Andrey Armiagov
Frances Woolley is a professor of economics at Carleton University, where she teaches public finance. Her recent Economy Lab posts can be found here.
Twenty years ago, undergraduate economics students rarely, if ever, had a chance to do original research. Today, everything from the Canadian census microdata files to hockey player statistics and the World Bank's data catalogue is readily available on-line. Statistical packages are cheap and/or easy to use -- and this generation of students learns how to use software at lightning speeds.
This past term, students in my honours research essay class have written original papers on topics ranging from the determinants of obesity to the impact of inheritance expectations on savings decisions. One of the most interesting papers was one on electronic commerce, written by Zinan Li.
The 2007 Canadian Internet Use Survey asked respondents, "During the past 12 months, have you ordered a good or service over the Internet? (For your personal or household use, not business use)." Mr. Li used a multivariate regression analysis to find out what types of people were most likely shop online. Predictably, people with more education, and higher incomes, were more likely to buy goods over the internet. The demographic most likely to buy goods on-line was 25-34 year olds – after that, the odds of shopping on-line gradually decline with age. Separated or divorced people are more likely to shop on-line than anyone else – though whether being divorced causes you to spend more time on the internet, or spending time on the internet causes you to get divorced, is an open question.
The finding that I found most surprising, however, was that people in Quebec are less likely to shop online than people in any other province – even after controlling for age, education, and income. My first thought was: language. Perhaps people in Quebec are less likely to shop online because American websites like llbean.com are English-only, and that makes it just that little bit more difficult to shop online?
Mr. Li's more detailed analysis, however, casts doubt on a purely linguistic explanation. Mr. Li broke down the internet shopping data further, and looked separately at electronic purchases of travel, books, music, electronics, clothing and flower gifts. This finer breakdown revealed that the Quebec effect is greatest for clothing and travel. Residents of Saskatchewan, for example, are 5 percentage points more likely to buy clothes and travel online than Quebeckers, even after controlling for a range of demographic factors, but only one or two percentage points more likely to buy flowers, electronics or music online. Why would that be?
My own personal theory about Mr. Li's clothing result is simple (though not supported by rigorous statistical analysis): Québécois don't buy clothes online because Montreal has great shopping. There may be a virtuous circle here: Montreal's retail scene entices people to go out and buy the latest fashions, one fashionably dressed person induces others to update their wardrobes, and all of these shoppers allow independent retailers to stay in business. By way of contrast, Newfoundland and Labrador – the province with the highest rate of on-line clothing purchases – risks falling into a vicious cycle: with relatively few local retailers, people buy clothes online, thus putting even greater pressure on local establishments.
What about Mr. Li's travel result? Why are people living in Quebec less likely to shop online for travel? For decades, people have been leaving Quebec for opportunities elsewhere. All provinces lose some people to interprovincial migration – British Columbia has its share of displaced Albertans, Alberta is home to more than a few British Columbians-in-exile. Québec is different because relatively few people move to Québec from elsewhere in the country, presumably because there are few good jobs for people who cannot speak French. The Canadian Internet Use Survey was conducted in October. In Alberta and British Columbia – the provinces where people are most likely to shop for travel online -- people were making plans to fly home for the holidays. People in Québec didn't have to – they were already there.
Economy Lab, winner of the 2011 Eppy Award for best business blog. Follow Economy Lab on twitter
Frances Woolley