Wasted Efforts

A whopping one-third of all food in the U.S. ends up in landfills, where it generates more greenhouse gas emissions than the entire aviation industry—an economic loss of $1.2 trillion each year.  

So when Assistant Professor Luyi Yang learned about the Too Good to Go app, which allows customers to pre-order $3.99 “surprise bags” of unsold food, he was intrigued by the chance to reduce waste while getting good deals. But was the approach truly effective? 

Working with Man Yu from The Hong Kong University of Science and Technology, he built a mathematical model that compared the impact of surprise-bag clearance, traditional mark-down clearance, and no clearance sales at all. 

The results, published in the journal Management Science, surprised him: Rather than reducing waste, both types of food clearance sales could generate more waste, because retailers overproduce and consumers overbuy—or toss what they don’t need or like.

“Once you take everything into account, these clearance strategies are not necessarily a universally beneficial market-based solution for combating food waste,” Yang says. 

Yang found surprise bags can work when they target a specific customer segment: people who are particularly price-sensitive and unlikely to overbuy. It’s most likely to backfire on food products with narrow profit margins. 

“For high-margin products—think food sold at high-end restaurants or premium bakeries—it seems that surprise clearance is going to be very effective,” Yang says. 

Yang also found that mixing disparate products creates more waste, while more homogeneous products—bagels with different flavors, for example—lead to less waste. “When consumers have strong preferences, they’ll just throw away whatever they find in a surprise bag that’s no good for them,” Yang says. 

While carefully targeted clearance strategies can be effective, profitable, and a benefit to consumers, sometimes the most effective overall pricing strategy for reducing food waste is managing production with straightforward predictions of sales numbers, not clearances.