Canadian Heritage Minister Steven Guilbeault, who says his top legislative priority is to “get money from web giants”, has discussed creating a new mandated licence for social media linking to news articles. Guilbeault calls the practice “immoral” and envisions using the Copyright Board of Canada to create a tariff for linking to news articles accompanied by new government powers to levy penalties for failure to comply. Last week, I looked at a single Toronto Star article to see its engagement on Facebook. Using the CrowdTangle Chrome extension, I found that virtually all public engagement with the article came from a Facebook post that the Toronto Star posted itself.
Today’s post expands on that approach by examining a larger group of articles all taken from single day. This post looks at 36 original articles posted to the Star’s website on the morning of October 14, 2020. The day was selected at random and while most Facebook posts take place within the first 24 hours, I waited five days before examining the social media engagement with the articles to give time for potential posts and shares. The Star’s website changes throughout the day, so recreating the day’s paper is difficult. Instead, I endeavoured to include all new Toronto Star-originated articles that had been posted over the prior 20 hours (thereby including articles posted on both October 13th and the morning of the 14th). Articles from wire news services (which is the majority of foreign news stories, sports stories, and some national stories) were excluded as they are licensed by the Star and can be found from many sources online.
The 36 articles demonstrates the value of the Toronto Star with many original local stories that might not otherwise be reported. Yet claims that Facebook is taking the Star’s content or that the situation amounts to “blackmail” is simply not born out by the data. First, the Star itself is responsible for the majority of public postings on Facebook. Of the 36 articles, 27 articles or 75% were publicly posted at least once by the Star. As I noted in my previous post on the issue, all of these articles are licensed since the Toronto Star grants Facebook a licence to host and distribute the content as part of its terms of service. Eleven of those articles were only publicly posted by the Star with no other public postings. This article on high school exams in Ontario is typical.
Second, the Star often publicly posts articles to multiple Facebook accounts. 13 of 36 articles or 36% were posted to two or more Facebook accounts. Some articles were publicly posted to six different Toronto Star Facebook accounts. For example, this article on the CCLA and the Ford government was publicly posted six times by the Toronto Star and once by a non-Star account. All of the public engagement came from the Star’s own postings.
Third, there were only two articles that were not publicly posted by the Star but were publicly posted by a non-Toronto Star account. One of those articles generated good engagement and one had none.
Fourth, it is worth noting which non-Toronto Star accounts typically publicly post its articles. In many instances, the posters are the subject of the article (or directly related to the story). For example, this article on affordable housing in Parkdale and the potential leasing of land from the University Health Network was one of the more active article with six non-Toronto Star public posters. The poster included the Parkdale Community Updates, University Health Network, the United Way, and the University of Toronto Medicine. The largest number of interactions still came from the Toronto Star’s own post.
Fifth, politicians are often the source of the public postings, which is curious given Guilbeault’s claim that the activity is “immoral.” For example, this article on Conservative leader Erin O’Toole resuming in-person fundraising events was publicly posted six times by the Toronto Star and by Liberal MP Hedy Fry.
In addition to the postings, there are two other data points that merit highlighting. The first involves public engagement from the postings. The so-called “immoral” activities is premised on Facebook benefiting from media content without paying for it. The primary means of profiting comes from ad revenue generated through engagement, which includes likes, comments, and shares. Yet the data shows that the majority of engagement comes from Toronto Star posts itself. The 36 articles generated a total of 5668 public engagements (notably 1/3 of those engagements came from a single article on electricity pricing) with 3712 of those coming from Toronto Star posts. In other words, 65.5% of all the public engagement on Facebook came from the Toronto Star itself. In fact, of the 29 posted articles (7 were not posted at all), 16 of them only had engagement arising from Toronto Star posts.
Second, this data is focused on public postings. CrowdTangle also provides aggregate data that incorporates non-public posts (ie. private sharing). The number of public and private article shares (along with all the data on posts) is posted below. The total number of article shares – both public and private – for all 36 articles was 2,663 for an average of 74 shares per article. Even leaving aside the fact that many of these shares come from the Toronto Star’s own posts, Canada’s largest newspaper by circulation generates just 74 shares per article on Facebook. The loss of ad revenue may be a significant threat to Canadian media organizations, but the claim that this can be attributed to Facebook “taking” media content simply does not accord with the data on how articles make their way onto Facebook nor what happens once those articles are publicly posted.