocols, produce unreliable results, no matter how complex the data processing model is. In other words, one can have “garbage in – garbage out.” Information is integral to agriculture risk management, not only in the short term to hedge against large adverse events, but also in the medium and long term to adapt to climate change and adopt climate smart agriculture practices. Climate-smart agriculture programs and agriculture risk management policies are toothless unless farmers have reliable information to implement changes on the ground.

Investing in agriculture risk information systems is a cost-effective way of making sure that farmers--and other actors along the food supply chain-- make the right decisions. But agriculture risk information systems in most countries suffer from lack of capacity and funding. Mexico, a country with an important agriculture sector, does not have information on market prices of agriculture products like maize, which is why a new Bank project aims to strengthen their capacity in this area. Mexico is not alone. Argentina solved this same problem recently with World Bank support, creating a market price information system for basic grains.

In Northeast Brazil, weather forecasts in rural areas are slightly worse than flipping a coin. Most weather reports are wrong. However, the problem is the delivery of the information, not the source. FUNCEME, the institution that collects and models the agroclimatic information for the Northeast region of Brazil, has an accuracy of more than 70%, but their mandate does not involve disseminating the information to individuals and broadcasters sometimes interpret it incorrectly. Here again, a World Bank project is looking to change that by improving the information delivery channels directly to farmers through mobile phone technology.

The good news is that governments are realizing the importance of agricultural risk information-- not just as a public service, but also to agribusinesses, which want production sources to have the best information possible to avoid business interruptions. The private sector is also seeing opportunities in agricultural information—there are startups like Agrinsurance, which offer informational services and work with agriculture insurance companies. Furthermore, the World Bank and others have been developing agriculture risk management assessments and strategies for countries across the world. Clearly, there is a need to strengthen information systems. What’s the best way to go about it?

There are several models. One is to build it entirely in house: The public sector collects, cleans, disseminates, and even creates applications and channels for farmers to access information. Embrapa in Brazil is following this approach, developing quite complex models like CONPREES, with World Bank support, to more accurately predict whether a specific crop in a specific municipality will be subject to agroclimatic losses in any given year.

A second approach, is to collect the information, clean it and make it available in raw form for the private sector to develop analytical tools to present to farmers and other actors. In the US, the National Oceanic and Atmospheric Administration (NOAA) does this with remote sensing information.

Finally, there is an “integration” approach, where the government integrates information systems spread out through public and private institutions, consolidating them and presenting them to the public in a coherent manner. This is something that Argentina has been doing, entering into public-private partnerships (PPPs) with local private institutions that have their own agroclimatic information networks in order to integrate the data with the public network and gain more density and historical time series.

Some countries may have a combination of such models or approaches, but it is important to notice that the private sector is more and more involved in the risk information business. This cannot be ignored by governments, and in fact should be embraced. But in order not to have “garbage out” from agriculture risk information services, it is important that public institutions ensure that what goes in is gold and that dissemination to farmers is done properly. Below is a proposed structure of components that need to be in place in the agriculture information service of any given country to ensure that the risk information is of good quality. Meanwhile, the next time you listen to the weather forecast, know that for some, this is an important affair. Farmers –and the food system—depend on it.
 
Integration Ensuring that the existing networks within a country are integrated, even if they are from different institutions. This can require inter-institutional agreements, PPPs, and others to ensure protocols and homogenization of data collection and processing
Integrity Once the data is collected, can it be tampered with?
Completeness Ensuring that historical and spatial data series are complete. If gaps exist, establishing protocols to fill out those gaps with industry-recognized methodologies, validated by experts.
Accessibility Users should have access to current and historical data and be able to reuse the data in any form in order to undertake analysis and services. This may require creating a platform for linking with online tools created by others.
Modularity Enabling the system to add modules or sections as new variables and/or information becomes available.
Timeliness The information needs to be available relatively quickly in order to have value for services such as weather insurance, updating of forecasting, etc.
Auditability The information system needs to be easily traceable to the source, with clear protocols in place.