A Supply Chain Approach to Finding Win-win Sustainable Solutions for Edible But Unharvested Produce

The work of the project team with resulting outputs and outcomes, by Project Objective.:

Objective #1: Provide produce growers with easy-­to-­utilize protocols to determine the quantity of LPP and CI product left in their fields.

Background: Produce lost at the farm-­level can be broadly classified as three types: Cosmetically Imperfect (CI) produce (outside the tolerance for USDA #1s or other buyer specifications due to surface scarring, shape and size, but that is otherwise edible, at the correct stage of maturity, and safe for consumption); product that meets USDA #1 and other highest-­quality buyer specifications, but is left in the field because harvesting costs exceed the market price at a given point in time (here referred to as Less than Price Point product (LPP)); and inedible product only suitable for animal feed, energy production, or field-­tilling. This project focused on the two categories appropriate for human consumption. Quantities of LPP and CI produce left in the field can be significant. Yet at the initiation of this project there was no available reliable means to measure this volume. Even the sizable and widely reported figure of 40% food waste in the U.S. does not account for on-­farm losses. Our belief was that by quantifying these losses our work could provide important missing information to policy makers, help to facilitate new market development for producers, and give growers field-­specific data to assess whether or not additional marketing or more intensive harvesting would benefit their particular business

Applied research and findings: This objective was accomplished with in-field measurement on cooperating produce farms, with field measurement of 13 commonly grown southeastern produce crops on a total of 60+ fields (plus an additional 40 fields from work funded by a SSARE Graduate Student Research grant). This research was used to create a field measurement protocol for growers. The team published a North Carolina Cooperative Extension publication and two short-how to videos detailing the protocol and how to use it on-farm. The measurement protocols have been incorporated into the Stewardship Index for Specialty Crops (https://www.stewardshipindex.org/) a national metric for furthering sustainable agriculture. The measurement work has been highlighted and championed by the World Wildlife Fund, the World Resources Institute, and the Produce Marketing Association of the US, all of whom aim to increase uptake of the protocol in order to ensure that produce buyers understand how purchasing policies affect food loss and food waste. 

In addition to the creation and testing of an in-field protocol, adoption of the protocol by national and international bodies, and practical use by growers to inform production and harvest practices, the field measurement revealed the tremendous quantities of field loss for commonly-grown southeastern US produce crops. These findings were published and widely shared and utilized in our economic analysis (Obj #2)  and in discussions with farmers and produce buyers along the supply chain (Obj #3).

For more information, see the published results in:

Johnson, Dunning, Bloom, Gunter, Boyette, and Creamer. 2018. “Estimating On-farm Food Loss at
the Field Level: A Methodology and Applied Case Study on a North Carolina Farm.”
Resources, Conservation & Recycling, 137: 243-250.

Johnson, Dunning, Gunter, Bloom, Boyette, and Creamer. 2018.  “Field Measurement in Vegetable
Crops Indicates Need for Reevaluation of On-Farm Food Loss Estimates in North America.”
Agricultural Systems, 167: 136-142.

Johnson, L., Bloom, J., Dunning, R., Gunter, S., Boyette, M., and Creamer, N. 2019.
”Farmer Harvest Decisions and Vegetable Loss in Primary Production.”  Agricultural Systems, 176: 1-11. https://doi.org/10.1016/j.agsy.2019.102672

Johnson, L., and Dunning, R. 2019. “Food Loss on the Farm: Lessons Learned from
Conversations with Produce Growers”. In Minor, Travis, Mishra, Ashok, and Thornsbury,
Suzanne. Economics of Food Loss in Produce: From the Farm to the Retail Dock. Taylor and
Francis.

Objective #2: Conduct economic analysis to understand the impact on farmer profitability of harvesting/selling LPP and CI product

Background: For most southeastern U.S. produce crops, farmers have harvest crews make repeated passes through a field over a period of days or weeks. Our field measurement research (Obj #1.) identified the quantity of produce that remains in the field after the final harvest.  

Applied research and findings: This objective was accomplished by combining our measurement findings with information from repeated interviews and farm visits with cooperating farmers. We conducted interviews to understand the process and criteria that farmers use to make decisions on harvest, particularly decisions about returning a harvest crew to the field. In year 1 of the project we conducted initial interviews and data collection with six large (300-1000 acres) produce farmers in NC and drafted a spreadsheet analysis. The project team aligned this farm-level data collection with interviews and analysis of the potential buyer demand for CI and LPP product (Obj #3). In year two we fact-checked our estimates with growers and buyers and created final estimates of costs and returns for eight commonly grown southeastern produce crops based on different packaging and final-market scenarios. The project team published these findings in peer-reviewed and trade publications and presented them at grower conferences and research meetings. We found that, dependent on product and volume remaining in the field, farmers could earn revenues above costs even at $0.10 paid per pound, the average price that we discovered that food banks pay directly to growers (see Obj #3), which served as one marketing scenario. 

For more information, see the published results in:
Dunning, R., Johnson, L., and Boys, K. 2019. “Putting Dollars to Waste: Estimating the Value
of On-Farm Food Loss.” Choices Quarter 1 (special issue on food waste). Available online:
http://www.choicesmagazine.org/choices-magazine/theme-articles/food-waste-reductionstrategies/putting-dollars-to-waste-estimating-the-value-of-on-farm-food-loss 

A second research initiative related to this objective was to identify the possible substitution effect that demand for cosmetically imperfect product could have on demand (and prices) of grade-A "perfect" product. Farmers have voiced concern that, should it become more widely available, demand for ugly produce will come at the expense of demand (and affect prices for) their “perfect” product (U.S. No. 1 grade or higher).  This portion of the study offers insight into potential demand and market implications of the availability of “cosmetically imperfect” produce.  An online survey concerning options about and willingness to pay for “ugly” relative to No. 1 grade produce was developed and administered to consumers who were also the primary food purchasers for their households.  A two-step process was used to identify these participants.  First, a screener survey was developed which asked general information about the respondent’s shopping habits and household food consumption.  From this, individuals who are the primary household food shoppers and whose households recently purchased fresh fruit and/or vegetables, were identified.  These participants were then invited to participate in the survey directly focused on cosmetically imperfect produce.  As relatively little is known about this issue, a mix of qualitative and quantitative questions were used in this full survey to allow for the collection of more nuanced information. A total of 415 individuals completed the screener survey, and 100 respondents completed the full survey.

As variety of terms are used to refer to cosmetically imperfect produce, as a starting point consumers were asked what term(s) they use to refer to these products.  This matter is important for use in both future consumer research and outreach efforts. Responses indicate that there is no generally used term used by consumers to refer to these products.  Responses included “cosmetically imperfect” (23.4% of respondents), “imperfect” (23.4%), “ugly” (13.8%), and seconds (9.6%).  Interestingly, however, the largest portion of respondents indicated that they don’t have a name for these products or had not previously thought about this issue (29.8%). 

Respondents were also queried as to their willingness to pay for fruits and vegetables with a variety of common cosmetic imperfections.  These results who also are purchasers of locally sourced food are presented in Table 1 below.  The price changes reflect the percent change in price the purchaser would require to buy a product with the imperfection relative to a No.1 graded fruit or vegetable of the same type.  Negative price changes indicate that a discount would be required.  Overall, these results indicate that some consumers are not bothered by cosmetic imperfections as they would be willing to pay the same price for products with these imperfections as a ‘perfect’ product.  Indeed, some consumers, would actually be willing to pay a larger price for products which are larger in size or abnormal shaped.  In addition, these results indicate that fruits and vegetables which are smaller in size require the largest discount.  This result makes intuitive sense for products which are priced on a per piece basis (e.g. $/apple).  For those products which are priced on basis of per-unit of weight (e.g. $/lb) this result may be somewhat unexpected relative to imperfections due to scarring, coloring, and shape. 

Table 1.  Averages and Range of Price Changes Purchasers would Require to be Willing to Purchase Cosmetically Imperfect Produce

There was also variation in purchaser willingness to pay for products imperfections by product type.  Overall, imperfections in leafy greens would require the largest, root vegetables the smallest discount.  The relative perishability of these products, and impact of these imperfections on the quality of these products, likely is driving these results.

The question of whether consumers would be likely to substitute CI for No.1 products is important and was explicitly asked of respondents.  Results indicate that a minority of shoppers (29.3%) are currently buying some CI products. A notable share of respondents indicated that they don’t currently buy these products but that they would be willing to (35.9%), and a similar share indicated that they either would not (25.0%), or are unsure if they would be interested (9.8%) in buying these products. Importantly though, for respondents who do currently or would consider purchasing CI products, the particular products they are willing to buy varied considerably.  For some, they would be willing to purchase a large assortment of CI fruits and vegetables, while others would opt for only a few select items. These results suggest that while growers of some crops may be at risk of losing a relatively small portion of their customers for No.1s to CI products, this is only likely to occur for root vegetables and if the items are heavily discounted.  Overall then, these results indicate that most farmers should not view the increasing popularity of CI products as a competitive threat, but rather as an opportunity to potentially generate additional income from otherwise abandoned, discarded or very low value products.  In the future, additional research on this topic will be needed to assess the extent to which these findings are generalizable beyond our respondents and whether they persist in instances of major income shocks such as that caused by the 2020-2021 COVID-19 pandemic.

The above survey information was presented at the July 2020 annual meeting of the Applied Agricultural Economics Association meeting.

Objective #3: Capitalize on existing relationships with produce buyers including processors to identify win-­win scenarios to bring edible but unharvested produce to market.

The project team strengthened existing and created new working relationships with supply chain intermediaries (e.g., produce buyers, processors, food service, grocery stores) to address this objective. We leveraged these relationships to investigate the potential for various market channels to absorb CI and LPP product at a price and under conditions that made economic sense for growers. Our applied research approach and findings are briefly summarized by supply chain channel, below.

1. Farms - Food Bank. During our preliminary research of possible market channels for CI and LPP, we heard anecdotal descriptions of farmers selling produce, sometimes on a forward-contracting basis, to large food banks. These food banks sought freshly harvested—rather than donated produce from supermarkets or wholesalers—to fulfil their growing missions to provide healthy, fresh foods to those in need. Food banks can have wider tolerances for size, shape, and surface scaring, making it more likely that a broader swath of the harvest will be utilized. To investigate this as a potential market we conducted a survey with food banks in the 13 states in the Southern SARE region. The results of this survey confirmed that purchase of produce directly from growers is an increasingly common practice among food banks. Additionally, food banks typically repackage product, often using volunteers, into single or family-sized packages. As such, farmers do not need to pack product in cartons. Rather, the product can be transported to the food bank in soft or hard-sided totes straight from the field. This substantially reduces packaging cost (both the labor and cost of the carton). In year 2 we utilized these findings as one scenario for our analysis of estimated costs/returns of CI and LPP project (Obj #2). 

For more information, see the published results in:
Dunning, R., Bloom, J.D., Brinkmeyer, E. 2020. “Making a Market for On-farm Food Loss: ~
Food Banks as a Market for Southeastern Produce.”Journal of Agriculture, Food Systems, and
Community Development. https://doi.org/10.5304/jafscd.2020.092.014.

2. Farms - Food Hub – Corporate Food Service. For this supply chain we analyzed the feasibility of moving CI organic produce from small-scale organic growers (average size 10 acres) via a food hub (annual sales >$2.5 M) to food service. The project’s role was to facilitate discussions between sales personnel at the food hub and the food service chefs at a large Compass Group-managed corporate cafeteria. The cafeteria chefs had occasionally ordered small quantities of organic produce from the food hub for catered events, but noted that cost could be prohibitive for larger-volume cafeteria dining. A pilot project was executed to test whether or not a lower cost of CI product would lead chefs to increase their demand of this product, creating additional income for growers. The project team facilitated the connections and discussions and gathered data for two summer seasons. The food hub provided information on availability and pricing of “seconds” (the CI product) each week on its product availability list circulated to the chefs. The price for CI product averaged 60% of the price of top-grade product.
The Food Hub-Food Service pilot yielded the following conclusions:
a. Having a CI/LPP (colloquially known as “seconds” by chefs in the study) component in the weekly list did not lead to larger total weekly purchases from the food service accounts.
b. Purchasing of “seconds” did occur, but this occurred at the expense of the top quality category. In other words, if “seconds” were purchased, “firsts” of the same product were not--there was a substitution effect.
c. Related to this substitution effect, we did see instances where chefs purchased organic "seconds" in lieu of conventional firsts (because the organic second product was approximately the same price as the commodity "firsts" product).

Generalized findings and recommendations from the pilot:
a. The assumption that we tested was than an opportunity existed for smaller-scale growers to sell unutilized CI produce harvested at the same time as their top-quality produce. The additional channel for CI produce--the connection between the growers unsold CI produce and a buyer with expressed interest in utilizing CI produce--was manifested through the action of the food hub, which provided the communication between the growers and the buyer by adding CI product to its weekly produce availability list. We found, however, that the availability of CI product did not increase buyer demand (because of the substitution effect). More significantly, the major source of food loss on these farms derived from LPP, rather than CI, product.  The grower's problem is not the need to sell a steady supply of CI harvested along-side top-quality product. Rather, the grower's problem is sporadic bursts of product in substantial volumes that go to waste without a ready buyer.
Food hubs can contribute to farmer profitability and sustainability by communicating with their buyers about the existence of on-farm food loss and set a communications channel to quickly identify sporadic bursts of product at farms and communicate this to buyers. In some cases, food service can adjust menus to use this product. However, we caution that this is more difficult to execute than advocates of this approach generally assume because of the constraints of managed food service operations, which can be inflexible.
b. There is an opportunity for organic CI products to be a substitute for more expensive top-quality product.  Chefs preferred (and purchased) organic CI product over similarly priced top-quality conventional product.

3. Farms - Local Produce Distributor – Restaurant Commissary Kitchen & Restaurants. For this pilot the manager of a commissary kitchen servicing a restaurant group in the Raleigh metro area worked with the project team to analyze the potential to purchase North Carolina tomatoes in season, preparing and preserving these tomatoes for use throughout the year, in lieu of purchasing non-NC tomatoes. The economic analysis indicated that this would generate savings (approximately 20%) in the cost of the final product. While this included additional refrigeration costs, the needed square footage to incorporate additional refrigeration was a limiting factor for this operation. This pilot illustrated that with support (e.g., engagement from an extension agricultural economist) commissary kitchens for restaurant groups are a potential partner to increase the use of LPP product (excess product during the season that becomes food loss because of insufficient demand). We found that kitchen managers are eager to analyze the potential for sourcing local product and will incorporate the findings into their operations if the results make economic sense.

4. Farms - National Produce Distributor – Food Service Clients (restaurants and university cafeterias). The projects role was to identify local farmers with the supply and desire to provide CI products on a regular basis for the distributor to provide to its client. The distributor was already working to try and move more CI product through its supply chain, marketing these products as “uglies” that could be cheaper for buyers while also reducing food waste. The distributor was exclusively sourcing small amounts of “uglies” on 1-3 California produce items. The pilot idea was to link local (within NC) small-mid-scale local growers (< 150 acres in produce) into the supply chain with local “uglies” that could be bought at a reduced price. Despite sincere efforts on the part of the distributor to “pull” these seconds through the supply chain, there was little demand from their primarily restaurant clients. The distributor noted that although the difference in the cost between first and second quality on a per-box basis could be substantial, restaurants would rather rely on the certainty of the quality they had come to associate with top-quality product. Buyers associated “seconds" with product that was substandard in the sense of being more likely to spoil more quickly. Just the possibility of this was enough to close the door on purchasing CI produce as it would, in the minds of the food service buyers, create more chaos in what is already a high-pressure kitchen environment. 

5. Farms - Small Processor – Chain Grocery Store (frozen produce section)
The project’s role in this relationship was to conduct initial economic (costs/returns) analysis for strawberries to understand if the processing (capping--removing the top, freezing, bagging) of local LPP strawberries would make economic sense for growers and the processor. Strawberries were selected because they are a relatively high-priced item grown by many smaller-scale farmers and generate large amounts of in-field LPP loss because of the short window in which berries ripen and are viable for harvest. Our findings were that local LPP berries from small-scale operations could not meet the needs of the processor at a price economically viable for both because of: (1) costs of transport and maintaining the cold chain on these perishable fruits (2) high cost of the freezing technology used by the processor (3) labor costs to cap (remove the top of) the berries before freezing.

Ultimately, the processing company became a 3^rd-party marketing and distribution company for regionally sourced produce processed in various co-packing facilities across the eastern seaboard. Small producers are worked into the processing schedule, but the company relies largely on large producers due to the volumes needed to produce at a low enough per unit cost to be profitable.

6. Farms – Chain Grocery Stores (fresh produce section)
Members of the project team worked with a locally-owned regional grocery store chain to facilitate the movement of CI product into these local grocery stores. Members of the project team had already worked over several years with the chain to facilitate value-chain-building between local farms, individual grocery stores, and the chain's regional distribution center. Our grocery chain partner was unwilling to create a separate “seconds” area in the grocery store, even for a pilot-phase purpose, reasoning that this would reduce sales of higher-priced top-quality product, while also negatively reflecting on the produce section.

7. Farms – Salvage Grocery Stores
Salvage grocery stores sell food at heavily discounted prices, which suggests that the could serve as a market outlet for CI produce. To address this question, we undertook exploratory research to begin to characterize these stores in terms of the produce they offer and how they source it. Team members visited nine salvage grocery stores in the Western NC Appalachian region, used an observation protocol to identify type and quality of produce sold, interviewed six produce managers at these stores, and interviewed a produce buyer for the distribution center for a regional chain of salvage grocery stores. We found that stores sourced produce from a variety of sources, including product rejected by local supermarkets or  overages at regional wholesale farmers markets. It is not uncommon for salvage grocery stores to communicate directly with brokers or drivers who transport produce destined for other outlets. Produce managers were willing to accept CI produce (misshapen, off-color, etc.), as long as it was in overall good condition. Our findings suggest that salvage grocery stores have the potential to serve as a market outlet for farmers’ and distributors’ CI produce. An interesting area of future research would be to map the logistics of flows between "regular" groceries and these secondary "salvage" markets.

Objective #4: Develop, field test, and economically evaluate a mechanical harvest-­aid to efficiently clear and sort product from fields after major harvesting is completed
The Glean Machine harvest-aid was constructed by research specialists and students in the Biological & Agricultural Engineering Department at North Carolina State University. A report including build specifications is listed in the project outputs. The Glean Machine is now housed at the NC Harnett County Extension Office and can be loaned to gleaning organizations. The Glean Machine was demonstrated to farmers and extension agents at two field days. These events and on-farm tests conducted with cooperating farmers provided feedback to make improvements to the system design. Following tweaks to the design, the Glean Machine had its first “real world” test at a fall 2019 gleaning event hosted by the Society of St. Andrew, the largest gleaning organization in the US. Their Harvest of Hope (https://endhunger.org/hoh/) event educated teens on hunger and food waste issues, then brought them to an eastern North Carolina farm to glean kale. Approximately 50 middle and high school students participated in the two-day event. A second event utilizing the harvest aid was the 2019 Harnett County 4-H Sweet Potato Gleaning Event. Approximately 80 kids and adults participated in the event, gleaning over 18,000 pounds of sweet potatoes. All participants that wanted to use the Glean Machine were able to ride and harvest. At both events there were participants that would not have been able to harvest produce if not for being able to ride on the harvesting aid. Based on field observations of volunteers using the Glean Machine compared to those who collected produce by hand, the Glean Machine did not appear to increase the rate of harvest. Rather, it provides accessibility to the field for more volunteers. The most dedicated volunteers at typical gleaning events are recently retired, which means some may have limited mobility or stamina. The Glean Machine enables these gleaners to continue to participate. Secondly, the continuous movement of the harvest-aid encouraged continuous harvest, in contrast to a typical hand harvest which results in short bursts of harvest, then rest. Additionally, there is an intangible “WOW” factor, especially in gleaning events involving youth.