Online Grocery’s Double-Edged Sword: Drastic Price Swings
(Bloomberg) -- People shopping online for groceries at Amazon.com Inc. or Walmart Inc. are more likely to encounter big differences in prices based on their location -- even down to the ZIP code -- than shoppers who venture into physical stores, researchers found.
Software that adjusts an item’s price based on factors like availability, pricing at competitors and even individual shopping behavior, is almost as old as e-commerce. But as the pandemic supercharges online grocery shopping, a study from researchers at the Massachusetts Institute of Technology has a warning: Buyer beware.
In one example from early 2019, a 12-pack of Coca-Cola sold by the Amazon Fresh grocery service cost $6.99 for residents of a ZIP code in New York City’s Upper West Side and Harlem and $4.99 for the slice of Cambridge, Massachusetts, that’s home to MIT. The identical item could be had at Walmart for $3.33 in downtown Atlanta, but $5.48 for shoppers in rural Washington state.
“The same product, to be delivered the same day, will have a different price,” said Diego Aparicio, one of the study’s authors. “As a consumer, this is a bit surprising to me. Maybe they’re changing the price in Seattle, but not Chicago, and yes in Miami.”
Shoppers aren’t going to travel far to save a buck or two on soda, but those gaps are a reminder that retailers have much more flexibility to experiment with pricing online than in stores, where applying a new price tag to a shelf costs time and money, Aparicio said. Brick-and-mortar retailers tend to change prices less frequently and apply those changes uniformly across a region. Not so online, he said.
Price gaps between retailers also tended to be larger online than for physical stores. “In Manhattan, in this given ZIP code, we know that the price difference for Oreos between Amazon and Walmart is much greater than the difference you would find in two different stores in Manhattan,” he said.
That’s not for lack of trying. The study, which used software to track 88 items across 30 ZIP codes and then compared that to physical stores, found that Amazon appeared to closely track Walmart’s prices, responding in kind to nearly three-quarters of the largest retailer’s price changes, often within hours.
“They’re experimenting,” said Aparicio, now a professor at University of Navarra’s business school in Spain. Aparicio, along with co-authors Zachary Metzman and Roberto Rigobon, found a region’s affluence didn’t appear to explain the price variations. Instead, stores appeared to set prices based on real-time demand in a very specific area.
The study, which was presented at a National Bureau of Economic Research conference earlier this year and is awaiting publication, didn’t examine whether online services were offering different prices for different users shopping from the same place.
©2021 Bloomberg L.P.