Suzanne Thornsbury, chief of the Crops Branch in the Market and Trade Economics Division, U.S. Department of Agriculture Economic Research Service, served as moderator for a panel on the impacts of food loss and waste on food prices and farm incomes.
Timothy Richards, Marvin and June Morrison Chair of Agribusiness and Resource Management at Arizona State University, presented highlights of a study funded through the Agriculture and Food Research Initiative to look at food waste driven by price discrimination by retailers. Retailers price discriminate by charging different prices for items of different quality. Retailers have minimum quality standards to maintain a reputation for high-quality produce, but farms grow produce of varying quality by nature (such as imperfect appearance or variance in size), which results in excess supply. The amount is substantial: about 19.5 million tons of edible, perishable food products annually. More value can be generated on the farm if this surplus has a market. As an example, a California company called Imperfect Produce picks up and resells surplus produce by subscription at a lower cost.
Dr. Richards and his colleagues developed an economic model to explain retail loss as a consequence of optimizing behavior, devise an identification strategy for supply chain loss, estimate retail loss due to quality-base
price discrimination, and estimate the impact on value lost in the supply chain due to willingness to pay for quality.
The model assumes that consumers demand produce with higher quality and retailers maximize profit subject to the grading standard. The researchers derived an equilibrium standard and developed two cases. In the first case, farmers do not produce enough to meet the standard, in which case there is no food waste when grading costs are sufficiently low. In the second case, however, farmers produce more than enough to meet the standard, and graded food sent to the retail channel is priced out of consumers’ reach, which may stimulate potential for loss in the retail channel. With reasonable parameters, they estimate the retail loss in the second case at 37.5 percent.
Using apples as an example, retail data were derived from Nielsen Scantrack to describe prices and sales volumes for bagged fresh apples. Data came from every store of a major U.S. retail supermarket chain for the 52 weeks from October 2014 to October 2015, with 6 varieties of apples and 14 different items by universal product codes. Data from agronomic literature1 and wholesale prices from the Washington Tree Fruit Association were also consulted. The results showed that customers are willing to pay for quality, an important aspect in demand for fresh produce. The gap between the willingness to pay for quality apples and the total amount of quality apples produced on the farms is what could be thrown away—about 10 percent, which mirrors the estimates in the Loss-Adjusted Food Availability series of the Economic Research Service (see presentation by Jean Buzby, Chapter 2), or a loss of $100 million for farmers. Among major types of fruit and vegetables, the economic losses for apples are found to be much higher. Moreover, if consumers have higher quality standards, the losses would be greater. Dr. Richards drew the following conclusions from the study:
1 See Miller, S., R. McNew, R. Belding, L. Berkett, S. Brown, J. Clements, J. Cline, W. Cowgill, R. Crassweller, E. Garcia, D. Greene, G. Greene, C. Hampson, I. Merwin, R. Moran, T. Roper, J. Schupp, and E. Stover. 2004. Performance of apple cultivars in the 1995 NE-183 regional project planting: II. Fruit quality characteristics. Journal of American Pomological Society 58(2):65-77; Miller, S., R. McNew, R. Crassweller, D. Greene, C. Hampson, A. Azarenko, L. Berkett, W. Cowgill, E. Garcia, T. Lindstrom, M. Stasiak, J. Cline, B. Fallahi, E. Fallahi, and G. Greene II. 2007. Performance of apple cultivars in the 1999 NE-183 regional project planting: II. Fruit quality characteristics. Journal of American Pomological Society 61(2):97-114; and R. E. Henriod, D. S. Tustin, K. C. Breen, M. Oliver, G. A. Dayatilake, J. W. Palmer, S. Seymour, R. Diack, and J. Johnston. 2008. Thinning effects on “Scifresh” apple fruit quality at harvest and after storage. IX International Symposium on Integrating Canopy, Rootstock and Environmental Physiology in Orchard Systems 903:783-788.
- Quality-based price discrimination can generate surplus food. Farmers produce a continuous distribution of quality, and retailers have an incentive to truncate the quality distribution of fresh produce at the farm level;
- Testing the hypothesis using store-level scanner data, fresh produce sold through retailers is horizontally and vertically differentiated;
- Using variety, package, and market variation to identify willingness to pay for quality, distribution of quality preference is recovered via kernel density;2
- Retailer behavior is responsible for a 10 percent loss in apples; and
- Loss due to retail intermediation is likely similar for other products.
Throughout the workshop, presenters and participants identified different stages of the value chain where food loss and waste occur. Rob Vos, director of the Markets, Trade and Institutions Division at the International Food Policy Research Institute, focused on the initial stages of the food supply chain in developing countries, particularly in the poor, small farm context. Dr. Vos agreed that reducing food loss and waste potentially will produce significant gains, but acknowledged that those may be complicated. The analysis by the International Food Policy Research Institute looks at losses in primary production (including both pre- and post-harvest losses), distribution, and processing, but not at food waste in the consumption stage. In developing countries, most food losses occur in the initial stages of the supply chain, while food waste at the consumer level is more prevalent in developed countries.
Evidence from Ecuador, Peru, Guatemala, Honduras, Ethiopia, and China shows that food losses range from 6 to 26 percent of production, mostly at the farm level (Figure 5-1). Delgado, Torero, and Schuster (2017) used four methodologies to estimate food loss along the value chain, including a self-reported method (S) based on reporting by producers, middlemen, and processors; a category method (C) based on the classifica-
2 The kernel density was estimated with an Epanechnikov weighting function to maintain maximum flexibility. See Epanechnikov, V. A. 1969. Non-parametric estimation of a multivariate probability density. Theory of Probability and Its Applications 14(1):153-158.
tion of a crop into quality categories; an attribute method (A) based on the evaluation of a crop according to inferior visual, tactile, and olfactory characteristics; and a price method (P) related to the price discrimination market (a decrease in price is a deterioration in quality).3 The determinants vary by crop, context, and stage in the value chain, but common factors relate to inclement weather; poor knowledge of how to avoid pests and plagues; poor technology; poor infrastructure for storage, transportation, and processing; and low and volatile prices. Lack of sufficient labor or high costs of labor also may lead to crops not being harvested, especially when market prices are low. These factors are important to take into account when looking at the impacts or benefits of preventing food losses at the farm level and other stages of the value chain.
Dr. Vos considered the hypothesis that less food loss is associated with higher farm income. The International Food Policy Research Institute’s analysis shows that the outcome depends on whether the “saved” food can be sold in the market without a price decrease. There can be gains if handling, storage, transportation, and processing capacity are improved; however, such improvements are not a given. A key point for policy research, he said, is the need to work across the value chain and not just at the farm level.
Next, he considered the question of whether reducing food loss would induce farmers to use some of their available land to produce other crops. There is little evidence about patterns of crop substitution in response to reduction in farm-level food loss. What is known is that this would depend on market conditions, Dr. Vos stated. If only a small group of farmers adopts a new technology or improved practice that reduces food loss, the impact on market prices may be small, such that farmers may prefer to keep the same acreage for that crop, which would now be more profitable at the same price because of the reduction in food loss. If the technology that reduces food loss is widely adopted, market prices may fall if the demand for that crop is relatively price elastic. In this case, farmers may have incentives to shift to higher value crops, Dr. Vos explained.
Reducing food waste at the consumer level may also lead to lower market prices and farm incomes. Consumers would benefit, of course, because the price of food would fall. Consumer purchasing power would increase and consumers could spend some of the real income gain on higher value–added
3 Delgado, L., M. Torero, and M. Schuster. 2017. The reality of food losses: A new measurement methodology. IFPRI Discussion Paper 01686. Washington, DC: International Food Policy Research Institute.
foods (such as fruits, vegetables, and animal-source foods), which could then have positive spinoffs for farm incomes.
Dr. Vos concluded that those benefits are not guaranteed, but are a likely outcome depending on further development of the overall value chain. Better inputs, practices at the farm level, packaging, cold chain and dry chain, processing capacity, and market development will help food loss reduction translate into welfare gains for farmers and consumers alike.
Pete Pearson, director of food waste at World Wildlife Fund, explained that the organization became involved in food loss and waste because 70 percent of biodiversity loss is due to food production. Eliminating or coming close to eliminating food waste can help feed a growing and more affluent population without more habitat loss. World Wildlife Fund’s strategy goes across five areas: hospitality and tourism, restaurants and other food services, retail grocery, farms, and schools and universities.
Its focus for the past 18 months has been on farms. Mr. Pearson referred workshop participants to the World Wildlife Fund report No Food Left Behind,4 the first in a series of planned reports. The study focused on four crops with differences in production volumes, methods, locations, and end markets: tomatoes, peaches, potatoes, and leafy greens. Quantitative measurements were done in the field; researchers also did a qualitative study in which they talked with growers of different crops.
The study found a 56 percent loss in the field for romaine lettuce. When there was an oversupply, lettuce was not harvested. Tomatoes had a 40 percent loss in the field and a 15 percent loss in packing houses. Peaches had a 39 percent loss in the field and a 14 percent loss in packing houses. Potatoes had a 2.5 percent loss in the field and about a 2 percent loss in packing houses. Reasons for loss included not meeting quality or retail standards, too ripe, labor costs and labor shortages, market dynamics, and grower-buyer relationships.
The point of the work, Mr. Pearson said, is take action. He reflected what other presenters had said: Solutions require looking at the system
4 Pearson, P., M. McBride, and L. Prezkop. 2018. No food left behind, part 1: Underutilized produce ripe for alternative markets. Available at https://c402277.ssl.cf1.rackcdn.com/publications/1170/files/original/WWF_No_Food_Left_Behind_111018.pdf?1542040595.
level. He said that World Wildlife Fund does not want to create incentives that would result in more agricultural expansion but rather to encourage the use of surpluses without expanding the footprint. He also suggested reimagining the notion of “fresh” and applying it to produce frozen directly after harvest to lock in nutrition.
A participant asked about the costs of water and energy production as part of the cost of production in the analyses by World Wildlife Fund. Mr. Pearson stated that his team tried to look at environmental metrics, including greenhouse gas emissions, but could only calculate them for a small number of farms. He noted the variability each year with weather and other environmental changes. Another participant, referring to Mr. Pearson’s comments about avoiding agricultural expansion, asked about Dr. Richards’ point that apple acreage might increase in his model. Dr. Richards responded that this growth was conjecture because producers may not make decisions based on today’s prices and might make this decision over a number of years. Mr. Pearson said that World Wildlife Fund recognizes that every bit of loss and waste cannot be eliminated, but tolerance rates can be found. Dr. Vos stressed the need to look at the broader food system and take account of tradeoffs. Mr. Pearson suggested looking beyond delivery of whole fruits and vegetables, even through means such as Imperfect Produce, to different ways to value and process items. They can be moved more inexpensively and with more nutrients intact through refrigeration and other ways, for example.
A participant brought up the demand for crops for biomass, which many food security studies do not consider. Another aspect, commented a participant, is that new technologies are often profitable for early adopters but less so for the later adopters, and she wondered whether food-loss technologies would follow this pattern. Dr. Richards said that there may be scale bias in some of the innovations toward larger farmers. Dr. Vos agreed that scale is always an issue. However, the dynamic can change with changes in infrastructure such as roads to improve market access or better storage capacity, for instance. Mr. Pearson added that better information flows would also help, both in developed and developing countries.
The final commenter in the session suggested that vertical farming can result in increases in nutrient density, a smaller footprint, fresher produce, and other benefits. Shortening the distance between production and farm-
ing can also be a missing link in conversations about food loss, food waste, and agricultural production. Mr. Pearson agreed; he urged re-thinking the production system of some crops and not being afraid to make bold moves.