Springboard Atlantic


Lost your password?

StFX Prof Says Mining Google Trends Could Help Predict Recessions

Economists and policy makers may be able to predict recessions sooner by analyzing Google Trends, says StFX economics professor Dr. Greg Tkacz in his latest paper “Predicting Recessions in Real-Time: Mining Google Trends and Electronic Payments Data for Clues,” published Sept. 4, 2013 by the CD Howe Institute in Toronto.

It’s the first paper from a Canadian to look at Google data for economic prediction purposes, and the first anywhere to look at the specific issue of recessions. And it’s created much media buzz.

Just hours after the report’s release, the story was picked up by the Canadian Press and appeared in Maclean’s, the Toronto Star, and the Huffington Post, Victoria Times-Colonist and Winnipeg Free Press. Dr. Tkacz also spoke with radio stations in Calgary and Vancouver, did a call-in show on CFAX radio based in Victoria, BC, spoke with a reporter from the Wall Street Journal’s “Canada Real Time,” which ran its own original story about the paper, and published an op-ed on this topic in the Globe and Mail.

In the report, Dr. Tkacz found a relationship between Google searches and economic activity when he looked at the 2008 recession. Knowing that many people are keying in search terms such as “recession” and “jobs” can give an indication of economic changes as they’re happening, he says, and the data can also help policy makers respond more quickly to these changes.

“New sources of electronically recorded data are both timely and reflect the real-time intentions of millions (or billions) of agents,” says Dr. Tkacz, who has expertise in short-run economic forecasting, and has been working for several years on innovative data sets to study the movements of the economy through debit and credit card transactions.

“Most policymakers and economists failed to predict the last recession because of the lag in traditional economic data. I look at the search data in Google Trends, as well as electronic payments data, to see if the recession signals were there in real time. ”

He says economic indicator data can often lag several months behind, making it hard to forecast future trends.

“It’s like trying to predict the weather when you only have temperatures for two days ago,” he says.

“This report looks at how we can get more timely information. Getting a more timely measurement of the economy is the ultimate objective here.”

Going forward, there’s still much work to be done to understand how to best use and optimize the data, he says.