By The Canadian Press
ANTIGONISH — A St. Francis Xavier University professor has published a paper that argues economists and policy makers may be able to predict recessions sooner by analyzing data from Google Trends.
In the paper posted Wednesday on the C.D. Howe Institute website, Greg Tkacz, an economics professor at the Antigonish-based university, says he found a relationship between Google searches and economic activity when he looked at the 2008 recession.
Tkacz says tracking the usage of search terms such as “recession” and “jobs”, could have predicted the recession up to three months in advance of its official onset. He also says it may be possible to track consumer spending intentions by following how often people are searching big-ticket items.
“If you can capture the intention of ... millions of consumers simultaneously, you can start getting some indication of where the economy is going as a whole,” he said.
“For example, if lot of people are looking for big-screen televisions, and they start typing the search into Google, that could be an indication the economy is booming if people have those kinds of spending intentions.”
The professor says timely information could help policy makers respond more quickly to changes in the economy.
“Right now it’s September and we only have official GDP data going back to the end of June,” he said. “If you can track something that comes out every day like Google trends or real time debit or credit transactions, you might have a better idea of how the economy is actually doing.”
He cautions that Google query data has only been available since 2004 and the time span in his study is very short in terms of the overall business cycle. Tkacz’s paper is titled “Predicting Recessions in Real-Time: Mining Google Trends and Electronic Payments for Clues.”