With about one-third of all food produced around the world lost or wasted each year, reducing food loss and waste is a key component in ending hunger and malnutrition. A major hurdle, however, stems from the fact that food loss and waste are complex issues, and thus properly measuring them and identifying where in the food system they occur remain a challenge. Food loss and waste have been defined in various ways in the literature, and there has thus far been no single measurement methodology agreed upon. As a result, data regarding food loss and waste remains inconsistent and success stories of decreased food loss remain few.

In a recent conference paper funded by the CGIAR Research Program on Policies, Institutions, and Markets (PIM), researchers from IFPRI, the University of Antwerp, and the World Bank propose several new methodologies to reduce measurement errors and to assess the magnitude of food loss across multiple nodes in the food value chain. The methods account for losses from the pre‐harvest stage through product distribution and include both quantity loss and quality deterioration. The instrument is applied to producers, middlemen, and processors in seven major staple food value chains in five developing countries: potatoes in Ecuador and Peru, beans and maize in Guatemala and Honduras, and teff in Ethiopia.

In the first methodology (the self-reported method), the research team designed a set of surveys to capture detailed data regarding farmers’, middlemen’s, and processors’ different activities and to collect self-reported measures of the volumes and values of food lost during each of these activities. This allowed them to quantify loss along each node of the value chain.

The second methodology (the category method) collected data from each value chain actor regarding the quality of agricultural commodities that they use as inputs and outputs. This allowed the researchers to estimate losses based on commodity damage.
Finally, the research team captured information regarding different types of commodity attributes (size, broken grain, impurities, etc.) and determined the price penalty that these types of crop damage cause, allowing them to quantify food quality loss based on market conditions (the attribute method and price method). While all of these methodologies were tailored to specific countries and commodities and commodity varieties, they provide consistent approaches that are comparable across commodities and regions, thus making them useful for developing countries beyond the five studied specifically.

Across all of the studied value chains, losses ranged between 6 and 25 percent of total production and of total produced value. Losses were consistently largest at the producer level and smallest at the middleman level. Across the different methodologies, producers’ losses ranged from 60 to 80 percent of all losses, while middlemen’s losses ranged from 7 and 19 percent. At the producer level, the largest losses were consistently found at the pre-harvest stage of the value chain, suggesting that addressing food loss at this node of the staple food value chain is particularly important.

Interestingly, self-reported losses were systematically lower than losses estimated using the other three methodologies. This gap was particularly large in the Honduran bean value chain, where self-reported loss estimates were between 10 and 15 percentage points lower than estimates found using any of the other methods. This finding supports the authors’ argument that a range of data collection methods is needed to gain a true understanding of the reality of food losses.

The study also attempts to identify major causes of food loss along different nodes of the studied value chains. For pre-harvest losses, farmers reported pests, disease, and lack of rainfall to be the leading factors behind their losses; the exception to this finding were teff producers in Ethiopia, who reported only lack of rainfall as the major determinant of pre-harvest losses.

When asked about product left in the field, farmers identified lack of appropriate harvesting technique as the determining factor. Workers’ lack of training and experience in crop selection was identified as the most common reason for damage and loss during post-harvest, followed by pests and disease.

The study also found that farmers’ education and experience is correlated with a reduction in losses in many cases. In the potato value chains in Ecuador and Peru and the maize value chain in Honduras, a farmer’s years of education were particularly significant. Similarly, the number of years in which a farmer has been involved in the production of a specific crop was significantly correlated with a reduction in losses in the potato value chains in Ecuador and Peru, the maize value chain in Guatemala, and the teff value chain in Ethiopia.

The cost to reach the market also appears to play a role in producers’ food losses; this factor is significantly correlated with increased losses in Peru, Guatemala, and Ethiopia. This finding supports previous research suggesting that access to improved roads and rural infrastructure can reduce food losses.

Improved seeds can also play a role in reducing food losses; the use of pest- and drought-resistance seed varieties reduced losses in Ecuador’s potato value chain and in maize and bean value chains in Honduras. However, use of improved seeds varies from country to country. More than 65 percent of producers from Peru, Ecuador, and Ethiopia used improved seeds in the last crop season (for potato and Teff, respectively); however, less than 20 percent of maize and beans producers from Guatemala and Honduras used improved seed during the study period.

Finally, the authors emphasize that credit constraints and lack of education appear to be underlying macro-causes behind food losses in staple value chains in developing countries. These constraints can prevent farmers from becoming aware of, understanding, and affording food loss reduction technologies.

The study has numerous policy implications. First, it emphasizes the important role that properly measuring food loss plays in understanding and solving the problem. The finding that farmers often appear to underestimate the magnitude of losses on their farm suggests that research and policies dependent only on farmers’ self-reported data will also underestimate the extent of the problem.

Second, it is clear that pests, disease, too much or too little rainfall, and poor harvest and post-harvest techniques appear to be the most common causes of food loss at the producer level. Thus, it will be important to improve farmers’ ability to handle these challenges; this could include increasing extension services to educate farmers about proper pest management and harvesting techniques and improving farmers’ resilience to rainfall shocks.

Finally, policymakers will need to address broader underlying factors that contribute to food loss, including poor road infrastructure, lack of market access, and lack of credit.

By: Sara Gustafson, IFPRI

Post new comment
The content of this field is kept private and will not be shown publicly.
Share