Cracking the code on food waste: A tricky balancing act

ASU researchers uncover the complex challenges of turning unsellable produce into a sustainable business model.

When it comes to sustainable living, we have a problem: Nearly 45% of the Earth’s habitable land is devoted to producing food, either by growing it or raising livestock. But current agricultural practices are leading to soil erosion, water pollution, and global warming, and there still isn’t enough food to go around.

The environmental cost of food production

"That amount of usage has a huge impact, environmentally," says Elliot Rabinovich, the AVNET Professor of Supply Chain Management. "And so, if you want to improve sustainability, a big part would involve either being more productive in how you use that land to make more food or cutting down on the amount of waste generated after producing the food."

Because Rabinovich’s expertise lies in operations and supply chain management, not agriculture, he and his W. P. Carey colleague Sanghak Lee, an associate professor of marketing, along with Assistant Professor Stanley Lim at Michigan State’s Broad College of Business and Korbit Chaired Professor of Marketing Sungho Park of Seoul National University in Korea, decided to study ways to cut down on food waste, with the help of a grant from the U.S. Department of Agriculture.

Rabinovich, Lee, and their fellow researchers learned that more than 40% of the food produced for consumption in the United States every year goes uneaten. Some of it is the result of failed cooking experiments. Some of it sits uneaten in pantries and refrigerators until somebody finally throws it away. However, nearly 20% of that food, primarily fruits and vegetables, never reaches consumers. It gets discarded before it even leaves the farm because, for whatever reason — color, shape, size — it fails to meet preset standards.

The role of standards in food waste

"These kinds of standards have a purpose," Rabinovich explains. "For instance, retailers want to make their stores look great, appealing, and clean and entice customers to buy. Also, using standards helps with storage. Having fairly uniform produce in sizes and shapes makes it easier to store and stock throughout the supply chain."

Still, that’s a lot of wasted apples and oranges and bananas.

In the past decade, several online companies have devised ways to save some aesthetically unappealing but perfectly edible fruits and vegetables. They obtain the ugly produce from the growers and sell them directly to customers at a deep discount. Rabinovich and his colleagues partnered with one of those companies, Imperfect Foods (IF), for their study.

A new business model for ugly produce

Based in San Francisco, Imperfect Foods was founded in 2015 and originally known as Imperfect Produce. Initially, IF offered customers uniform boxes of produce based on what was available, much like a CSA (community-supported agriculture). However, as time passed, its founders adjusted the model so customers could choose what they wanted in their deliveries.

However, this led to another set of problems rooted in basic inventory management. Unlike traditional retailers, IF can’t choose what to order from its suppliers or adjust its orders based on customer demand; instead, it has to sell what is available. It also has to keep its prices low so the service remains attractive to customers primarily interested in saving money. Finally, it has to be especially careful about its inventory. In other sectors, an overstock might not be a big problem: The company can hang onto the merchandise and sell it sometime later. This approach doesn’t work with produce because of its limited shelf life. But suppose IF doesn’t have enough stock. In that case, this is also a problem: Customers may not receive the products they ordered, which may make them unhappy enough to take their business elsewhere, even if IF offers compensation.

The hidden cost of stockouts

Still, the general thinking was that customers wouldn’t care about a stockout because the product was defective anyway and they were getting a good deal. Overstocking, which left IF with rotting produce, was worse than understocking.

But this, the researchers discovered as they delved deeper into the data, was not actually true. Customers did care. And stockouts did have a significant impact on IF’s business.

Quantifying the cost of waste

The data they used came from IF’s largest market, San Francisco, for five months, from Oct. 1, 2016, to Feb. 28, 2017, and included 18,314 transactions from 2,153 customers. At the time, IF was offering an average of 38 products per week. The researchers examined the cost of stockouts for 50 different products.

The cost of a stockout has three components: direct cost (loss in profit for the out-of-stock product), indirect cost (the loss in profit of products that are in stock but will be less likely to be chosen because consumers tend to buy them together with the out-of-stock product), and compensation cost (what IF would need to spend to make things right with customers if they can’t get the product they ordered). Some products, like Lacinato kale, were low-cost. Others, like curly kale, were priced significantly higher. But for 49 of the 50 products, compensation cost was higher than either direct or indirect cost, roughly $3 per item, when the total stockout cost was between $3 and $4.

The researchers also noticed that stockouts had much greater consequences for some products than others. For some groceries, there were acceptable substitutes. But for others that were usually ordered with other complementary ingredients, like, say, lettuce, the cost was compounded.

"By quantifying those stockout costs and showing that they are significant," Rabinovich says, "we show that by being shy in how much inventory you order, thinking that stocking out is not going to be a problem, you're actually leaving money on the table."

The solution, then, is for IF to have more inventory than less and risk leaving customers unsatisfied. There may be food waste, but that’s offset by attracting more customers, which will help IF in its original mission of selling all the ugly fruits and vegetables that would otherwise be thrown away.

This principle extends beyond fruits and vegetables. Rabinovich says it can also apply to products such as electronics or clothing in secondary markets, which are often sold at a deep discount; the glut of unsold electronics and discarded clothing have caused environmental problems.

What’s next for Imperfect Foods?

After the research was concluded, IF made changes in its business. It started offering more choices to customers, both in terms of the size of their boxes and what they could put in them, and it began selling retail-unfriendly pasta, grains, and nuts (hence the name change from Imperfect Produce to Imperfect Foods). And other companies began selling food that might otherwise go to waste.

Which leads, Rabinovich and Lee say, to yet more questions that they hope to address in the future: How will broader assortments affect stockout costs? How will more competitors in the imperfect food space affect that market and the supply of produce overall? And what effect will this have on the overall problem of food waste and the impact agriculture is having on the environment?

"Whoever tells you this is easy to solve," Rabinovich says, "I don't think they truly thought about it."