LITTLE DATA ON BIG MEDIA. CORCORAN,

According to the newly released Heritage Committee report on Canadian media, we are a nation smothered under the power of excess media concentration. “The data available on this subject indicates that media concentration is increasing in Canada” to the point where “Canada has the highest vertical and horizontal media concentration in the world.”

Wow. That’s terrible. There’s nothing worse in liberal economic circles than corporate concentration, unless the concentration is in the hands of government. Then it’s okay, even desirable. In this case, the committee is zeroing in on alleged concentration of private industry control over the telecommunications, broadcasting, social media, web browsers, wireless, desktop operating systems, newspapers, pay TV, etc.

The committee says it has “the data” to prove that Canada is under the heels of the most concentrated media structure in the world. If it has the data, then it must be fact, right? Unless, of course, the data is a little shaky — as is often the case in the area of economics and law that covers anti-trust, monopolies, competition enforcement, market power, industrial structures, control and concentration and other bogeys of the alt-left.

So what is this data and how did the (Liberal majority) politicians on the Heritage Committee come up with its alarming conclusion? It gets complicated. Some math is involved, along with a lot of massaging and manipulation of media ownership and market share numbers.

The supplier of the data, it turns out, was Dwayne Winseck, director of the Canadian Media Concentration Research Project at Carleton University’s School of Journalism and Communications. The institute’s name sort of gives away its lack of objectivity and Winseck has been cranking out claims of media concentration for many years.

For the committee, Winseck employed a basic data headspinner called the Herfindahl-Hirschman Index. It was developed by U.S. researchers more than 60 years ago and adopted by U.S. Justice Department anti-trust enforcers.

HERITAGE MEDIA COMMITTEE CONCENTRATION CLAIM LACKS SOLID SCIENCE.

Here’s the basic math exercise: To determine market concentration, begin with a list of all the firms in the business and their market shares. Let’s say there are four firms supplying widgets: Firm One has 40 per cent of the market, Firm Two has 30 per cent, Firm Three 20 and Firm Four 10.

From this data, the Herfindahl-Hirschman Index of industry concentration (widely known as the HHI) calculates the square of each firm’s market share. Firm One’s data point then becomes 1,600, while the square of each of the others is 900, 400 and 100 respectively. The total, 3,000, is the HHI for the widget industry.

Under HHI theory, if the total of the squares of the market share in an industry exceeds 2,500, then it is already excessively concentrated, even oligopolistic.

Whether this makes any sense or not has hard to discern from the economic literature. There are dozens of YouTube videos on how to calculate the Herfindahl index, but none that explains why it matters.

The foundational point appears to be the old “perfect competition” myth of corporate organization. If there are 10 firms in the market, then the market is perfectly competitive. Ten times 10, 10 times, equals a 1,000 HHI score, perfect competition. If there’s only one firm with 100 per cent of the market, the HHI is 10,000, monopoly.

If this strikes you as the mathematical flapdoodle that it is, never mind, because it gets more complicated. A key to the Herfindahl index is determining the market that’s being measured and squared. In Winseck’s analysis, cited by the Heritage Committee, the definition of markets are narrowed down to the point where just about every nook and cranny of the media world is a self-contained concentrated market. When it comes to today’s media industries, that approach makes no sense.

For example, Winseck runs an HHI on “Social Network Sites” and comes up with a score of 2,762, thereby creating the idea that Twitter, Facebook and a few other sites are standalone examples of “highly concentrated” media. Separate HHI calculations are done on broadcast television, mobile web browsers, pay and specialty TV channels, newspapers, mobile wireless, desktop operating systems, and many more. All but a few (magazines, radio) hit HHI scores of below 2,100 ( moderately concentrated) while most top 2,500 and up to 8,300 (highly concentrated).

But breaking off each little segment and sub-segment of today’s media industries makes no sense. Many of these isolated industries are part of vicious inter-industry and inter-market competition. Consumers are constantly shifting from one medium to another, one technology to another, from YouTube videos to magazine sites to print newspapers to radio to cable channels to Netflix to Twitter to wireless videos — it’s an endless and giant smorgasbord of content and technology.

In a 2014 law review paper, California attorney Toby Roberts commented that Orris Herfindahl, creator of the index, warned that there are many factors that play into the complex business of industrial competition and that the index “suffered the deficiency of considering only two indicia of industry behavior.” Roberts added another point: “The precision and sophistication of the HHI may cloak its limitations and create a false impression of scientific accuracy in the courts.”

And, one might add, among researchers and politicians who are in the business of policy-based evidence making, one of the hallmarks of junk science.