Follow priorities in the field to find a topic of interest
In recent years, food loss and waste have been high on the global agenda. The United Nations’ Sustainable Development Goal 12 included an ambitious target to cut food waste and loss in half by 2030 (SDG Target 12.3). The statistics presented by the Food and Agriculture Organization of the United Nations (FAO) were stunning: globally, a third of the food produced for human consumption is lost or wasted and enough food is lost or wasted to feed 2 billion people around the world (1). If food lost and wasted were a country, it would be the third largest emitting country in the world (2). Seeing these statistics, I became very invested in this topic and wanted to find data so that I could learn more about food loss and waste.
Search the existing literature to find out what is known
First, I delved into the definitions – what are food loss and waste? Food loss is formally defined as all edible food discarded along the supply chain before the retail stage (3). On the other hand, food waste is all edible food discarded during food retail, food service, and household consumption (4). Food loss is estimated to be a bigger problem than food waste (1), so I decided to focus on food loss.
Next, I went to Google Scholar to search for relevant literature and came across an article by Cattaneo and colleagues (5). This article, along with other literature, showed me the difficulties in doing research on food loss and waste. The definitions of food loss and waste vary across organizations; most governments do not collect sufficient data on food loss and waste; and the available data are of poor quality because of weak monitoring and management (5–7).
Use background knowledge to search for data
Through this background research, I figured out that FAO and the United Nations Environment Programme (UNEP) work together to monitor progress to reduce food loss and waste. Progress on SDG Target 12.3 is monitored through two indicators: the Global Food Loss Index, overseen by FAO, and the Food Waste Index, overseen by UNEP. Since I am focusing on food loss, I dug into FAO’s data.
FAO, in collaboration with national statistical offices, tracks food supply, utilization, and consumption at the country level over time using Food Balance Sheets (FBS) (8-9). FBSs can also be used to look at food loss from harvest to retail, though most of the food loss data is imputed due to limited data availability. Thus, while comprehensive, FBSs only provide an aggregate picture of food loss.
Access data about food loss
I decided to extract data from FAO's Food Balance Sheets learn more about food loss. On the SCANR Research Guidance Finding data section, there is a list of data repositories where you can find publicly available data related to agriculture, food systems, nutrition, and health. There, I found FAOSTAT, an online repository of food and agriculture data collected and maintained by the FAO.
From the main page, I went to the “Data” page, on which many domains of data are listed. From there, I typed “food balance” in the search box. You can also scroll down to find the “food balance” dataset.
For obtaining food losses, I used the New Food Balance Sheet.
As noted in the SCANR Research Guidance section, it is always good practice to check the metadata, which provides users with information about the dataset. For instance, this is where I found the definition of FBS, the units of measure, and the frequency of data collection.
Under the “Download data” tab, I selected “Losses” from the “Elements” box by filtering or scrolling down for “Losses.” There are other boxes to select from, such as countries, regions, or special groups (e.g., United Kingdom, EU, low-income country), items (food commodities, such as cereals), and years (2014-2018).
After making a selection from each box, I downloaded the data. The default options are fine, but if you have preferences, you can adjust them by “output type,” “file type,” “thousand separator in ‘show data,’” and/or “output formatting options.”
Once the download was complete, I opened the dataset in Excel, and this is what it looked like: the first row listed the variable names, and each column had its own values. The “Area” column tells me who, the “Item” column tells me what, the “Year” column tells me when, and the “Unit” and “Value” columns tell me how much food loss was occurring.
From this dataset, I decided to visualize how much food is lost in the Least Developed Countries (LDC) by food item from 2014-2018.
*Tip: If you are wondering which countries are classified as Least Developed Countries, you can look it up under the “Definitions and standards” tab on the FAOSTAT homepage.
On the SCANR Research Guidance Analysing data page, I found a list of online resources for using statistical software, including tips on making graphs in Excel. First, I identified that my x-axis on the graph would be the years, and my y-axis would be the food loss values for each item from 2014 to 2018. I selected the values for the cereals and then clicked “Charts” under the Excel “Insert” tab.
Once I selected a bar graph, the default bar graph pops up with the correct y-axis and x-axis. Even if this was not the case, I can easily correct it by right clicking on the graph and selecting “Select data.”
Another window pops up, which allows me to add, edit, and even remove data for the y-axis (“legend entries”) and x-axis (“horizontal (category) axis labels”).
I wanted to provide the item names for the each of the values so that later the legend for each bar will be automatically named, so I designated the “Series name” as “Starchy roots” under the “Items” column and the “Series values” from the “Value” column from 2014 to 2018. I repeated this step for each item.
This is what my graph looked like after I added all of the food items and values. Then, I decided to add a legend and axis titles.
Under the “Chart design” tab, there is an option to add elements to the graph by clicking “Add chart element.”
For this graph, I added both “Primary horizontal” and “Primary vertical” titles. Then, I added a legend to the right by going to “Legend” and selecting the “Right” option.
From this graph, I can learn a lot about food loss in the Least Developed Countries. For example, food losses were greatest among cereals and starchy roots, and the magnitude of food loss did not change too much between 2014 and 2018.
Keep researching and exploring data
Through this process, I was able to learn about food loss and waste, find data related to food loss, and visualize food loss data in different countries and regions. These datasets have the potential to be combined with other datasets to answer more specific questions. For instance, future research projects could investigate which food group is wasted most frequently and how food loss and waste relate to food production; illustrate food loss in relation to food insecurity; explore the relationship between food loss and climate change; or examine how food loss is related to health and nutrition. If this topic interests you, I recommend that you to explore the NUTRI-P-LOSS project (11), as well as the African Postharvest Losses Information System (APHLIS) (12). The NUTRI-P-LOSS project demonstrates how to combine food loss data with food composition data to examine nutrient loss and calculate lost opportunity for eating healthy diets. APHLIS is another online data repository specifically for sharing data on postharvest losses in Sub-Saharan Africa. In APHLIS, you can retrieve food loss data for specific regions, crops, and years. This dataset is created via postharvest loss profiles from peer reviewed literature, as well as contextual observations from local experts (12).
- FAO. global food losses and food waste. Extent, causes and prevention [Internet]. 2011 [cited 2021 May 31]. Available from: http://www.fao.org/3/i2697e/i2697e.pdf
- FAO. Food wastage footprint: Impacts on natural resources - Summary report [Internet]. 2013 [cited 2021 May 31]. Available from: www.fao.org/publications
- Fabi C, English A. Working Paper Series ESS / 18-13 METHODOLOGICAL PROPOSAL FOR MONITORING SDG TARGET 12.3. SUB-INDICATOR 12.3.1.A THE FOOD LOSS INDEX DESIGN, DATA COLLECTION METHODS AND CHALLENGES [Internet]. [cited 2021 May 31]. Available from: http://www.wipo.int/amc/en/mediation/rules
- United Nations Environment Programme. Food Waste Index. 2021. 8 p.
- Cattaneo A, Sánchez M V., Torero M, Vos R. Reducing food loss and waste: Five challenges for policy and research. Food Policy. 2020;98(September 2020).
- Lusk JL, Ellison B. Economics of household food waste. Can J Agric Econ. 2020;68(4):379–86.
- Parfitt J, Barthel M, MacNaughton S. Food waste within food supply chains: Quantification and potential for change to 2050 [Internet]. Vol. 365, Philosophical Transactions of the Royal Society B: Biological Sciences. Royal Society; 2010 [cited 2021 Mar 24]. p. 3065–81. Available from: https://royalsocietypublishing.org/
- Tayyib S, Golini N. The FAO approach to food loss concepts and estimation in the context of Sustainable Development Goal 12 Target 3. Pap Present to 17th Int Conf Agric Stat (Rome, Italy 26-28 Oct 2016) [Internet]. 2016;(2011):7pp. Available from: http://www.fao.org/3/a-bt613e.pdf
- Food Balance Sheets (FBS) | INDDEX Project [Internet]. [cited 2021 May 31]. Available from: https://inddex.nutrition.tufts.edu/data4diets/data-source/food-balance-sheets-fbs
- FAO. New Food Balances Description of utilization variables [Internet]. [cited 2021 May 31]. Available from: https://www.oilworld.biz/t/publications/data-base.
- NUTRI-P-LOSS: Nutritional Postharvest Loss Estimation Methodology | ANH Academy [Internet]. [cited 2021 May 31]. Available from: https://www.anh-academy.org/immana/grants/grants-round-2/natural-resources-institute-nri-university-greenwich-0
- APHLIS+ [Internet]. [cited 2021 May 31]. Available from: https://www.aphlis.net/en
This use case was prepared by Hyomin Lee, MS Candidate at the Friedman School of Nutrition Science and Policy at Tufts University. June 21, 2021.