by Policy Options. Originally published on Policy Options
September 1, 2007
In March 2007 the Canadian Radio-television and Telecommunications Commission (CRTC) invited the public to comment on questions about concentration of ownership, cross-media ownership and minority-group ownership. It also announced plans for a public hearing on these questions beginning in September.
In announcing this ”œdiversity of voices” proceeding, the CRTC did not publish any information about the cur- rent state of broadcasting ownership, how ownership in this sector has changed over time or what impact owner- ship has on broadcast content. This is contrary to the CRTC's own past practices. In 1978, it commissioned seven reports from a special ownership study group, which described the status of ownership in broadcasting and addressed the effects of concentration on advertising rates, selling prices and total programming. Twenty years later it also published ordinal rankings of the revenues of the top ten television broadcasters, but it did not link ownership with advertising or program content.
Although the CRTC published some ownership data between 1997 and 2005, the information is limited and dis- courages comparisons between the broadcasting sectors. For radio, it publishes the total hours Canadians are tuned into the nine largest operators, along with the total number of stations and the revenues obtained by the ten largest opera- tors, but it publishes no information about the Canadian musical content aired by these ownership groups. For TV, the CRTC publishes little information about total tuning to the largest television operators, the number of television stations they control, the amount of Canadian program- ming they air or their Canadian content expenditures. While in 2006 it published the total revenues of the four largest English-language television operators and of the two largest French-language operators separately for the year 2005, it is unclear whether these oper- ators are the six largest television oper- ators in the country. It does not list the individual radio and television sta- tions of any ownership groups in these ownership reports. As for pay and spe- cialty services, it publishes no tuning information about these services and does not publish their total revenues by ownership group but by individual service, requiring some calculating dexterity from those who are truly curious about such matters.
The CRTC's failure to publish com- plete information about its owner- ship policies and their effects leaves the general public at a clear disadvan- tage relative to the CRTC itself as the decision-maker whose views the public might want to influence, and relative to Canada's privately owned broadcast media, whose long-established lobby- ist, the Canadian Association of Broadcasters (CAB), likely has ample empirical information through its members.
An obvious irony is that while the CRTC's ownership proceeding will deal with a sector that broadcasts information about current events and the weather to millions of Canadians every day, relatively little empirical information about trends in broadcast- ing ownership and its effects are easily available for the public to review and analyze. The Web site of the Department of Canadian Heritage offers only limited information about current levels of broadcast ownership. While there are several detailed sources of historical information about media ownership, such as the Massey and the Fowler Royal Commissions and the work by Senator Davey's com- mittee on the mass media in 1970, Canadians simply have no easy way to determine whether ownership of broadcasting in this country has become more or less concentrated.
The lack of information is a source of concern due to longstand- ing fears that more concentrated media ownership may affect the con- tent by limiting the diversity of news, analysis and debate available to audi- ences. As the Supreme Court of Canada held in the seminal 1951 case of R. v. Boucher, Canadians are accus- tomed to the ”œclash of critical discus- sion on political, social and religious subjects.” The Court concluded that ”œ[f]reedom in thought and speech and disagreement in ideas and beliefs, on every conceivable subject, are of the essence of our life.” The effects of broadcast media on behaviour are not merely theoretical, as Tannis MacBeth Williams and her colleagues estab- lished in the mid-1980s in their ground-breaking quantitative research into what happened to indi- viduals and the community at large when television was first introduced in a small Canadian community.
In this article I offer aggregated quan- titative information about changes in the control of Canada's conventional, over-the-air radio and television services over time. There are occasional gaps in information about ownership and con- trol, which mean the research is a work in progress. The analysis reported here is based on an ownership dataset describ- ing the control of individual, privately owned radio and television stations from 1922 to 2007. The data consist of some 25,500 lines, each line describing control of an individual, over-the-air radio or television broadcasting under- taken in the September-August broad- cast year, by a company or individual.
The licensing decisions, public announcements and public notices of the CRTC and its predecessors consti- tute the primary source of information for ownership data. Using these documents to identify ownership and control in broadcasting pres- ents a variety of problems, not the least of which is the fact that although the CRTC was established in 1968, its licensing and other related documents have been available on-line only from 1984 on.
Perhaps the most important chal- lenge involves the deceptively simple task of identifying stations that broadcast in Canada. The CRTC does not currently publish lists of these stations, and even if it did, these lists would not necessarily help identify them. Stations' callsigns have changed: after radio station CJCJ- AM went on air in 1929, for example, its callsign changed five times – in 1964 (CKXL-AM), 1987 (CISS-AM), 1991 (CFXX-AM), 1992 (CFXL-AM) and 1996 (CKIS-FM). Even determining newly licensed stations' callsigns typically requires a review of subsequent licence renewal decisions to determine the new stations' callsigns, since this information is not typically disclosed in the original licensing decisions.
Fortunately, the Canadian Commu- nications Foundation (CCF) offers an excellent on-line source of information about this aspect of broadcast owner- ship. Established in 1967 by the CAB, it is operated primarily by volunteers who write histories of individual stations, often including details about individual owners and callsign changes. In the data collected for this research, a station's most current callsign is used to describe the same undertaking over time, so as to facilitate comparisons over the years. If the same station inadvertently appears in the dataset more than once under dif- ferent callsigns, however, concentration of ownership may be overstated. This is likely particularly true of radio stations before the mid-1990s, since their call- signs tend to change more often than those of television stations.
A second challenge in creating a media ownership dataset is that regula- tory announcements typically discuss individual licensees without identifying the person or company with whom control of the licence ultimately rests. Although licensees' original applica- tions to the CRTC might disclose own- ership, virtually all of the CRTC's early records from this period appear to have been destroyed as lacking historical value, and they are therefore no longer available even in the National Archives. While the CRTC has maintained paper-based charts describing licensee ownership and control for several decades and updates these as ownership or control changes, it is unclear whether charts exist for each undertaking or for each owner: for instance, the CRTC's ”œdetailed index” of multiple ownership charts does not include Bayshore Broadcasting, which holds several radio station licences. Other problems with the ownership charts are that there is no simple search mechanism to link specif- ic stations with ownership charts for spe- cific licensees, and many charts do not identify changes in stations' callsigns.
Moreover, since the CRTC does not post its historical ownership charts on-line, those wishing to review those charts have to obtain permission to review the historical ownership charts on-site, as I did when undertaking this research in the late 1990s. Even then, one challenge of the his- torical charts is that for at least a decade they did not distinguish between the callsigns of the originating stations and transmitters that simply rebroadcast the programming of the originating stations. Overall, however, if doubt existed about the owner- ship or control of a station, ownership was not identified. The result is that concentration of ownership is likely underestimated until the mid-1990s, due to missing ownership information.
The resulting dataset describes the ownership over time of more than 600 conventional, privately owned radio and television stations in Canada. The results show that over time, the number of groups that com- pete against each other has decreased. In the 12-month period ending in August 2006, 112 individuals or com- panies controlled Canada's 539 private radio stations and 107 private televi- sion stations, a substantial decrease from the 215 ownership groups that existed in 1976 (see table 1).
As the number of owners has decreased, the average number of sta- tions they control has increased, to roughly 5 stations per ownership group by the turn of the new century, com- pared to just over 1 in the 1960s. In radio, 108 ownership groups now con- trol more than 500 stations. In televi- sion, 21 ownership groups now control just over 100 stations.
Simply knowing the number of stations and ownership groups is not the best way to measure concentration of ownership, however. Competition analysis often uses two statistics to quantify the degree to which a sector's ownership is concentrated: the n-firm concentration ratio (CRn) and the Herfindahl-Hirschman Index (HHI).
The CRn is a straightforward meas- ure of the proportion of a specific vari- able (such as revenues, hours tuned or number of undertakings) controlled by a specified number (n) of the largest firms in a sector. The CRn ranges from just above 0 to 1, with 1 meaning that the n largest firms enjoy full control of the sector. Analysis of broadcasting ownership data using the CR4 statistic shows that ownership has always been more concentrated in Canada's private TV stations than it is in radio; current- ly it is roughly twice as concentrated (see table 2). The four largest owner- ship groups now control more than half (63 percent) of Canada's over-the- air television stations, while in radio the four largest groups control just over a third (37 percent) of the stations.
An interesting aspect of broadcast- ing ownership is that knowing how many stations a group controls pro- vides little information about the group's economic power. As noted earli- er, the CRTC has published total rev- enues of some broadcasting groups since 1997. Calculating the CR4 for radio and TV revenues shows that the proportion of revenues obtained by the four largest groups in each sector has grown substantially over the last decade (see table 3) compared with the CR4s for radio and TV station ownership.
The disconnect between numbers of stations owned and ownership groups' revenue likely reflects the truth of that old real estate maxim, ”œlocation, location, location.” The largest ownership groups control sta- tions in Canada's largest cities where the majority of our population lives, and advertisers are willing to pay more for access to these audiences. Six own- ership groups control almost one hun- dred radio and TV stations in Calgary, Edmonton, Halifax, Ottawa, Toronto, Montreal, Winnipeg and Vancouver (see table 4).
Although knowing the CR4 for radio and TV stations and revenues is interesting, the statistic still tells us little about the underlying market structure of this sector. The HHI index can help fix this gap ”” it indicates the degree to which oligopolistic power exists. In highly competitive markets, the HHI ranges close to zero, and in pure monopolies reaches a maximum value of 10,000. Moderately concentrated sectors are said to have HHIs of between 1,000 and 1,800 HHI points, while highly- concentrated sectors have HHIs exceeding 1,800. A 100-point shift in the HHI in a moderately concentrated sector, or of 50 points in a highly con- centrated sector, is said to warrant concern from competition authori- ties. Unfortunately we lack sufficient information to calculate the HHI in Canadian broadcasting because the CRTC has published only ownership group revenue for just a few of the TV and radio ownership groups.
An interesting issue raised by these ownership data involves long-term control in Canadian broad- casting. Parliament has not granted the broadcasting regulatory authority the right to grant licences in perpetu- ity, only for limited terms. Since 1991, for instance, the CRTC may grant licences that last a maximum of seven years. Nothing requires the CRTC to automatically renew licences, although as part of the administrative ”œstreamlining” it has practised for more than 25 years, the CRTC's current renewal policy is to renew for a full seven-year term the licences of licensees that have fully complied with the regulations and their conditions of licence, and to grant shorter terms to licensees with minor noncompliance issues.
The data show that many owner- ship groups now enjoy licences in near perpetuity. Between 1922 and 2007, 208 ownership groups con- trolled a radio station for more than two licence terms (i.e., 14 years). More than 100 groups (125) held such con- trol for more than three licence terms (i.e., 21 years), while 36 groups held such control for more than four decades. We can certainly question whether it was really Parliament's intention that a select few retain licences for so long, effectively limit- ing the entry of new broadcasters. At a minimum, it raises legitimate con- cerns about the effect on news coverage of such prolonged control. Say, for example, a long-established broadcast- er in a small or mid-sized community consistently supported candidates from a particular party, or opposed candidates of a particular linguistic, ethnic or sexual orientation ””would such support or opposition affect elec- toral outcomes? For sure, further research is needed to shed light on this issue.
Another decisive issue is whether more concen- trated ownership results in higher-quality program- ming. Private television broadcasters have long argued that larger owner- ship groups benefit our broadcasting system by facilitating the production and exhi- bition of more, higher-quality Canadian television programs. So far, the only evidence offered to support such claims is anecdotal.
Broad-based empirical evidence is missing from debates about this question because the CRTC does not publish statistics about the broad- casting of Canadian programming by Canada's radio and television sta- tions or ownership groups. The CRTC now keeps confidential the limited data about individual radio and TV stations' expenditures on Canadian programming that it once made available on licensees' public files in the 1980s and early 1990s. At least, however, the Commission has published aggregated national and regional Canadian television pro- gramming expenditure data since 1993; limited program expenditure data are also available for 1977- 78 and for 1985 through the CAB and the Task Force on Broadcasting Policy, respectively.
Comparing the CR4 in TV stations, ownership with Canadian and foreign program spending suggests that the allegedly positive relation- ship between expenditures and own- ership size is at best tenuous and, at worst negative (see figure 1).
In the case of private TV broad- casters, Canadian programming expenditures increased in real terms from the mid-1970s to 1985 (from $80 to $389 million) while the CR4 in ownership decreased (from 0.31 to 0.28). From 1993 to 2000, Canadian programming expenditures declined (from $426 to $431 million), while the CR4 increased (from 0.48 to 0.65). And from 2001 to 2005, Canadian programming expenditures increased, while the CR4 decreased slightly from 0.65 to 0.63. Meanwhile, private television sta- tions' spending on foreign programs simply increased from the early 1990s, regardless of changes in the CR4. These results lend little support to arguments that larger ownership groups will naturally yield increased spending on Canadian programming.
Since the late 1960s Parliament has required broadcasters to make ”œpredominant” use of Canadian resources to create and present their programming. Although cultural sov- ereignty is often held responsible for this requirement, the truth is that Canada's cultural sector plays an important economic role, generating almost 4 percent of Canada's GDP in the early 2000s, according to Statistics Canada. Figures from the Canadian Film and Television Production Association also indicate that in 2006 Canada's radio, televi- sion and pay and specialty services employed 22,841 people directly and thousands more indirectly. Broadcasters' programming invest- ments amount to hundreds of mil- lions of dollars annually.
Given the economic importance of this sector, the effect of changes in media ownership structures should be assessed empirically, or if such assessments have been made (say, by the Competition Bureau, which reviews large media trans- actions for their effects on advertisers), they should be published.
At a bare minimum, however, the CRTC should begin by disclosing own- ership information in all of its licens- ing decisions, even in the administrative ”œletters of authority” it now uses to grant transfers of owner- ship for licensees' estate-planning pur- poses, without an open public process.
The commission should also pub- lish any research it has conducted or commissioned about the impact of media ownership on Canadian pro- gramming expenditures and broad- casting. Consider that the CRTC has commissioned several content analy- ses of broadcasters' programming in the past to measure gender portrayal and levels of broadcast violence: pre- sumably it retained the recorded pro- gramming on which this research was based, recordings that would constitute invaluable and irreplaceable doc- umentation of our broadcasting histo- ry. Given the importance of diversity in news and information to informed democratic participation and the seri- ous concerns about the effects of high- ly concentrated media ownership on news content, could the CRTC not use such recordings of our broadcast his- tory to determine whether broadcast- ers carry more or less news and information as ownership becomes more concentrated?
As a famous, or perhaps infamous, character on a popular drama program used to remark, ”œThe truth is out there.”
What is somewhat frustrating in debates about important policy mat- ters such as broadcasting is that the public is being asked to participate in such debates, but is simultaneously denied the empirical information required to assess supporting or opposing arguments. The CRTC ”” the agency that has such information ”” has the data, but does not disclose it.
In my view, Canadians deserve more if they are to participate mean- ingfully in debates about media own- ership policy in Canada: simply put, the public deserves more data, more analysis and ”” as Parliament itself has stipulated in our broadcasting leg- islation ”” a clearer understanding of how more concentrated broadcasting ownership results in the allocation of more resources to Canada's cultural industry.