In early September, news outlets reported that the price of onions in India had suddenly spiked nearly 300 percent over prices a year before. Analysts warned that the jump in price for this food staple could signal an impending economic crisis, and the Research Bank of India quickly raised interest rates.
A startup company called Premise might’ve helped make the response to India’s onion crisis timelier. As part of a novel approach to tracking the global economy from the bottom up, the company has a daily feed of onion prices from stores around India. More than 700 people in cities around the globe use a mobile app to log the prices of key products in local stores each day.
Premise’s cofounder David Soloff says it’s a valuable way to take the pulse of economies around the world, especially since stores frequently update their prices in response to economic pressures such as wholesale costs and consumer confidence. “All this information is hiding in plain sight on store shelves,” he says, “but there’s no way of capturing and aggregating it in any meaningful way.”
That information could provide a quick way to track and even predict inflation measures such as the U.S. Consumer Price Index. Inflation figures influence the financial industry and are used to set governments’ monetary and fiscal policy, but they are typically updated only once a month. Soloff says Premise’s analyses have shown that for some economies, the data the company collects can reliably predict monthly inflation figures four to six weeks in advance. “You don’t look at the weather forecast once a month,” he says.
Premise uses a mixture of in-store logging and online prices to measure U.S. inflation. The company has continued to publish new figures each day even as the federal shutdown triggered by Congress has halted the U.S. government’s indexing program.
The people who check store shelves for Premise are located mostly in cities in Latin America and Asia, including the largest emerging economies: China, India, and Brazil. Soloff says that custom price indexes for food or specific crucial commodities will be of interest to many financial and consumer product companies. “In these developing, huge economies this isn’t an alternative—it’s primary,” he says. “We’re building a map for the first time.”
Each day, Premise’s workers receive a kind of to-do list on their smartphones via a mobile app for Android devices. They’re asked to go to particular stores and log the prices of certain products, and to snap a picture of that product on the shelf. Workers receive between 5 and 15 cents for each price logged and may be asked to record as many as 250 in a day, a workload that would take a couple of hours. The tasks, done with the permission of store managers, are designed to be something workers can fit in around another job or university studies.
Premise can also collect customized data for clients by sending specific questions to its crowd of workers. Consumer goods companies might be interested in paying for information on how its own and competing products are being priced in certain markets, Soloff says.
Soloff says Premise is already working with some consumer goods companies and has begun to feed data to financial companies including Bloomberg, which has since August offered access to certain Premise data feeds on inflation indexes. The company is backed by venture capital firms Harrison Metal, Andreessen Horowitz, and Google Ventures. Google’s chief economist, Hal Varian, is an advisor.
Premise’s data could be a useful new way to monitor prices and unearth new information about how they are changing within economies, says Alberto Cavallo, an assistant professor of economics at MIT and a cofounder of PriceStats, a company that calculates inflation and price indexes by monitoring online prices for goods in over 70 countries.
However, Cavallo says that Premise will have to constantly check that its workers are recording the price of the exact same items each time they log prices in a store. Research projects that looked into using mobile apps to track prices in stores have sometimes produced unreliable data because people sometimes record the price of a different version or packet size of a product by mistake, says Cavallo.
“That’s potentially the greatest challenge of crowdsourcing,” says Cavallo. “If they can control for that and make sure it is always the same items, it could be a very useful approach.”
Soloff says he is confident his company can do that, because it offers multiple layers of quality control. The company’s workers are carefully trained to be reliable, he says, and Premise can check the location fix and other metadata gathered by its app each time a price is recorded.
The company is also investigating whether image processing software could be used to automatically check the images uploaded with each price check. That could ensure that the correct item was logged and might even allow more sophisticated measures to be added—recording, for example, what condition fresh food is in, or whether shelves are running low on stock.
Premise’s data may have other uses outside the financial industry. As part of a United Nations program called Global Pulse, Cavallo and PriceStats, which was founded after financial professionals began relying on data from an ongoing academic price-indexing effort called the Billion Prices Project, devised bread price indexes for several Latin American countries. Such indexes typically predict street prices and help governments and NGOs spot emerging food crises. Premise’s data could be used in the same way. The information could also be used to monitor areas of the world, such as Africa, where tracking online prices is unreliable, he says.
The dark secret behind those cute AI-generated animal images
Google Brain has revealed its own image-making AI, called Imagen. But don't expect to see anything that isn't wholesome.
Inside Charm Industrial’s big bet on corn stalks for carbon removal
The startup used plant matter and bio-oil to sequester thousands of tons of carbon. The question now is how reliable, scalable, and economical this approach will prove.
The hype around DeepMind’s new AI model misses what’s actually cool about it
Some worry that the chatter about these tools is doing the whole field a disservice.
How Charm Industrial hopes to use crops to cut steel emissions
The startup believes its bio-oil, once converted into syngas, could help clean up the dirtiest industrial sector.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.