Posts Tagged ‘Internet concentration’

Growth and Concentration in the French-language Network Media Economy in Canada, 2000-2012

Cross posted from the Canadian Media Concentration Research Project website. 

This post focuses on the development of and concentration trends in eight sectors of the network media economy in French-language regions of Canada from 2000 until 2012: i.e. wireline telecoms, mobile wireless services, internet access, broadcast tv, pay and specialty tv channels, total tv, radio and online advertising. It is a follow up to previous posts that looked at these matters across Canada as a whole (see here and here for the last two)(for a downloadable PDF version of this post please click here).

As with the previous posts, the data and methodology underpinning the analysis in this post can be found through the following links: Media Industry Data, Sources and Explanatory Notes, French Media Economy, CR and HHI French Media and the CMCR Project’s Methodology Primary. Excellent additional resources for further analysis of the media in French-language regions of Canada can be found through the GRICIS research project at Université du Québec in Montreal and the Centre d’études sur les médias at Laval. Journalist Steve Faguy is also very knowledgeable about the media industries in Quebec.

So what did we find?

The Growth of the French-Language Network Media Economy, 2000-2012. 

The media economy in French-language Canada has expanded greatly since 2000. Revenues rose from $9.8 billion to $14.5 billion in the last dozen years and, indeed, the French-language media grew faster than in the rest of Canada. The relatively fast pace of growth, however, has slowed considerably since the “great financial crisis” of 2008, just as has been the case with the rest of Canada and indeed for much of the Anglo European world,

The faster rate of growth relative to the rest of Canada likely reflects the fact that, historically the French-language media economy has been smaller than what its population alone would dictate. For instance, while Quebec’s population accounts for about 23% of the national total, in 2012 it’s media economy accounted for just over a fifth of the total Canadian media economy (20.6%) – although that was up from just 18% at the turn-of-the-century. 

Figure 1 below shows the trends.

Figure 1: The Growth of the French-Language Network Media Economy, 2000-2012


Growth of French NME 2012 (wo total$)

Sources: French Media Economy, Sources and Explanatory Notes.

The fastest growing sectors of the French-language media economy, again similar to patterns in the rest of Canada, have been in internet advertising (2,442%), internet access (524%), mobile wireless services (237%), cable, satellite and IPTV (118%) and, less so, television (34%). By and large, it is the platform media industries and, again to a lesser extent, television that are driving the growth of the network media ecology, adding both to its size and structural complexity. 

At the opposite end of the spectrum, wireline telecom has fallen by more than a quarter. Newspapers also saw their revenues decline after seeming to peak in 2008, from an estimated $1,036 million then to $907 billion last year – a drop of 12%.

While these trends areconsistent with the course of events in the rest of the country, and indeed throughout much of the Anglo European world, one important thing distinguishes French-language dailies form the rest of the country: paywalls. Unlike the English-language press where twenty-four dailies accounting for two-thirds of circulation have put up paywalls in a bid to stem the tide, only two dailies out of ten in Quebec representing just under half of average daily circulation – Quebecor’s Le Journal de Montréal and Le Journal de Québec – have done so.  

Power Corporation’s La Presse has resisted the temptation. This difference in the extent to which English- and French-language dailies have embraced paywalls likely reflects the fact that Radio Canada/CBC looms larger in Quebec than elsewhere in Canada, and perhaps cultural considerations as well.

Radio has grown only modestly since 2000. In fact, since 2008, the medium has seen revenue stagnate, largely because of a combination of budget cuts and restraint in government funding of the CBC and flat advertising spend, with the latter largely being a function, once again, of the economic uncertainty since the financial crisis. 

One more thing that stands out from Figure 1 is the extent to which the growth of several media flattens or goes into decline after the onset of the “great financial crisis” in 2008. Indeed, several sectors see a dogleg in growth at this time: cable, satellite and IPTV as well as internet access, notably. Even fast growing mobile wireless services slowed, while newspaper revenues dropped.

Figure 2 below gives a snapshot of these conditions based on trends since 2000. 

Figure 2: Growth, Stagnation and Decline in the French-language Media Economy, 2000-2012.


Growth Stagnation Decline French Media 2000-2012 

Leading Telecoms, Media and Internet Companies in Quebec

Every study of the Canadian media industries highlights the colossal role that Quebecor plays in French-language media, and rightly so. The company’s reach across the telecoms, television, newspaper, magazine, book and music retailing landscape is enormous. With over 80% of the sprawling media conglomerate’s $4 billion in revenues — $3.3 billion — coming from Quebec in 2012,[i] the company single-handedly accounts for over one-fifth of all French-language network media economy revenue. 

While Quebecor no doubt cuts an imposing figure within French-language media, it is not the largest media conglomerate in this respect; Bell is — by a large margin. Figure 3 illustrates the point and shows the top 14 companies and their revenues from the eight sectors canvassed in this post. Figure 4 immediately after that shows what conditions would have looked like if the CRTC had approved Bell Astral Version 1.0.

Figure 3: Leading Media, Internet and Telecoms Companies in Quebec, 2012 (millions$).



Figure 4: Leading Media, Internet and Telecoms Companies in Quebec, 2012 (millions$) — Post Bell Astral Version 1.0


Sources: Media Industry Data, French Media Economy, Sources and Explanatory Notes.

Several interesting points stand out from Figure 3. First, taking all their holdings into account in Quebec, Bell’s 2012 revenues of just over $5 billion outstripped Quebecor’s $3.3 billion by a large margin. In fact, BCE accounted for more than a third of all revenue in Quebec, which was roughly equal to the next three biggest players combined: Quebecor, Rogers and Telus.

If the CRTC had approved the 2012 version of Bell’s bid to take-over Astral, as Figure 4 illustrates, the gap would be larger yet. Under the first version of that failed transaction, BCE’s total share of the French network media economy would have been 37.2% versus 22.6% for Quebecor (the consequences of the Competition Bureau and CRTC’s approval of Bell’s revised bid to acquire Astral in early 2013 will be discussed in next year’s post when the effects based on 2013 data will be discernible).

As Figures 3 and 4 show, Bell and Quebecor are in a league of their own. The two vertically-integrated giants tower over their peers, most of whom operate in only one or two sectors. Cogeco is a partial exception because it too is vertically integrated because of its stakes in high-speed internet access, basic phone service, cable tv and radio, but its revenue ($445 million) and market share (3.1%) across the ‘total network media economy’ are puny by the standards of Bell or Quebecor.  

Joining Cogeco are another half-dozen or so second tier players: Rogers, Telus, the CBC, Power Corp, Astral and Google with French-language media revenues in $200-$950 million range. Telus and Rogers’ stakes in Quebec are mostly limited to mobile wireless services, although the size of the mobile wireless segment, and the fact that after internet advertising and internet access, it is the fastest growing sector, means that the two comapnies loom large in the province. Google is ranked ninth based on estimated revenues of $268.4 million from online advertising in 2012 and just under two percent share of the entire network media economy (versus $242.2 million in 2011). 

The CBC still cuts a formidable presence in the province as well. Indeed, it is the largest player in TV and radio, with a 40% and one-third share of both media markets, respectively, compared to Quebecor with one-quarter of the French tv market and Astral’s 27% of the radio market in 2012. The CBC/Radio Canada’s major role is probably one reason, as mentioned earlier, why the French-language press has been more hesitant to introduce paywalls, as noted above; it is also why the CBC is so vilified by Quebecor and others in the English-language press.

Power Corps’ place as the seventh largest media enterprise in French-language markets gives a sense of the continued importance of the press within the overall mediascape and of the scale of its newspaper interests (i.e. La Presse, Le Nouvelliste, La Tribune, La Voix de l’Est, Le Soleil, Le Quotidien, Le Droit). Power Corp’s share of average daily circulation is equal to that of Quebecor’s two French-language dailies, Le Journal de Montréal and Le Journal de Québec: 47% — the basis which I use to estimate newspaper revenues for both companies in French-language markets. Independents pick up the remaining six percent of the circulation and revenues.

The French-language newspaper market, in short, is extremely concentrated, and more so than the national situation. Given the importance of newspapers amongst political and business elites, it is this dominance that no doubt draws a critical eye to both companies, and especially to Quebecor given it’s sprawling grasp across media, while Bell’s relative absence from ‘opinion influencing media’ seem to give it a freer hand in this regard.

Finally, a number of smaller players with less than one percent market share round out the ranks: Eastlink (.7% market share), V Interactions (.5%), Facebook (.2%) and Shaw (.2%).  Together these four companies account for less than two percent market share.

Concentration in the French-language Network Media Economy, 2000-2012

Beyond the individual companies and their ranking, the most notable point with respect to the French-language media is the extent to which just two entities — BCE and Quebecor – dominate the landscape. Together, they account for well over half of all revenues (57%) (BCE’s market share in 2012 was 34.7%; Quebecor’s 22.6%). And this was before the Competition Bureau and CRTC blessed BCE’s take-over of Astral, the 8th largest French-language media company, earlier this year.

The second observation to be made is that concentration trends across the board are considerably higher for French-language media markets than in their national counterparts, except wireless. This is important for several reasons.

For one, it shows that the national measure we rely most on can be insensitive to conditions on the ground at the local/regional level. To put this more bluntly, we under-estimate concentration levels, not exagerrate them. This in turn makes the case that there is a media concentration problem in Canada even stronger.

Table 1, below, depict the trends over time in French-language network media economy on the basis of two standards methods for analyzing concentration: Concentration Ratios (CR4) and the Herfindhahl – Hirschman Index (HHI) (see methodology discussion in the last post and the CMCR project’s methodology primer).

Tables 1: CR and HHI Scores for the French-language Media Economy, 2000-2012

CR & HHI French Network MediaEcon, 2012

Sources: CMCR Project CR and HHI French Media.

Table 1 shows that every single sector of the media, telecom and internet examined here is very highly concentrated in Quebec, except for radio which slipped under the threshold for the designation in 2012.

The Platform Media Industries

One notable trend moving gradually in the opposite direction is the steady decline in the extremely high levels of concentration in mobile wireless services. One thing that stands out in this regard is that Quebecor has emerged as a significant rival to Bell (33% market share), Rogers (29%) and Telus (28%) since entering the market after acquiring spectrum in the last round of spectrum auctions in Canada in 2008.

Quebecor’s share of the market has grown to 5.3% (based on revenues) in the four years since it entered the market, effectively demonstrating the viability of the 4th player strategy. Other newcomers, notably Wind, have picked up about 4.7% market share, as well. As a result, Quebecor, Wind and other newcomers now account for 10% of the market, while the big three’s share has dropped to 90% since 2008. While the mobile wireless market is still highly concentrated by the CR4 (95.3%) and HHI (2742) measures, Quebec stands out as (1) the province with the highest levels of competition and (2) indicating the viability of a “4th mobile wireless carrier” strategy.

In contrast, a less unusual trend can be seen when we turn our attention to internet access. In this case, the levels of concentration are much higher in Quebec than they are across the country. Indeed, concentration levels for internet access have risen steadily and sharply since 2008, reaching an HHI of 2726 in 2012, a score that is firmly in the very concentrated zone versus one that was more in the moderately concentrated region just four years earlier. The CR4 in 2012 was also high at 78%.

In contrast, Canada-wide, the CR4 was 59% and the HHI at the low end of the scale at 1051. It is the former measure that is the more accurate, though, while the gap between them a reflection of measuring things nationally despite the fact that access to the internet is arranged in light of the choices available locally. Both measures are useful, though, and this why we look at things from multiple angles.

In terms of the broadcast distribution markets (BDUs), IPTV services have steadily grown to become more significant rivals to incumbent cable and DTH companies since 2010. The CR4 and HHI scores both fell slightly between 2011 and 2012, but are still at the extremely concentrated levels they were two years earlier: CR4 = 91.5%; HHI = 3400.

Concentration has hardly budged over the past few years. In 2012, Quebecor and BCE accounted for 49.3% and 29.5% market share, respectively, or just shy of four-fifths of the BDU market. The big two have increasingly clashed over the past twelve years, however, as BCE’s share of the market nearly tripled, rising from 10% to 30%, while Quebecor’s slid from 65% in 2000 to just under 50% in 2012.

A key reason why concentration remains sky high in Quebec is that BCE began rolling out IPTV services in 2010, half-a-decade later than in the prairie provinces, and did so in a way designed to protect its investments in DTH satellite TV. Cogeco and Eastlink remain distant rivals behind the big two, with 9% and 3.5% market share respectively.

Figure 5: Cable, DTH & IPTV French-Language Market Share, 2012

French Lang BDU (2012) Sources: French Media Economy, Sources and Explanatory Notes.

The Content Media Industries

Casting the net a bit more broadly to bring television into view alongside the distribution side of this domain also illustrates the extent to which Quebecor and Bell, and their strategies of vertical integration, stand apart from the rest of the field by a very wide margin. As Figure 5 below illustrates, the “big two” account for about two-thirds of the total tv universe, including distribution platforms. The CR4 for this measure is just shy of 82%, while an HHI of 2377 puts it just under the threshold for highly concentrated markets.

Again, it is worthwhile to reiterate that such claims are based on 2012 data before consolidation increased yet further on account of Bell’s take-over of Astral Media.  Figure 6 illustrates the state of affairs with respect to BDU and the total television market in Quebec as of 2012.

Figure 6: Vertically-Integrated BDUs and Total Television by French-Language Market Share, 2012

French Lang BDU+TV (2012)

Sources: French Media Economy, Sources and Explanatory Notes.

French-language broadcast television consists of three main players: CBC/Radio Canada (60% market share), Quebecor/TVA (27.3%) and V Interactions (7.8%). Broadcast television is extremely concentrated with the top 3 players accounting for 95$% of revenues and a sky-high HHI of 4403. Bell has no interests in this segment, and did not acquire any either when taking over Astral.

Radio is not nearly as concentrated, but still sits at the high-end of the moderately concentrated spectrum with an HHI of 2407 and a CR3 of 84%. Once again, the biggest player on the radio landscape is the CBC/Radio Canada, with a third of all revenues. Astral was the top commercial radio broadcaster and 2nd after Radio Canada and ahead of Cogeco in 2012 with 27%.  Big 3 control 84.1% of radio, and the sector fell just under the threshold of highly concentrated in 2012 based on an HHI score 2407. The biggest change in recent years was Shaw (Corus) exit from French-language radio in 2011 after a radio station swap with Cogeco.  

The trend for French language pay and specialty TV services is followed a U-shaped pattern over the decade, with concentration declining between 2004 and 2010, but rising again after 2010. Despite the half-a-decade or so dip, this sector has always remained highly concentrated on both the CR4 and HHI measures, with the “big four” – Astral, BCE, Quebecor and the CBC – accounting for 95.3% of revenues in 2012, and an HHI score of 2670.

Bell was already the second largest player in specialty and pay tv services in 2012 with a 27.1% market share. It would have single-handedly held an extraordinary two-thirds (64.8%) of the market had its original bid for Astral Media — the biggest player in this sector — been given the green light by the CRTC in 2012. The CR4 would also have risen from 95.3% to 99.3%, and the HHI score soared to 4,715. While the outcomes from the 2013 Bell Astral transaction will be assessed in next year’s version of this post, the difference is a matter of slight degree, not in kind: the results are off-the–charts in terms of CR4 and HHI guidelines.

In 2012, the total French-language TV market was highly concentrated by either the CR4 (92%) or the HHI (2594). The biggest entity is still Radio Canada, with 40% market share, trailed by Quebecor (TVA)(24%), Astral (16%) and Bell (11%). Had Bell’s original proposal to acquire Astral been approved as planned without any divestitures, its market share would have risen sharply from 11.1% to 27.7% and the third largest player, Astral, would have disappeared.

Concluding Thoughts

Studying the media industries and their evolution over time is never easy, and it becomes more difficult the deeper one probes simply because so much data is not released to scholars or the public. However, based on what we do know and some reasonable estimates of revenues and market shares derived from that, we can arrive at a pretty detailed and reasonable portrait of the French-language network media economy.

And in this regard, several things stand out.

  • the French-language media economy is very concentrated and much more so than the Canadian media economy as a whole.
  • Bell is the largest player in French-language markets with over one-third of all revenues across a wide swathe of media, followed by Quebecor with just under a quarter of the market across an equally large span of media, telecom and internet markets. The two are in a league of their own.
  • The market shares of BCE and Quebecor are significantly larger within Quebec than they are on the national stage (e.g. for Bell, its total share of the French-language market is 35%, while nationally it is 28%; for Quebecor, the gap is more pronounced, with 23% of all revenues in Quebec versus a modest 6% at the national level).
  • Similar patterns are observable in terms of the structure of the network media economy as a whole, with an HHI of 1800 in the French-language media being 400 points higher than what it is at the national level.
  • Two important exceptions to this general portrait need to be made. First, radio nominally falls into the moderately concentrated zone. Second, the steady uptick in competition in wireless bodes well for those who suggest that a fourth wireless competitor strategy might be just what is needed to help the ailing mobile wireless sector in Canada, at least if international measures are our guide (see here, here, here and here)


[i] For the sectors covered by the CMCR project and not including music and books.


Competition Bureau Blesses Bell – Shaw Take-Over of Astral Media

Bell Astral Round 2 officially got under way today with an announcement by the Competition Bureau that it will conditionally approve the deal. In the Competition Bureau’s words, “Today’s agreement is essential to preserving choice for consumers and ensuring continued and effective competition in this area.”

The Competition Bureau and Bell place a great deal of emphasis on the pay and specialty tv channels and radio stations the latter agreed to sell off to get approval for the deal, as well as the modest restrictions that the Bureau imposed to prevent Bell from blocking rivals’ access to two marquee channels in the Astral line-up: The Movie Network and Super Écran.

The bottom line, however, is that no amount of divestitures can obscure the fact that already extremely high levels of media, telecom and internet concentration in Canada — by historical, international and anti-trust standards — will become a lot higher yet (see here). At least that will be the case, if the CRTC does not steel its spine for a second time to take a much more expansive view of the issues than the Competition Bureau’s myopic views of the deal’s impact on economic efficiency and “relevant advertising markets”. 

More important than the conditions placed on the deal is what Bell did get. Bell already owns thirty pay and specialty tv channels (e.g. CTV News, ESPN, Comedy Network, TSN, Réseau des Sports, Discovery Channel, etc.) and it will add eight more if its deal with the Competition Bureau sticks: the French-language SuperÉcran, CinéPop, Canal Vie, Canal D, VRAK TV, and Z Télé, and English-language services The Movie Network, HBO Canada, and TMN Encore. This, too, must be seen on top of the 28 conventional tv stations that Bell owns that make up its CTV1 and CTV2 networks across Canada.

Thus, even after the divestitures required, Bell will still hold 66 tv channels and its share of the pay and specialty tv market will rise sharply from 27.4% to 38.7%. But as I’ve always said, media and internet concentration is not about the market share of a single player but the structure of the relevant sectors and the telecom, media and internet (TMI) industries as a whole.

Thus, more important than just Bell’s dominant market share is that in the pay and specialty segment of the tv industry, the big 4 companies’ — Bell, Shaw, Rogers and the CBC, in that order if the deal succeeds — share of revenues will rise from 87.6% to 90.5%. This is far in excess of the CR4’s typical threshold for establishing a prima facie case of concentration of 50% and well above the Competition Bureau’s own standards set for banking (para 47)

An already sky-high Herfindahl – Hirschman Index (HHI) score of 2270 will move into uncharted territory at over 3000 (recall, that the U.S. Department of Justice typically uses an HHI of 1800 as a threshold for defining high levels of concentration) (on questions of the CR and HHI methodology, see here). Table 1, below, shows the results. 

Pay and Specialty Television Ownership Groups, Revenue, Market Shares and Concentration Levels, 1984-2011 (1)
2004 2006 2008 2010 2011 Post Comp Bureau Divestitures
Shaw/Corus (4)














CTV Globemedia
















Bell – Shaw

CBC/Radio Canada







Quebecor (5)














Fairchild (Chinavision)







MusicPlus/MusiqueMax (7)







Cogeco (as TQS from 2001-08)


0.1 (Remstar)
Spec and Pay TV $ (14)







Conventional TV $







Total TV $





















Sources: CRTC’s Communication Monitoring Report and its Pay and Specialty Statistical and Financial Summaries; Corporate Annual Reports.

While Bell’s take over of Astral will have minimal effect on conventional over-the-air television, its impact on the total tv market, an amalgamation that adds conventional tv stations to the pay and specialty tv segment, will be significant. Bell’s share of total tv revenues will rise from just under 26% to just under 32%. Sure, these figures fall beneath the CRTC’s threshold of 35% set out in the Diversity of Voices ruling in 2008, but that is more a measure of the weakness of the rules rather than a satisfactory state of affairs. The CR4 for the total tv market will rise sharply from 81% to just under 90%; the HHI will similarly shoot upwards from its current excessive level of roughly 1900 to 2284, as the following table shows.

Total Television Market






Post Comp Bureau Divestitures



Shaw/Corus (7)







CBC/Radio Canada (4)




















 Bell – Shaw

Quebecor (8)











Total TV $





















One of the more perverse outcomes of the state-of-affairs overseen by the Competition Bureau is how it plays to one of Canada’s other major TMI conglomerates: Shaw. Indeed, while there is much talk of divestiture, the arrangements brokered by the Competition Bureau effectively dismantles Astral Media — the ninth largest media company and most significant non vertically-integrated media enterprise in the country — in a way that allows Bell to keep the company’s crown jewels while handing over much of everything else to Shaw.

Indeed, Shaw is a major beneficiary of this transaction, moreso than citizens, consumers and the public will ever be. This is because Corus, which it controls through common ownership by the Shaw family, will pick up the two English-language radio stations as well as the half-a-dozen pay and specialty channels that Bell must sell: the bilingual Teletoon/Télétoon service, English-language Teletoon Retro and Cartoon Network (Canada), and French-language Télétoon Rétro, Historia and Séries+. Bell will also sell off ten other radio stations and another half-dozen specialty and pay channels: The Family Channel, Disney XD,Disney Jr. (English and French), MusiquePlus and Musimax.

This horse-trading amongst dominant players in the industry overseen and blessed by the Competition Bureau smacks of the worst in Canadian regulatory traditions, i.e. the state giving its seal of approval to incumbent interests in already concentrated markets. The matter is made all the more unsavoury by the fact that Shaw was Bell’s only industry ally in Round One of the Bell-Astral deal, supporting Bell’s application to the CRTC and largely sitting silent on the sidelines. The rest of the industry and many others — Quebecor, Cogeco, Telus, MTS, Sasktel, Eastlink, the Independent Broadcasters Association, public interest and consumer groups, etc — fought strategically and on principled grounds against the original deal. The upshot of these arrangements is the creation of two roughly equal behemoths, Bell and Shaw, with each accounting roughly for 38.7% and 35.1% of revenues in the pay and specialty tv sector, respectively, and about 31.6% and 25.4%, respectively, of revenues in the total television market.

Call it a duopoly, but it certainly is not competition in any normal sense of the term. On what should be the more exacting terms of creating the most diverse media possible in line with the ideals of the free press and democracy, such arrangements are a travesty.

Indeed, it is exactly this kind of insider coopetition that has defined Canada’s TMI industries for too long and which the original CRTC decision looked like it might undo. The Competition Bureau’s Consent Agreement certainly blunts that hope, if not kills it outright.

To be sure, this transaction has always been animated by the idea that Bell’s acquisition of Astral might just put it in a better position to undo Quebecor’s dominance of French-language media markets. Is we keep our eyes focused only on the ‘clash of titans’ scenario in which the end game is to pit an even bigger Bell against Quebecor, there is some truth to this, but focusing on only one or two players is not the proper way to assess the structure of any market, let alone media markets.

Looking at Table 3 below, we can see that on the basis of revenues, the CBC is currently the largest player in French language television markets, followed by Quebecor with roughly 24 percent market share and Astral with just over 17%. Bell, V Interactions and Shaw/Corus trail far behind with 8.2, 4.4 and 2.2 percent market share, respectively.

Table 3: French Language Total Television Revenues (Millions), 2007 – 2011






2011 Market Share

PCBD Mrkt Share)

































V Interactions



























Total French-language Conventional TV








French pay and specialty TV








Total French-language TV














Sources: CRTC (2012). Communications Monitoring Report and Aggregate Annual Returns and company Annual Reports.

If the scenario contemplated by the Competition Bureau’s Consent Agreement goes ahead, Bell will replace Astral as Quebecor’s biggest commercial rival.  Shaw/Corus’ place in the French-language market will also be strengthened on account of the increased share in French-language TV services that it will have. While such a scenario might put two of Canada’s largest TMI conglomerates on a more equal footing in Quebec, the elimination of Astral will reduce the number of independent media groups and further drive up already extremely high levels of concentration within Quebec and across the Canada as a whole.

That concentration is already extremely high in Canada there can be no doubt, with the big four firms (CBC, Quebecor, Astral and Bell), as Table 3 above shows, controlling 91% of all revenues. The CR4 will rise if Bell acquires Astral to just under 92%, while the already sky-high HHI will rise from an exceptional 2699 to 2818.  To be sure, these increases might appear modest, but it cannot be emphasized enough that this is only because concentration levels are already off-the-charts by any reasonable measure.

The claims that a bulked up Bell will make for a more formidable competitor to Quebecor is even less convincing when we look beyond the domain of television. In radio for example, while Bell will bulk up on French-language radio stations to complement its English-language stations, Quebecor isn’t involved in radio at all. Net outcome? More concentration in radio, but zero benefit in terms of competition and diversity.

The table below shows the results with respect to French-language radio.

French-language Radio Revenues  






Post CompBur Divestitures $ Mills)




166.2 (4)










Bell – Shaw






84.1 (1)









Total Fench Private Radio Rev



238.4 (2)


258.4 (3)

273.2 (5)

Total Fench Radio Rev







Sources and Notes:  CRTC (2012). Communications Monitoring Report and Aggregate Annual Returns and company Annual Reports; CBC figure for 2007 is based on estimate of 41% of CBC radio revenues allocated to French language services, as per 2008. For 2009, the Aggregate Annual Returns identifies French radio rev for CBC as 170.5, however it is 166.2 in the Canadian Media Monitoring Report; Cogeco data for 2011 from Annual Report differs (p. 29) from CRTC figure of $113.6 (Aggregate Annual Return).

Moreover, while Bell will divest ten English-language radio stations as part of its agreement with the Competition Bureau, more importantly it will retain 77 out of Astral’s 84 radio stations. Add that to the 30 that Bell will retain in its existing stable and it will have 107 radio stations across the country — a development that will, as I stated last year when this transaction was first announced, see Bell “catapult from being the fifth ranked player in radio to top dog”. It’s exact share of revenues can’t be precisely counted, but would be about 26% before the divestitures and likely somewhere around 21-23% afterwards by my estimation.

This is not terribly high, but it does reverse the trend of declining concentration in radio, which is pause enough for concern. Indeed, the best the Competition Bureau can muster in this regard is that it “is satisfied that the proposed divestitures are sufficient to ensure the transaction will not result in a substantial lessening or prevention of competition in any radio market.” That’s a far cry from saying that it will contribute anything positive. 

Finally, Quebecor’s dominance of French-language newspapers and magazines will remain completely unscathed by Bell’s acquisition of Astral, since neither of them is involved in either of these areas, except for Bell’s minority stake in the Globe and Mail. Given the protracted strife and lock-outs at Quebecor’s Journal de Quebec and later the Journal de Montreal in recent years, and Pierre Karl Péladeau’s commitment to using his media outlets to push a clear political and ideological agenda, there is no doubt a great deal of antipathy toward Quebecor in Quebec, across the country and amongst journalists in particular.

This has no doubt fomented a desire to undercut Quebecor’s ability to seemingly lord over the French press with impunity. While that no doubt plays well into Bell’s claims about increasing competition with its erstwhile rival, the fact that it has no stake in the French press further weakens its claim.

Ultimately, the CRTC might yet turn back Bell’s bid to take-over and carve up Astral Media by taking a more expansive view of these matters under the Broadcasting Act and, more importantly, from within the traditions of a free press and democracy. At the same time, however, the fact that the Competition Bureau moved on its own today does not bode well.

Two years ago in the United States, by contrast, the Department of Justice and FCC worked hand-in-glove in relation to the closest parallel to the Bell-Astral agreement: Comcast’s acquisition of NBC – Universal in 2011. To be sure, both regulators gave the green light in that instance, but the terms were a far cry from the weak measures that appear to have been adopted by the Competition Bureau on its own.

We still await details of the Competition Bureau’s Consent Agreement, but so far, its actions seem woefully myopic and unhinged from even its own standards of assessing market concentration. This, however, is probably the price we play when fundamental matters of communication and democracy are left to those who see the world only through a constrained economic lens.

The net outcome of this transaction will be demonstrably higher levels of concentration in both French and national pay and specialty tv markets as well as the total tv market overall. The same will be true with respect to radio.

It will also further the extremely high levels of vertical integration across the entire sweep of the TMI industries. That, in turn, will, at the very least, solidify our dubious honour of having the second highest levels of cross media ownership concentration among the 14 comparable countries surveyed by the International Media Concentration Research Project. In fact, it will likely make us Number 1 on this measure.

At the same time, the idea of carving up the market between Bell and Shaw smacks of too much that is unsavoury of how media policy in this country has worked for far far too long. This has to change. There was hope that such change might be in the air last year when CRTC spiked the first incarnation of the Bell Astral deal.  That hope just got dimmer.


Media and Internet Concentration in Canada, 1984 – 2011

As my last post explained, the media economy in Canada has grown immensely and become far more complex in the past twenty-five years with the rise of the Internet and digital media. In this post, I ask whether the media have become more or less concentrated amidst all these changes?

While opinions are rife on the issue, as McMaster University professor Philip Savage (2008) observes, the debate over media concentration in Canada “largely occurs in a vacuum, lacking evidence to ground arguments or potential policy creation either way” (p. 295).

The need for good evidence on the question has been obvious over the past year in the context of Bell Canada’s bid to buy Astral Media, the ninth largest media company in Canada. Indeed, the CRTC’s decision to kill the deal in late October turned in a big way, although not entirely by any stretch of the imagination, on the evidence about media concentration.

The same question will be front-and-centre in Bell Astral Round Two. While nobody knows what version 2.0 of the deal looks like outside of the two companies’ inner sanctum, and the CRTC staff currently vetting it before it is opened for public interventions (probably in the new year), the issue of concentration will undoubtedly loom large in whatever discussions, and regulatory actions, do occur.

That said, however, we must make no mistake about it, studying media and internet concentration is not about Bell or Astral, or any specific transaction. In fact, the issue in the Bell Astral case is not if Bell is too big but whether telecom, media and internet markets in Canada are already too concentrated as a whole? How do we know one way or another? This post helps to address these questions.

Competing Views on Media Ownership and Concentration

Grappling with these issues is not just about remedying the ‘missing evidence’ problem, but thinking clearly about how the issues are framed.

Many critics point to media concentration as steadily going from bad to worse, but with little to no evidence to back up such claims. Perhaps the best known example of this is Ben Bagdikian, who claims that the number of media firms in the U.S. that account for the majority of revenues plunged from 50 in 1984 to just five by the mid-2000s. Similar views also exist in Canada, where critics decry what they see as the inexorable trend towards greater media concentration and its debilitating effects on “democracy’s oxygen”, for instance, or vilify the media moguls behind such trends who have, in these critic’s words, created “Canada’s most dangerous media company”.

A second group of scholars set out to debunk the critics by quantitatively analyzing reams of media content only to find the evidence about how changes in media ownership and market structure effect content to be mostly “mixed and inconclusive” (Soderlund, et. al al. 2005). The problem with this conclusion, however, is that it proceeds as if media concentration’s ‘impact on content’ is the only concern, or as if preserving the existing status quo might not be a significant problem in its own right (Gitlin, 1978). Undeterred, this line of scholarship trundles on so that, half a decade later, similar studies by many of the same authors, Cross-Media Ownership and Democratic Practice in Canada: Content-Sharing and the Impact of New Media, reach pretty much the same conclusions (Soderlund, Brin, Miljan & Hildebrandt, 2011).

A third school of thought mocks concern with media concentration altogether. According to this school, how could anyone believe that the media are still concentrated when there are thousands of news sources, social networking sites galore, pro-am journalists, user-created content and a cacophony of blogs at our finger tips, 700 television channels licensed by the CRTC, ninety-four newspapers publishing daily and smartphones in every pocket? Ben Compaine (2005), a media economist at MIT, has a one-word retort for those who think that concentration still matters amidst this sea of plenty: internet!

Those in this camp also argue that focusing on concentration when traditional media face the perilous onslaught of global digital media giants such as Google, Amazon, Netflix, Facebook, and so on is akin to rearranging the deck chairs on the Titanic – foolhardy and doomed to fail (Thierer & Eskelen, 2008; Dornan, 2012). Journalistic accounts often share this view, routinely invoking, in mantra-like fashion, the idea that media are more competitive than ever. Like their acdemic counterparts, such accounts offer little to no evidence to support such claims, other than pointing to the same roster of foreign digital media goliaths as if examples equals evidence. It does not.

While some might find it hard to fathom, there’s a fourth school of thought, and one that I largely subscribe to, that accepts that fundamental changes have occurred, but rejects claims that this renders concern with media consolidation obsolete. For all those who guffaw at charges of media concentration, it is easy to point, for example, to the fact that only about a third of the 94 daily newspapers said to exist are actually still publishing original content on a daily basis. Of the 700 television channels listed on the CRTC’s books, just over 200 actually filed a financial return last year. And half of those tv channels belong to just four companies — Bell (33), Shaw (46), Rogers (11) and QMI (12). Their share of the market, as we will see, is much higher yet. Keeping our eye on these facts also highlights, for example, how dominant incumbent players use price (usage-based billing) and bandwidth caps, for example and among other tactics, to protect their legacy television businesses (i.e. CTV, Global, CityTV, TVA), while hobbling rivals (Netflix) and limiting people’s choice as a result.

This school also suggests that core elements of the networked digital media – search engines (Google), Internet access (ISPs), music and book retailing (Apple and Amazon), social media (Facebook) and access devices (Apple, Google, Nokia, Samsung, RIM) – may actually be more prone to concentration because digitization magnifies economies of scale and network effects in some areas, while reducing barriers in others. If this is correct, then we may be witnessing the rise of a two-tiered digital media system, with many small niche players revolving around a few enormous “integrator firms” at the centre (Noam, 2009; Benkler, 2006; Wu, 2010).

The more that central elements of the networked digital media are concentrated, the easier it is to turn these nodal points — Facebook, Google, ISPs, Twitter, and so forth — into proxies that serve other interests in, for example, the preservation of dominant market power in ‘legacy’ media sectors (e.g. television and film), the copyright wars, efforts to block pornography, and in law enforcement and national security matters. In other words, the more concentrated such nodal points are, the more potential digital media giants have to:

  • set the terms for the distribution of income to musicians, newspapers and books (Google, Apple, Amazon);
  • turn market power into moral authority by regulating what content can be distributed via their ‘walled gardens’ (Apple),
  • set the terms of ownership and use of user created content and how it is sold in syndicated markets as well as to advertisers (Google and Facebook) (van Couvering, 2011; Fuchs, 2011);
  • and set defacto corporate policy norms governing the collection, retention and disclosure of personal information to commercial and government third parties.

Whilst we must adjust our analysis to new realities, it is also true that long-standing concerns have not disappeared either. To take just one case in point, consider the fact that during the 2011 election campaign, every single newspaper in Canada, except the Toronto Star, that editorially endorsed a candidatefor Prime Minister touted Harper – roughly three times his standing in opinion polls at the time and the results of the prior election. When 95 percent of editorial endorsements for PM across the nation stump for one man – Harper — something is amiss.

Ultimately, talk about media concentration is really a proxy for bigger conversations about consumer choice, freedom of expression as well as democracy. While such discussions must adapt to new realities, the advent of digital media does not mean that such conversations should fall silent. Politics, values and heated debates are endemic to the topic, and this is how things should be (Baker, 2007Noam, 2009; Peters, 1999).


Discussions of media concentration will never turn on the numbers alone, and nor should they, but it is essential to be as clear as possible about the methods used to assess the issue. To begin, there is no naïve vantage point from which data about these issues can be innocently gathered and presented as if evidence is just out there laying in a state of nature, somewhere, waiting to be plucked like apples from a tree.

Data, in other words, does not serve as a one-to-one map of the reality it claims to describe. Nonetheless, there are good ways to make a good body of evidence and bad. An essential factor all down the line is the need for researchers to be open and reflexive about their methods and theoretical starting points.

A fuller discussion of the methodology that I use can be found here, here and here, but for now we can lay out the bare bones of the approach before turning to the analysis itself. I begin by selecting a dozen or so media sectors at the heart of the analysis: wired & wireless telecoms; cable, satellite & IPTV distributors; Internet access; broadcast tv; pay & subscription tv; radio; newspapers; magazines; search engines; social media sites; and online news services.

Data were collected for each of these sectors over a twenty-seven year period, 1984 – 2011, first at four-year intervals up until 2008 and annually since. For the DIYers among you, here’s a handy dandy list of sources.

Data for the revenues and market share for each ownership group in each of these sectors was then assembled. I then group each of the above sectors into three categories, assess the concentration level in each category, and then scaffold upward from there to examine the network media industries as a whole: (1) network infrastructure; (2) content: (3) online media.

I typically drop wired and wireless telecoms from the whole of what I call the network media industries because the size of these sectors means that they tend to overshadow everything else.

Lastly, I use two common tools — Concentration Ratios (CR) as well as the Herfindhahl – Hirschman Index (HHI) – to depict levels of competition and concentration over time. The CR method adds the shares of each firm in a market and makes judgments based on widely accepted standards, with four firms (CR4) having more than 50 percent market share and 8 firms (CR8) more than 75 percent considered to be indicators of highly levels of concentration.

The HHI method squares and sums the market share of each firm with more than a one percent share in each market to arrive at a total. If there are 100 firms, each with a 1% market share, then markets are highly competitive, while a monopoly prevails when one firm has 100% market share. The following thresholds are commonly used as guides:

HHI < 1000                                     Un-concentrated

HHI > 1000 but < 1,800             Moderately Concentrated

HHI > 1,800                                    Highly Concentrated

The Historical Record and Renewed Interest in Media Concentration in the 21st Century

There has always been, even if episodically, keen interest in media ownership and concentration in Canada and the world since the late-19th and early-20th centuries.

In 1910, for example, the Board of Railway Commissioners (BRC) broke up the three-way alliance between the two biggest telegraph companies — Canadian Pacific Telegraph Co. and the Great Northwestern Telegraph Co. (the latter an arm of the New York-based goliath, Western Union) – and the American-based Associated Press news wire service. Why?

In the face of much corporate bluster, the BRC did this because the two dominant telegraph companies were giving away the AP news service to the top newspaper in cities across Canada for free in order to bolster their stranglehold on the lucrative telegraph business. Allowing this to continue, stated the BRC matter-of-factly, would “put out of business every news-gathering agency that dared to enter the field of competition with them” (1910, p. 275).

Thus, in a conscious bid to use telecoms regulation to foster competition amongst newspapers, and to free up the flow of news on the wires, the BRC effectively dismantled the alliance. For upstarts such as Winnipeg-based Western Associated Press – which had initiated the case – it was a significant victory (Babe, 1990).

Media concentration issues arose episodically thereafter and came to a head again in the 1970s and beginning of the 1980s, when three inquiries were held: (1) the Special Senate Committee on Mass Media, The Uncertain Mirror (2 vols.)(Canada, 1970); (2) the Royal Commission on Corporate Concentration (1978); and (3) the Royal Commission on Newspapers (Canada, 1981).

Things lay dormant for more than two decades thereafter, but sprang to life again in the late-1990s and turn-of-the-21st century after a huge wave of consolidation thrust concerns about media concentration back into the spotlight. Three inquiries were held between 2003 and 2007 as a result: (1) the Standing Committee on Canadian Heritage, Our Cultural Sovereignty (2003); (2) the Standing Senate Committee on Transport and Communications, Final Report on the Canadian News Media(2006);[i] as well as (3) the Canadian Radio-Television and Telecommunications Commission’s Diversity of Voices inquiry in 2008.

Structural Transformation: Two (three?) Waves of Consolidation and the Rise of TMI Conglomerates

As I noted in my last post, for all sectors of the media economy in Canada, revenues grew immensely from $37.5 billion in 1984 to just under $70 billion last year (or from $12.1 billion to just under $34 billion when we exclude wiredline and wireless telecoms) (in inflation-adjusted “real dollars”). Between 1984 and 1996, new players meant more diversity in all sectors, except for newspapers as well as cable and satellite video distribution, where concentration climbed significantly.

Conventional as well as pay and subscription television channels were already expanding during this time. In terms of ownership, incumbents and a few newcomers – e.g. Allarcom and Netstar – cultivated the field, with their share of the market growing steadily in tandem with the number of services available (underlying data for these claims can be found here).

Concentration levels remained very high in wired line telecoms in the 1980s and early 1990s, while wireless was developed by two companies, Bell and Rogers. As had been the case in many countries, telecoms competition moved slowly from the ends of the network into services and then network infrastructure, with real competition emerging in the late-1990s before the trend was reversed and concentration levels again began to climb.

In the 1980s and early-1990s, consolidation took place mostly among players in single sectors. Conrad Black’s take-over of the Southam newspaper chain in 1996 symbolized the times. In broadcast television, amalgamation amongst local ownership groups created the large national companies that came to single-handedly own the leading commercial television networks – CTV, Global, TVA, CHUM, TQS – by the end of the 1990s.

While weighty in their own right, these amalgamations did not have a big impact across the media as a whole. There was still significant diversity within sectors and across the TMI sectors. The CBC remained prominent, but public television was being eclipsed by commercial television as the CBC’s share of all resources in the television ‘system’ slid from 46 percent in 1984 to half that amount by 2000 to just over twenty percent today (see the motion chart on CMCR website illustrating this point).

While gradual change defined the 1980s and early-1990s, things shifted dramatically by the mid-1990s and into the 21st century as two (and maybe three) waves of consolidation swept across the TMI industries. A few highlights help to illustrate the trend:

Wave 1 – 1994 to 2000: Rogers acquisition of Maclean-Hunter (1994). Peaks from 1998 to 2001: (1) BCE acquires CTV and the Globe & Mail ($2.3b); (2) Quebecor takes over Videotron, TVA and the Sun newspaper chain ($ 7.4b) (1997-2000); (3) Canwest buys Global TV ($800m) and Hollinger newspapers papers, including National Post ($3.2b).

Wave 2 – 2006-2007.  Bell Globe Media re-branded CTVglobemedia, as BCE exits media business. CTVglobemedia acquires CHUM assets (Much Music, City TV channels and A-Channel).  CRTC requires CTVglobemedia to sell City TV stations – acquired by Rogers (2007). Astral Media buys Standard Broadcasting. Quebecor acquires Osprey Media (mid-size newspaper chain)(2006). Canwest, with Goldman Sachs, buys Alliance Atlantis (2007) (Showcase, National Geographic, HGTV, BBC Canada, etc) – and biggest film distributor in Canada.

Wave 3 – 2010 – ? Canwest bankrupt. Newspapers acquired by Post Media Group, TV assets by Shaw.  BCE makes a comeback, re-buys CTV (2011) and bids for Astral Media in 2012, but fails to gain CRTC approval.

That the massive influx of capital investment drove consolidation across the telecom, media and Internet industries during these periods is illustrated in Figure 1 below.

Figure 1: Mergers and Acquisitions in Media and Telecoms, 1984 – 2011 (Mill$)

Sources: Thomson Financial, 2009; FPInformart, 2010; Bloomberg Professional; CRTC, Communication Monitoring Report.

Consolidation has yielded a fundamentally new type of media company at the centre of the network media ecology: i.e. the integrated media conglomerate. Extremely popular in the late-1990s in many countries around the world, many media conglomerates have since collapsed or been broken up (AOL Time Warner, AT&T, Vivendi, CBS-Viacom, and parts of NewsCorp, etc)(see, for example, Jin, 2011; Thierer & Eskelen, 2008; Waterman & Choi, 2010). The trend elsewhere has not, however, taken hold in Canada.

Indeed, in Canada, sprawling media conglomerates are still all the rage. Four such giants and a half-dozen other large but more specialized companies part their size make-up the core ‘big 10’ companies in the network media economy: Bell (CTV), Shaw (Global), Rogers (CityTV), QMI (TVA), CBC, Post Media, Cogeco, Telus, Astral, and Eastlink. A detailed chart of each by ownership, revenues, and sectors operated in is available here and will be addressed further in the next post.

Looking at media concentration from the vantage point of the ‘big ten’, the media have become more concentrated than ever. Their share of all revenues (excluding telecoms services) rose sharply in the 1990s and between 2000 and 2008 hovered steadily in the mid- to low-60 percent range. The big four’s share of the network media economy subsequently rose significantly to just under 68 percent in 2010 (after Shaw’s acquisition of Global) and rose again to just under 70 percent in 2011 (when Bell re-acquired CTV) — an all-time high and a substantial rise from 52% in 1992. The levels of media concentration in Canada are more than twice as high as those in the U.S., based on Noam’s analysis in Media Ownership and Concentration in America (2009).

Breaking the picture down into the following three categories and applying the CR and HHI tools provides an even better view of long-term trends:

  • ‘network infrastructure’ (wired and wireless telecom services, ISPs, cable, satellite and other OVDs);
  • ‘content’ (newspapers, tv, magazines, radio);
  • ‘online media’ (search, social, operating systems).

At the end of the post, I combine these again to complete the analysis of the network media industries as whole in a slightly different form.

The Network Infrastructure Industries

All sectors of the network infrastructure industries are highly concentrated and pretty much always have been, although Internet Access is a partial exception.

Table: CR and HHI Scores for the Network Infrastructure Industries, 1984 – 2011

CR & HHI Network Industries, 2011

Much the same can be said with respect to wireless services: they have consistently been highly concentrated, and still are until this day, despite the advent of four newcomers in just the past two years: Mobilicity, Wind Mobile, Public and Quebecor.CR4 and HHI measures for wired telecoms scores fell during the late-1990s as greater competition in wired line telecom services took hold. They reached their lowest level ever between 2000 and 2004 before after shocks from the collapse of the speculative bubble took out many of the new rivals (CRTC, 2002, p. 21). Competition grew more and more feeble for most of the rest of the decade before drifting modestly upwards since 2008. Concentration levels, however, still remain high by late-1990s, turn-of-the-century standards, as well as those of the CR and HHI measures.

Two competitors – Clearnet and Microcell – emerged in the late-1990s and managed to garner 12 percent of the market between them, but were then taken over by Telus and Rogers in 2000 and 2004, respectively. Whether the recent round of newcomers will fare any better it is still too early to tell, but with only 2.2 percent of the market as of 2011 they are a long way from the high tide of competition set a decade ago.

As the telecoms and Internet boom gathered steam in the latter half of the 1990s new players emerged to become significant competitors in Internet access, with four companies taking more than a third of the ISP market by 1996: AOL (12.1%), Istar, (7.2%), Hook-Up (7.2%) and Internet Direct (6.2 percent).

The early ‘competitive ISP era’, however, has yielded to more concentration since. Although the leading four ISPs accounted for a third of all revenues in 1996, by 2000 the big four’s (Bell, Shaw, Rogers & Quebecor) share had grown to 54 percent. Things stayed relatively steady at the level for most of the decade before inching upwards in the past few years to reach 57.1 percent in 2011.

HHI scores for internet access also moved upward between 1996 and 2000, but are still low relative to most other sectors. However, this is probably more an indicator of the limits of the HHI method in this particular case, since 93% of high-speed Internet subscribers rely on one or another of the incumbent cable or telecom companies’ ISPs to access the Internet, according to figures in the CRTC’s Communication Monitoring Report (p. 148).

ISP provision in Canada is effectively a duopoly, with the left over 6-7% of the market not dominated by the incumbents scattered among the 400 or so independent ISPs that still exist. This is a slight increase from last year, but it does not mark the return to competitive internet access.  Canada has relied on a framework of limited competition between incumbent telecom and cable companies for wiredline, wireless, internet access and video distribution markets and in all of these markets they dominate, with some other smaller rivals in each.

Cable, satellite and IPTV distribution is one of the only segments assessed where concentration has risen steadily from low levels in the 1980s (850) to the top of the scales in 1996 (2300), before drifting downwards by the turn-of-the-century to the low 2000s where it has remained ever since. It has dipped below that, to the 1900-range, for the last five years, but this is still at the very high end of the scale.

As I noted in the last post, the IPTV services of the incumbent telcos – Bell, MTS, Telus and SaskTel – are becoming a more significant factor in the distribution of television, after a slow and staggered start. By 2011, IPTV services accounted for 7.6 percent of the TV distribution market, based on my numbers, or 3.8 percent using CRTC data (see page 96).

While I have yet to get to the bottom of why this discrepancy exists, what can be said is that, on the basis of my figures, the growth of IPTV services has made small incursions into the incumbent cable and satellite service providers’ turf (i.e. Shaw, Rogers, Quebecor, Cogeco and Eastlink). However, this has done little more than nudge the CR and HHI scores, as the table above shows.

Over the last twenty-seven years, cable tv has become ubiquitous and new tv distribution infrastructures have been added to the fold – DTH in the 1990s, and now, slowly, IPTV. New players have emerged, but never have so few owned so much. New technologies have generally added to this and have not fundamentally disrupted the broad trajectory of development when it comes to tv distribution channels: more channels, and even some new players, but with more of the whole in the hands of the old. The wired society in Canada is probably the poorer for this.

The Content Industries

Until the mid-1990s, all aspects of the tv industry (i.e. conventional broadcast tv as well as pay and specialty channels) were moderately concentrated by HHI standards and significantly so by CR measures. Competition and diversity made some modest inroads from 1998 to 2004, but the trend abruptly reversed course and levels have climbed steadily and substantially since, and sharply in the last two years. Figure 2, below, shows the trend in terms of CR scores; Figure 3, in terms of the HHI.

Figure 2 CR Scores for the Content Industries, 1984 – 2011

 Figure 3 HHI Scores for the Content Industries, 1984 – 2011

The largest four commercial television providers control about 81% of all television revenues in 2011, up from 75% a year earlier. Levels of tv concentration were pushed to new extremes by

Shaw’s take-over of Canwest’s television assets in 2010 and Bell’s buy-back of CTV last year. The big four’s share of all tv revenue before these transactions in 2008 was 70%. A ten percent leap in concentration in two years is a lot.

If the CRTC had approved Bell’s acquisition of Astral Media – the fifth largest television company in Canada, ahead of Quebecor – the all-time high levels of concentration set in 2011 would have been surpassed by an even higher 89.5%. In contrast, the big four accounted for 61% of the tv biz in 2004, a time before major players such as Alliance Atlantis and CHUM were bought out by the now defunct Canwest and Bell/CTV 1 (circa 2000-2006), respectively.

The CR and HHI measures for tv were at all time lows in the 1990s. This was a time when newcomers emerged (Netstar, Allarcom), yet before the time when the multiple ownership groups that had stood behind CTV and Global for decades combined into single groups. The period was also significantly more diverse because the CBC no longer stood as a central pillar in tv and radio, while pay and specialty television channels were finally making their mark. Today, the latter are the crown-jewel in the tv crown.

Today the largest four tv providers after Bell and Shaw are: the CBC, Rogers, Astral, and QMI, respectively, and in that order. By 2011, these six entities accounted for ninety-five percent of the entire television industry. Similar patterns are replicated in each of the sub-components of the ‘total television’ measure (conventional television, pay and specialty channels), as the chart above illustrates.

In contrast, in 2004, the six largest players accounted for a little over three-quarters of all revenues. The run of HHI scores reinforces the view that the television industry is highly concentrated and has become markedly more so in just the past two years.

Like the cable industries, there has never been a moment when diversity and competition has flourished in the newspaper sector. Consolidation rose steadily from 1984, when the top four groups accounted for two-thirds of all revenues, to 1996, when they accounted for nearly three-quarters – a level that has stayed fairly steady since, despite periodic shuffling amongst the main players at the top. Levels declined slightly in 2011 from 2010, from 77% to 75%, likely on account of Postmedia’s decision to sell some of its newspapers (Victoria Times Colonist) and to cut publishing schedules at others.

Of all media sectors, magazines are least concentrated, with concentration levels falling by one-half on the basis of CR scores and two-thirds for the HHI over time. I have not been able to update the data for this sector for 2011, but there is little to suggest a need to change this view.

Radio is also amongst the most diverse media sectors according to HHI scores, but slightly concentrated by the C4 measure. In fact, in 2011, it became moreso, likely because of a shuffling of several radio stations between Shaw/Corus and Cogeco. Bell’s take-over bid for Astral – the largest radio broadcaster in Canada with 17.5% market share – would also have further pushed radio in the direction of concentration had it been approved last month by the CRTC. Had that scenario come to pass, levels of concentration would have still remained well-beneath the CRTC’s self-defined thresholds, but high by the CR measure and just moderately high by the HHI.

Online Media

So far, there’s little reason to believe that trends are any different in the online realm, as measures of the ISP segment showed. But what about other core elements of the increasingly Internet-centric media universe, such as search engines, social media, online news sources, browsers, and smartphone operating systems?

The trends are clear. Concentration in the search engine market continued to grow between 2010 and 2011, with the CR4 score rising from 94% to 97.6%. Google’s share of the market, however, seems to have plateaued, at just over 81 percent of this domain. Microsoft (8.6%), Yahoo! (4.2%), and (3.7%) trail far behind, yielding a CR4 of 97.6% and an off-the-charts HHI of 6,683.

Figure 3: C4 Scores for the Search Engines, 2004 – 2011

Source: Experien Hitwise Canada. “Main Data Centre: Top 20 Sites & Engines.” last Accessed October 11, 2012.

Social media sites display a similar but not quite as pronounced trend, with Facebook accounting for 63.2% of time spent on such sites in 2010, trailed by Google’s YouTube (20.4%), Microsoft (1.2%), Twitter (0.7%), and News Corp.’s MySpace (.6%) (Experien Hitwise Canada, 2010). Again, the CR4 score of 86% and HHI score of 4426 reveal that social networking sites are highly concentrated.

Similar patterns also hold for other layers of the media ecology. The top four web browsers in Canada – Microsoft’s Explorer (52.8%), Google’s Chrome (17.7%), Firefox (17.1%) and Apple’s Safari (3%) – have a market share of over 90 percent (Comscore, 2011).  There is no data available for Canada with respect to smartphone operating systems, but US data shows that the top four players in 2010 accounted for 93 percent of all revenues: Google’s Android OS (29%), Apple’s iOS (27%), RIM (27%) and Microsoft’s Windows 7 (10%) (Nielsen, 2011).

However, not all areas of the internet and digital media environment, of course, display such patterns. The picture with respect to online news services, for instance, is significantly different. Between 2003 and 2008, the amount of time spent on online news sites nearly doubled from 20 to 38 percent, with most of the leading 15 online news sites simply being the extensions of well-established media companies:, Quebecor, CTV, Globe & Mail, Radio Canada, Toronto Star, Post Media, Power Corp. The other major sources included CNN, BBC, Reuters, MSN, Google and Yahoo! (Comscore, 2009; Zamaria & Fletcher, 2008, p. 176).

While that trend meant that attention was consolidating around a few online news sites, and those of traditional journalistic outlets in particular, it nonetheless seems clear that Canadians have diversified their news sources relative to the traditional news environment (newspapers, tv, radio, magazines).  On either the CR or HHI measure, online news fall under the concentration thresholds and are diverse relative to any of the other sectors, except magazines.

However, the fact that concentration levels edged upwards between 2004 and 2007, after the rapid “pooling of attention” that took place between 2003 and 2007 (see immediately above), suggests that a certain plateau might have been reached in terms of the range of sources people are using. Nonetheless, online news sources are not concentrated on the basis of the measures used here. The following table shows the results.

Table: Online News Sources, 2004 – 2011

News website 2004 (N=1482) 2007 (N =1306 ) 2011 (N=1651 )
CBC 10.6 18.3 13.8
Google 5.3 9.2 10.4
MSN / Sympatico 18.2 11 14.7
Yahoo 9.3 7.4 6.5
CNN 9.3 9.4 6.1
CTV 6.2 2.9
Canoe 2.4 7.6 2.9
Cyberpresse 3.5 3.3 3.9
Globe and Mail 4.1 5.9 3.6
BBC 4.9 2.8
Toronto Star 2.6 2.4 1.5
Global 2
Other 32.6 14.4 31.1
CR4 43.4 45.9 45.4
97.9 100 100.2

Source: Table calculated by Fred Fletcher, York University, from the Canadian Internet Project Data sets (Charles Zamaria, Director).  Reports on the 2004 and 2007 surveys are available at

The Network Media Industries as a Whole (excluding wired and wireless telecoms)

Combining all the elements together yields a birds-eye view of long-term trends for the network media as a whole. As Figure 4 below shows, the HHI score across all of the network media industries is not high by the criteria set out earlier, but the long-term upward trend is clear and significant.

Figure 4: HHI Scores for the Network Media Industries, 1984 – 2010


While the HHI for the network media fell in the 1980s and early-1990s, by 1996 trends had reversed and levels were higher than they were a dozen years earlier. Thereafter, the number rose steadily to close to 600 in 2000, where it hovered for several years before falling again in 2008. Since then, however, the HHI score has shot upwards, rising from 510 in 2008 to 623 after Shaw acquired Global and then to 739 once Bell re-acquired CTV after having sold down its majority stake a few years earlier.

The effect of the Bell Astral deal would have been significant in terms of the network media as a whole, raising the HHI score to over 800 – an all time high. This is still low by HHI standards, but we must bear in mind that we are talking about concentration across the entire sweep of the network media industries, not just a random assortment of a few sectors.

The CR4 standard, as shown in Figure 5 below, reveals the trend even more starkly, with the big four media conglomerates – Bell, Shaw, Rogers & QMI – accounting for more than half of all revenues in 2011, a significant rise in a vastly larger media universe from just under forty percent held by the big four twenty-seven years earlier in what was a Lilliputian pond by comparison. While still only moderately concentrated by the CR4 standard, this is for all media combined.

In each and every single sector of the media that the big four operate, they dominate, as the earlier review of CR and HHI scores illustrated. Moreover, the trend in both scores is up, significantly so in the past three years from a CR4 of around 40% to its current level of just over 50%. If this really was a golden digital media age, as some like to contend, that number should be going firmly in the opposite direction.

Figure 5: CR 4 Score for the Network Media Industries, 1984 – 2010

Concluding Thoughts 

Several things stand out from this exercise. First, we are far from a time when studies of media and internet concentration are no longer needed. Indeed, theoretically-informed and empirically-driven research is badly needed because there is a dearth of quality data available and because, one after another, the press of events and specific transactions – Bell Astral in 2012, but Bell’s re-acquisition of CTV the year before and Shaw’s acquisition of Global in 2010 – demands that we have a good body of long-term, comprehensive and systematic evidence ready-to-hand.

This kind of data is still very hard to come by and data collection for 2011 reconfirmed that at every step of the way. The CRTC still needs a dramatic overhaul of how it releases information and of its website, as David Ellis has recently argued so eloquently. The underlying data sets it includes in seminal publications like the Communications Monitoring Report, Aggregate Annual Returns, and Financial Summaries needs to be made available in a downloadable, open format that allows people and researchers to use it as they see best. The regulated companies themselves must also be made to be more forthcoming with data relevant to the issues, and not less, as they so strongly desire.

The trajectory of events in Canada is somewhat similar to patterns in the United States. Concentration levels declined in the 1980s, rose sharply in the late-1990s until peaking around 2000. However, it would appear that whereas in the U.S. a process of deconsolidation set in thereafter, with the obvious exception of Comcast’s blockbuster acquisition of NBC-Universal last year, concentration levels in Canada have climbed, and steeply so, in the past three or so years.

Current media concentration levels in Canada are roughly two-and-a-half times those in the U.S. and high by global standards (Noam, 2009). Moreover, large media conglomerates straddle the terrain in Canada in a manner that is far greater than in any of the other thirty countries studied by the IMCR project, including the U.S., Germany, Japan, Australia, the UK, and so on, where media conglomerates are no longer all the rage as they once were a decade ago.

The assets from the bankrupt Canwest have been shuffled in recent years, and some significant new entities have emerged (e.g. Channel Zero, Post Media, Remstar, Teksavvy, Netflix, The Mark, Tyee,, Huffington Post). The overall consequnence is that we have a set of bigger and structurally more complicated and diverse media industries, but these industries have generally become far more concentrated, not less.

There is a great deal more that can and will be said about what all this means, but in my eyes it means that concentration in no less relevant in the “digital media age” of the 21st century than it was during the industrial media era of centuries’ past.

CRTC Kills Bell Astral Deal: What Happened and Why?

On Thursday this week, the CRTC killed the Bell Astral deal (news release, full decision). The decision was entirely unexpected by anyone, including me, although all along I have argued that Bell’s bid to acquire Astral Media, the 9th largest media company in Canada, gave the CRTC ample ground to do exactly what it did. I also argued that it was the right thing to do, and that the CRTC should stop Bell’s take-over bid for Astral “dead in its tracks”.

Several things stand out from the decision. First, it sets a precedent. To find the closest parallel to this case, we’d have to reach back more than a quarter-of-a-century to 1986 when the regulator quashed a bid by Power Corporation – owner (then and now) of Quebec-based newspaper group, Gesca – from acquiring Tele-Metropole, the cornerstone of what eventually became TVA: the “largest and most important private French-language television station in Quebec and one of the leading Canadian television stations in terms of local production”, as the decision noted at the time.

Second, the decision makes crystal clear that the CRTC, under new chair, J.P. Blais, will take a large view of media consolidation rather than its typically flinty-eyed view of the world. The CRTC will also look carefully at questions of market share and media concentration, and do so not just using audience ratings as its preferred method but also revenues in ways that capture trends within specific media sectors (e.g tv) and across the media as a whole (see paras 29, 51-54).

Of course, numbers are never determinative, according to the CRTC (see para 52), and nor should they be, I would argue. There is no ‘magic number’ upon which things turn, but measuring media concentration within and across the relevant telecom, media and internet sectors, across time as well as in relation to relevant trends elsewhere in the world, is an essential prelude to the conversation that needs to be had. The Commission now seems more ready than it has been in a long, long time to have that conversation. This is a very good thing.

Third, the CRTC rejected Bell’s claim about the threat of OTT services offered by Netflix, Apple, Amazon, etc., on the grounds that they were exaggerated. As the Commission (2011c) stated less than a year ago in its Results  of  the  Fact-Finding  Exercise  on  Over-­the-­Top  Programming Services,

“. . . the evidence does not demonstrate that the presence of OTT providers in Canada and greater consumption of OTT content is having a negative impact on the ability of the system to achieve the policy objectives of the Broadcasting Act or that there are structural impediments to a competitive response by licensed undertakings to the activities of OTT providers” (p. 8).

That evidence has not changed and the CRTC said so in this decision (para 62). In 2008, according to a Media Technology Monitor/CBC study about 3 percent of tv viewing occurred on the Internet (MTM/CBC, 2009, p. 49). According to their most recent study, “only 4% of Anglophones report only using new platforms to watch TV” (MTM/CBC, 2012, p. 4).

Netflix’s annual revenues, based on 1.2 million subscribers, can be an estimated $115 million in 2011, or about .7% of the total television universe (including BDUs). To this we can estimate that Google’s revenues in Canada last year would have been roughly $1.3 billion, or half of online advertising revenue (IAB, 2011). While that may have had an impact on the newspaper and magazine industries, there is no evidence it has done anything of the sort with respect to the broadcasting industry.

The CRTC also cast a jaundiced eye on Bell’s proposal for BellFlix – a new online, on-demand tv service for its subscribers — that would, so Bell argued, allow a combined BellAstral to effectively compete with foreign OTT operators like Netflix. Bell sprung the proposal on the CRTC on the opening day, but the CRTC didn’t buy it because, first, eleventh hour proposals do not follow the rules. The deadline for complete applications was August 9th, not Day 1 of the hearings.

More importantly, an online “TV Anywhere” service is now a requirement of the internet-centric media world, not a bolt on somehow dependent on Bell’s take-over of Astral (para 61). In other words, Bell will have to launch such a service regardless, if it wants to meet current realities and consumer demand.

Fourth, the CRTC rejected Bell’s argument that there was no need to worry about vertical integration because, “This issue was recently exhaustively canvassed by the Commission in its Vertical Integration proceeding” (Bell, Supp. Brief, para 59). In fact, the CRTC observed that consumer groups, non-integrated distributors (Telus, MTS Allstream, SaskTel, Cogeco, Eastlink, etc.) as well as independent broadcasters (VMedia, APTN, Zoomer, etc.) “filed evidence and argument” that cast significant doubt about the capacity of the new vertical integration rules to effectively constrain “BCE’s alleged anti competitive behaviour with respect to program rights negotiations and product launches” (emphasis added, para 32; all submissions can be found here). Put simply, Bell has been acting as a brute ever since it re-acquired CTV just last year, and for this it has now paid the price.

More importantly for the long-run is what the CRTC had to say about consolidation and vertical integration en route to squashing the deal. First, and to avoid over-stating the significance of what is going on, the CRTC noted that it has long been a fan of consolidation and vertical integration, and still is. Second, and with a big however, it also picked up on a point that I have made many times: greater consolidation and vertical integration has not been an unalloyed blessing (far from it); in fact, the process has been thrown into reverse in many other countries around the world.

In the U.S., the results of de-convergence have been remarkable. Aside from the mega-merger of Comcast and NBC-Universal last year, media companies have been beating a hasty retreat from vertical integration and “convergence”. The number of pay and specialty tv channels controlled by cable companies fell dramatically from the 50-55% range in the early 1990s to 15% by 2006 (Thierer & Eskelsen, 2008, pp. 55-56; Waterman & Choi, 2010).

As Viacom-CBS Chairman Sumner Redstone declared in 2005, “the age of the conglomerate is over” (Sutel, 2005). A year later, Time Warner President Jeffrey Bewkes called claims of convergence and synergy “bullsh*t”! Mainstream Media economist Alan Albarran (2010) summed up the lessons as follows: “Looking back, vertical integration was not a very successful strategy for media companies, and it was a very expensive strategy – costing billions of dollars over time. In the 21st century, the early trends have been to shed non-core assets that distract from the base of the company . . .” (Albarran, p. 47). Further examples could be piled up like leaves in autumn.

With this decision, the CRTC put Bell and the rest of the telecom and media industries on notice that claims about vertical integration and consolidation will no longer be taken as an article of faith, although it will still look upon such claims fondly.  This is critical and while it could put a halt to any more ‘blockbuster deals’ for the time being, I am more inclined to think that it’s too early to tell.

Fifth, the CRTC rejected Bell’s bid for Astral on the grounds that it did not pay sufficient attention to radio (paras 57&60).

Lastly, Bell’s benefits package was roundly criticized and rejected for being self-serving. Too many of the benefits would flow to activities that Bell was already doing (e.g. its otherwise laudable Mental Health promotion campaign) or to services that it had already been directed by the regulator to invest in, i.e. expanding broadband access in the North by its subsidiary Northwestel (para 59).

There is a bigger implication in this latter point too, however, a not-too-subtle slap not at Bell, but rather the independent television and production sector, J-Schools and others who line up at the trough for their share of the public benefits package, all the while soft-peddling their criticisms of ownership consolidation as a prerequisite to doing so, as the Canadian Media Producers Association and Canadian Writers Guild, for instance, did in this case and every other one like it in the past decade.

The CRTC’s decision, thus, interrupts the well-known cycle whereby independent television and film production community pull their punches in ownership cases in the hope that they will be in the acquiring company’s good books when it puts together its “public benefits package” as it seeks regulatory approval. This has created a seriously distorted and sordid cycle of dependency in which higher concentration and problems in the long run are sacrificed for short-term gains. It is essentially taking scraps off the table in a strategic way instead of a principled stance on the matter, or one informed by any evidence one way or another about the desirability of such transactions.

It also could take the process out of the gutter insofar that it lifts the chill over independent broadcasters and those in the creative community who will no longer have to cower out of fear that they will be frozen out of the big vertically integrated players’ programming schedules, or denied access to essential distribution facilities, if they speak out against a deal like this one. Those who stood opposed to the Bell Astral deal jeopardized their own access to the schedule of what is already the second largest tv operator in Canada, and which would have been the largest if the deal had been consummated (see para 28).

This is what economists call the ‘monopsony problem’, where there are many sellers and very few buyers. This problem is acute enough already, with the ‘big four’ – Shaw, Bell, Rogers and Quebecor, in that order – already dominating 81 percent of the ‘total tv market’. That number would have grown to just under 90 percent, if Bell had its way.

The last point I want to address for now is the claim being bandied about that the CRTC’s decision to kill the Bell Astral deal reflects a new activist regulator under the stewardship of its new chair, J. P. Blais.  The claim seems to have first emerged in a Globe and Mail article by Steve Ladurantaye at the beginning of the hearings when Blais read aloud a series of public criticisms of the Bell Astral deal.

Since Thursday when the decision came down, the claim that the CRTC has become an activist commission with a consumer bent has gained a great deal of fuel. Michael Geist, writing in the Toronto Star, says that this ain’t your mom and dad’s old CRTC, but one that has put the consumer back in the drivers’ seat. A piece in the Globe and Mail by Steven Chase today makes the same case. Thursday night, and over at the National Post, Terrance Corcoran bemoaned the turn-of-events, seeing the CRTC as playing the populist card and pushing its activist agenda behind the “shadowy concept” of the public interest.

I have several reservations about this view. First, I am uncomfortable that most of the references are to consumers, with none to citizens and just a few to ‘the public’, and then in disparaging terms (Corcoran). These decisions are not just about cable and satellite bills (Globe & Mail); they are about citizens’ and the public’s access to the maximum range of entertainment, news and information sources possible. They are also about “the Public’s” ability to use these media, especially the internet, without having that use hedged about by restrictions and limits imposed by TMI giants bent on protecting their legacy television businesses and transforming the open internet into the pay-per model, where usage based billing and bandwidth caps run roughshod over citizens’ communication rights. This is about communication rights, democracy and pleasure, not just cable and satellite bills.

Lastly on this point, in contrast to seeing the CRTC as suddenly having been remade in a consumer activist mould by J. P. Blais, I think we need to entertain a more critical view.

In this view, as social and political theorists have long shown and discussed (see, for example, C. Wright Mills, The Power Elite), the room for significant changes and unexpected outcomes increases immensely when there is a split amongst elites. And in this case, that split was on full display, with Bell standing on one side arrayed against not just citizens and consumers wary of yet even more telecom-media-internet concentration, but the biggest players in the biz, indeed, almost all of the rest of the industry except Shaw, who sat on the sidelines.

Bell may be a behemoth, but pitted against the rest of the industry and the public, the CRTC had a massive opening through which to think outside the box. And it did, and make no mistake about it, this is a big decision. However, the real test will be whether that continues to be a trend when the industry once again closes ranks, as it so often does, or most of the key players involved do like Shaw did this time around: sit on their hands. Will the CRTC be as emboldened then to pursue “the people’s” interest? For that, we’ll have to wait and see.

The Significant Impact of the Bell Astral Deal on Media & Internet Concentration in Canada

Today was a good day. An unbelievably frantic one, but a good day nonetheless. I’ve been pouring blood, sweat and tears into a submission to the CRTC’s hearings on Bell’s bid to buy Astral Media to be held in Montreal next month. Today was the deadline for submissions to the CRTC.

My submission is part of an intervention by the Public Interest Advocacy Centre, Consumers’ Association of Canada, Canada Without Poverty, and Council of Senior Citizens’ Organizations of British Columbia opposing the Bell/Astral deal. The documents were filed with the CRTC today.  All submissions to the CRTC can be found on its website here.  

Bell claims in its application to the CRTC that a combined Bell/Astral “will not exercise market dominance in any sector of the broadcasting industry” (emphasis added, Bell, Reply, A14c). My submission on behalf of PIAC et. al. argues otherwise and that the transaction deserves very close scrutiny, and that key elements of it should be stopped dead in their tracks.

The key findings in the submission can be summarized as follows:

  1. a successful bid by Bell to acquire Astral would catapult it to the top of the ranks in radio, with revenues of $500 million, 106 radio stations, just under 29 percent of the market – twice the size of its nearest competitors: Rogers, CBC and Shaw (Corus). Notwithstanding such an outcome, this would not trigger regulatory intervention under the CRTC’s new ownership rules or its Common Ownership Policy. Consolidation in radio increased in the early 2000s before drifting downwards in recent years. Radio is unconcentrated by conventional measures. The Bell/Astral deal, however, would reverse the tide and result in the highest levels of concentration in the past twenty-five years
  2. there would be no direct impact on traditional television broadcasting.
  3. in the specialty and pay television market, Bell’s market share would rise sharply from 28% in 2011 to over 42%. This gives the CRTC ample grounds to intervene.
  4. across the total television universe, Bell’s position would be reinforced, rising sharply from 27% in 2011 to 35%. This, too, provides grounds for intervention.
  5. television markets worldwide tend to be more concentrated than often assumed. Canada is, at best, a middle-of-the-road performer on this measure, and often at the high-end of the scale. While concentration is slowly declining elsewhere, in Canada it is rising sharply; the Bell – Astral deal will compound the trend.
  6. Canada currently has the second highest level of cross media ownership and vertical integration among thirty-two countries studied by researchers in the International Media Concentration Research Project (Columbia University). It will be the highest amongst these countries if the CRTC does not pull the plug on the Bell — Astral deal.

The following figure shows the story.

Crossmedia Ownership/Vertical Integration Ratios — Canada # 1 amongst 32 Countries Surveyed Worldwide

Source: International Media Concentration Research Project with updates for 2011-2012 for Canada by author

Conclusions Drawn

Ultimately, the submission concludes:

  1. The CRTC probably has no choice but to give a pass to Bell with respect to its take-over of Astral’s radio assets. Bell meets the Commission’s requirements under the Common Ownership Policy, or at least will once it divests itself of ten stations in Vancouver, Calgary, Winnipeg, Toronto and Ottawa-Gatineau. This is unfortunate because, until now, radio has been one of the least concentrated and most diverse media in the country. The Bell-Astral deal will increase concentration significantly, whereas in most countries covered by the IMCR study, it is declining.
  2. Television is a different matter. There will be no direct effects on broadcast television. There will, however, be large and significant effects on the specialty and pay television and “total television” markets. Concentration levels in all of these areas are already very high by the CRTC’s own standards, historical norms, global standards and by CR and HHI standards used to measure media concentration in this submission.
  3. The impact will be most extreme in the specialty and pay tv market, where Bell will increase its share of the market from 26.6% to 42.2% — well in excess of every other major player in the market: Shaw (32.3%), Rogers (10.7%), CBC (4.1%) and QMI (3.2%). Together, these five companies will control 92.5% of this market. Out of the eighteen countries for which adequate data is available, Canada currently is the 11th most concentrated market. If the Bell – Astral deal is approved, we’ll fall down another notch to 12th place.
  4. The trend is similar with respect to the “total television” market, but not quite as pronounced. On the basis of the CR, it is already more concentrated than it has ever been in the last twenty-five years. In terms of the HHI, things could soon be right back where they were in 1984, when the HHI score was 2307.5 and the VCR all the rage. By my calculation, the HHI score is presently 1918, up significantly from three years earlier when it was 1,481. Should the Bell deal go through, it will have 35% of the market and the HHI score will be higher still at 2308.8 – one point more than twenty five years ago. The CRTC’s own concentration rules permit it to intervene actively in the face of such levels, and it should.
  5. Lastly, Canada already has the second highest levels of cross-media ownership consolidation and vertical integration in the 32 countries examined by the IMCR project. We don’t need to be first. The CRTC ought to oppose this venture on this ground alone, although it is unclear whether it even as the power, let alone the will, to do so. Concentration within and across the network media industries –  demonstrably and empirically – has been extremely high, and is set to get higher yet.

It is time to reverse the tide.

Weak Links and Wikileaks: How Control of Critical Internet Resources and Social Media Companies’ Business Models Undermine the Networked Free Press

I’ve written several times on Wikileaks over the past year-and-half. In this piece I draw together and update my thoughts on Wikileaks in light of recent developments, with a focus on how concentration of ownership and control over critical internet resources (internet access, domain name registries, webhosting sites, payment services, etc.) and the business models of social media companies such as Twitter compromise freedom of expression and the press on the Internet, with Wikileaks serving to illustrate the point.

What follows is a first draft of a chapter that I have written for a forthcoming book edited by Benedetta Brevini, Arne Hintz and Patrick McCurdy. Beyond Wikileaks: Implications for the Future of Communications, Journalism and Society. I would be delighted to hear any constructive comments and criticisms you might have.

In his seminal piece on Wikileaks, Yochai Benkler (2011) makes a compelling case for why Wikileaks is a vital element of the networked fourth estate, and why we should view its harsh treatment by the U.S. government as a threat to the free press. As he says, the case embodies a struggle for the soul of the internet, a battle that is being waged through both legal and extralegal means, with major corporate actors – Apple, Amazon, eBay (Paypal), Bank of America (Visa), Mastercard, etc. – using their control over critical internet resources to lean in heavily on the side of the state and against Wikileaks.

This piece reviews Benkler’s case for seeing Wikileaks as an crucial element of the networked free press, adds a few details to it, then presents an important new element to the story: the role that Twitter, the social media site, has played in what I will call the Twitter – Wikileaks cases. In contrast to the pliant commercial interests that Benkler discusses, Twitter fought hard in a series of legal cases during the last year-and-a-half to avoid having to turn over subscriber account information for several people of interest to the U.S. Department of Justice’s ongoing Wikileaks investigation: Birgitta Jónsdóttir, an Icelandic MP and co-producer of the Collateral Murder video whose distribution over the internet by Wikileaks put it, and her, on a collision course with the U.S. to begin with, Wikileaks’ volunteer and Tor developer, Jacob Applebaum, and the Dutch hacktivist Rop Gongrijp.

The DoJ’s “secret orders” raise urgent questions about state secrets and transparency, the rule of law, internet users’ communication rights, and the role of commercial entities that control critical internet resources. The Twitter – Wikileaks cases also cut to the heart of journalism in light of how journalists routinely use social media such as Twitter and Facebook, but also search engines and internet access services, to access sources, share information, and generally to create and circulate the news.

Wikileaks and the Emergence of Next Generation Internet Controls

Information filtering, blocking and censorship have been the hallmark of China’s model of the internet since the 1990s. Now, however, we are at critical juncture in the evolution of the Internet, with the United States government’s anti-Wikileaks campaign showcasing how such methods are being augmented by a wide range of legal and extra-legal methods in capitalist democracies. Indeed, governments the world over now rely on multidimensional approaches that use technical tools to filter and block access to certain kinds of content while normalizing internet control through legislation and by out-sourcing or privatizing such controls to commercial internet companies (Deibert & Rohozinski, 2011, pp. 4-7). Among other things, the Wikileaks case shows that such actors are often all-too-willing to serve the state on bended knees, albeit with some important exceptions to the rule as the Twitter – Wikileaks cases discussed later in this chapter illustrate.

Three intertwined tendencies are stoking the shift to a more controlled and regulable internet. First, the concentration of ownership and control over critical internet resources is increasing: incumbent cable and telecom firms’ dominate internet access, while a few internet giants do the same with respect to search (Google), social media platforms (Facebook, Twitter), over-the-top services (Apple, Netflix), webhosting and data storage sites (Amazon) and payment services (Visa, Master Card, Paypal), among others. Simply put, more concentrated media are more easily regulable than many players operating in a more heterogeneous environment. Second, the media and entertainment industries have scored victories in Australia, UK, NZ, US, Taiwan, South Korea, France and a handful of other countries for three-strikes rules that require Internet Service Providers (ISPs) to cut-off internet users who repeatedly run afoul of copyright laws. A 2011 UN report condemned these measures as disproportionate and at odds with the internet’s status under the right to communication set out in Article 19 of the Universal Declaration of Human Rights (1948), but they remain operative nonetheless (La Rue, 2011). Lastly, the internet is being steadily integrated into national security and military doctrines, with thirty or so countries, notably the US, Russia and China, leading the push (U.S. Congressional Research Service, 2004). The U.S. Department of Defense’s revised “information operations” doctrine in 2003, for instance, defines the internet (cyberspace) as the fifth frontier of warfare, after land, sea, air and space (United States, Department of Defense, 2003). National security and law enforcement interests are also central in new laws currently being considered in the US (CISPA), Canada (Bill C-30) and the UK (Communications Data Bill).

These trends are increasing the pressure to turn Internet Service Providers (ISPs) and digital intermediaries into gate-keepers working on behalf of other interests, whether of the copyright industries or law enforcement and national security. This drift of events is already bending the relatively open internet, with its decentralized architecture pushing control to the ends of the network and into users’ hands, into a more closed and controlled model. Such trends are not new, but they are becoming more intense and firmly entrenched in authoritarian countries and liberal capitalist democracies alike. This is the big context within which the anti-Wikileaks campaign led by the U.S. government has unfolded.

Wikileaks and the Networked Free Press

There are counter-currents to these trends as well, and one of those is the rise of Wikileaks in the heart of the networked free press, just at a time when the press is struggling to find a sturdy footing in the internet-centric media ecology. While it is common to bemoan the crisis of journalism, Benkler (2011) strikes a cautiously optimistic note, laying the blame for the ongoing turmoil among traditional news outlets on their own self-inflicted wounds that have festered since the 1980s. The rise of the internet and the changing technological and economic basis of the media magnifies these problems, he argues, but the internet is not responsible for them. In fact, nascent forms of non-profit, crowd-sourced and investigative journalism may be improving the quality of journalism.

Wikileaks is part and parcel of these trends. In the events that put it on a collision course with the U.S. government, the whistle-blowing site burnished its journalistic credentials by working hand-in-glove, at least after the “collateral murder” video, with The Guardian, the New York TimesDer SpeigelLe Monde and El Pais to select, edit and publish the Afghan and Iraq war logs and embassy cables. By cooperating with respected journalistic organizations, Wikileaks material was selected, edited and published according to professional news values rather than driven solely by the logic of hactivism or being an indiscriminate and irresponsible dump of sensitive state secrets into the public domain. The collaboration between traditional news outlets and Wikileaks also demonstrated that gaining access to large audiences in a cluttered media environment still requires ‘big media’. Altogether, these efforts set the global news agenda four times in 2010. For its efforts, Wikileaks chalked up a bevy of presitgious awards for its significant contributions to access to information, transparency and journalism, adding to the long list of honours that it had already won from press and human rights organizations, including from British-based Index on Censorship, Amnesty International and Time Magazine, among many others, since its inception (see Wikileaks Press, nd).

Interestingly, while Wikileaks had been offering journalists free access to the war logs and embassy cables for some time, it was only after it offered exclusive national rights to The Guardian, New York Times, and other major newspapers around the world that journalists showed much of an interest. Rights, money, and market power are still important lures, and are cornerstones of market-based media, with or without the internet – although it is important that Wikileaks certainly does not follow the conventional commercial model, and offers an alternative to it.

The more important point for now, however, is that investigative journalism is not the exclusive preserve of the traditional press, but it is the signature feature of what Wikileaks does. That the interjection of Wikileaks into the journalistic process led to outcomes that are probably better than the ‘good ole days’ is underscored by the fact that while the New York Times consulted with the Obama Administration before publishing the war logs and diplomatic cables, it did not withhold the material for a year. Indeed, this is a big and important difference from its behaviour in 2005 when, at the behest of the Bush Administration, the New York Times delayed James Risen and Eric Lichtblau’s (2006) expose of unauthorized, secret wiretaps conducted by the National Security Agency in cooperation with AT&T, Verizon and almost all of the other major telecom-ISPs in the U.S. (Calame, 2006). The war logs and embassy cables stories likely became headline news in 2010 faster than would otherwise have been the case because of Wikileaks role in these events, and its strategy of playing news organizations’ competitive commercial interests off of one another. Moreover, with little need to maintain good standing with the centres of political, military and corporate power, Wikileaks never assumed levels of deference similar to the New York Times and other established news sources.

All-in-all, Wikileaks is emblematic of a broader set of changes that, once the dust settles, will likely stabilize around a new model of the networked fourth estate, an assemblage of elements consisting of (1) a core group of strong traditional media companies; (2) many small commercial media (Huffington Post, the Tyee, Drudge Report, Global Journalist, etc.), (3) non-profit media (WikiLeaks, Wikipedia), (4) partisan media outlets (, Daily Kos, TalkingPointsMemo), (5) hybrids that mix features of all the others and (6) networked individuals (Benkler, 2009). The fact that WikiLeaks is so central to these developments, and so solidly at one with journalistic and free press traditions, helps to explain why neither it nor any of the newspaper organizations it partnered with have faced direct efforts by the U.S. to suppress the publication of WikiLeaks’ documents (Benkler, 2011). If the story ended here, it would be a happy one, a triumph of a plucky, determined watchdog willing to take on the powers-that-be, without fear or favour, a testimony to the power of the internet to contribute to freedom of expression, the free press and the public’s right to know – in other words, democracy.

Using Ownership and Control of Critical Internet Resources to Cripple Wikileaks

Unfortunately, however, the story does not end there. The problem, as Benkler (2011) states, is that what the U.S. was not able to obtain by legal measures, it gained with remarkable ease from private corporations and market forces. Thus, buckling under the slightest of pressure, Amazon banished Wikileaks’ content from its servers the same day (December 1, 2010) that Senator and Senate Committee on Homeland Security and Governmental Affairs Chair, Joe Lieberman (2010), called on any “company or organization that is hosting Wikeleaks to immediately terminate its relationship with them”. Wikileaks quickly found a new home at webserver firm OVH in France but lost access to those resources after France’s Industry Minister warned companies on December 4 that there would be “consequences” for helping keep Wikileaks online. A day later, the Swedish-based Pirate Party stepped in to host the “cablegate” directory after they were taken off line in France and the US.

Yet, Wikileaks’ troubles didn’t end there because just a day before it was kicked out of France, the U.S. company everyDNS delisted it from its domain name registry. As a result, Internet users who typed into their browser or clicked on links pointing to that domain came up with a page indicating that the site was no longer available (Benkler, 2011; Arthur, 2011). The Swedish DNS provider, Switch, faced similar pressure, but refused to buckle. It continues to maintain the address that Internet users still use to access the site, but has faced a barrage of Distributed Denial of Service (DDoS) attacks for doing so.

As Amazon, OVD and everyDNS took out part of WikiLeaks technical infrastructure, several other companies moved into to disable is financial underpinnings. Over the course of four days, Paypal (eBay) (December 4), MasterCard and the Swiss Postal Office’s PostFinance (December 6), and Visa (December 7) suspended payment services for donors to the site. Two weeks later, Apple removed a Wikileaks app from the iTunes store (Apple removes Wikileaks, 2010). Thus, within a remarkably short period of time, a range of private actors cut-off Wikileaks’ access to critical internet resources. The actions did not kill the organization, but the financial blockade did contribute mightily to the fact that Wikileaks funding plummeted by an estimated 95 percent (Wikileaks, 2011).

Privacy Rights Online and Internet Companies’ Business Models: Weak Foundations for the Networked Fourth Estate and Communication Rights

One important entity has stood outside this state-corporate triste on the outskirts of the law: Twitter. Indeed, it has stood alone among big American corporate internet media brands in refusing to assist the United States’ anti-Wikileaks campaign. Faced with a court order to secretly disclose subscriber information for three of its users, it said no.

In December 2010, at the same times as Wikileaks was being cut-off from critical internet resources, the US Department of Justice demanded that Twitter turn over subscriber account information for Birgitta Jónsdóttir, Jacob Applebaum and Rop Gongrijp as part of its ongoing Wikileaks investigation. The information sought was not innocuous and general, but intimate and extensive: i.e. subscriber registration pages, connection records, length of service, Internet device identification number, source and destination Internet protocol addresses, and more (United States, 2011a, pp. 7-8). Twitter was also told not to disclose the request to the people concerned, and to stay quiet about the whole thing. It did none of this.

Instead, the company mounted a serious legal challenge to the Justice Department’s “secret orders” and pushed the envelope in interpreting what it could do to protect its subscribers’ information (McCullagh, 2011). In Twitter – Wikileaks Case #1, the social media site won a small victory by gaining the right to at least tell Jónsdóttir, Applebaum and Gongrijp that the DoJ was seeking information about their accounts (United States, 2010). They were given 10 days to respond before it was compelled to comply with the DoJ order. It also took the extra step of recommending that they seek legal help from the Electronic Frontier Foundation (EFF), a public interest law watchdog on all matters digital and about internet/cyberspace governance (a copy of Twitter’s letter to Gongrijp is available at Gongrijp, 2011).

The EFF has represented Josdottir on the matter since, while Twitter’s lead counsel, Alex MacGillvray, has stood for the company. Interestingly, Iceland has also weighed in by strongly criticizing the US over Jónsdóttir, while a group of 85 European Union Parliamentarians condemned the United States’ pursuit of Wikileaks. They were especially critical about how the US was harnessing internet giants to its campaign. They “failed to see” how, among other things, the Twitter Order could be squared with Article 19 of the Universal Declaration of Human Rights. More to the point, they worried the United States’ actions were contributing to the rise of a

“national and international legal framework concerning the use of . . . social media . . . [that] does not appear to provide sufficient . . . respect for freedom of expression, access to information and the right to privacy” (Intra-Parliamentary Union, 2011).

The first Twitter – Wikileaks case, or “Twitter Order”, was a shallow victory. It allowed the company to inform Jónsdóttir, Applebaum and Gongrijp that they were of interest in the DoJ’s ongoing Wikileaks investigation, but did not prevent the disclosure. Yet, even this shallow victory looks positive relative to how easily Amazon, Apple, eBay (Paypal), Mastercard, Bank of America (Visa), everyDNS, etc. enlisted in the United States’ campaign against Wikileaks. Twitter staked out a decidedly different position that insisted upon the rule-of-law, speaking out in public and going beyond what was necessary to help its subscribers ensure that their rights, and personal information, are respected.

The full perversity of these circumstancs only came fully into light in the Twitter – Wikileaks Case #2, when Jónsdóttir, Applebaum and Gongrijp appealed part of the first case to overturn, and thus prevent, the requirement that Twitter hand over their account details to the DoJ (United States, 2011a). The U.S. District Court‘s decision in the case in November 2011 had direct results and some potentially far sweeping implications.

The first direct result, as we have seen, is that Twitter had to hand over Jónsdóttir, Applebaum and Gongrijp’s subscriber information. Another, however, is that they have no right to know whether the DoJ has approached Facebook, Google or other Internet companies with secret orders, and if so, for what kinds of information, and with what results (p. 52). The courts seem to believe that neither they nor the public-at-large have the right to know the answers to these questions. For their part, Google, Facebook and Microsoft (Skype) have stayed silent on the affair despite their frequent pontification about internet freedom in a generic sense and mostly in relation to ‘axis of internet evil’ countries, such as Saudi Arabia, China, Russia and Iran, among a rotating cast of others.

If these results are not discouraging enough, more sweeping implications flow from two other directions in the second Twitter – Wikileaks ruling. The first is the poor analogy the court draws between the internet and banks to ground its decision as to why companies of the former type must hand over subscribers’ information just as readily as the latter do when served with a court order. There is a lot of potential discussion in this point alone, but for now it suffices to say that thinking about social media in terms of banking, insurance and clients is a long way from comprehending the internet as a public communications space.

Of more interest for here is the mind-boggling claim that internet users forfeit any expectation of privacy – and hence, privacy rights – once they click to accept internet companies’ terms of service policy. As the court put it, Jónsdóttir, Applebaum and Gongrijp “voluntarily relinquished any reasonable expectation of privacy” as soon as they clicked on Twitter’s terms of service (United States, 2011a, p. 28). Thus, instead of constitutional values, law or social norms governing the situation, the court ruled that privacy rights are creatures of social media companies’ business models. Social media users, according to the court, would have to be woefully naive to expect that privacy is a priority value for advertising-driven online media, given that almost the entire business model of major Internet companies is about collecting and selling as much information about audiences as possible.

But this is ridiculous because Twitter, Facebook and Google’s terms of service policies are about maximizing the collection, retention, use and commodification of personal data, not privacy. It is as if the ruling is intentionally out of whack with the political economy of the internet so as to give the state carte blanche to do with digital intermediaries as it pleases. Christopher Soghoian (2011) captures the crux of the issue in relation to Google, but his comments apply to Internet companies in general:

Google’s services are not secure by default, and, because the company’s business model depends upon the monetizaton of user data, the company keeps as much data as possible about the activities of its users. These detailed records are not just useful to Google’s engineers and advertising teams, but are also a juicy target for law enforcement agencies.

Conclusions and Implications: Wikileaks, the Networked Fourth Estate and the Internet on Imperiled Ground

Things don’t have to be this way. The idea that privacy rights turn on the terms of service policies of commercial internet companies rests upon a peculiarly squinty-eyed view of things and leverages the mass production and storage of personal data enabled by Twitter, Facebook, Google and so forth for the advantage of the state. But even if we took corporate behaviour as our moral compass, Twitter has occasionally distinguished itself, as it did during the London riots/uprising in August 2011 by refusing to comply with the UK government’s requests to shutdown its service and handover users’ information, while Facebook complied. Thus, even by the standards of corporate behaviour, Twitter’s behaviour could cultivate a higher sense of privacy amongst its users.

Concentrated Internet Markets and Small Details: Changing the business model of internet companies to minimize the collection, retention and disclosure of personal information, as the EFF recommends and as some non-commercial sites such as IndyMedia sources do, would be helpful., a small ISP with 45,000 internet subscribers in the San Francisco area, and which is also implicated in the Wikileaks case because Jacob Applebaum, a key figure in the Twitter – Wikileaks case, as we saw above, has been one of its subscribers, does just this. Most ISPs, in contrast, take the opposite view, as a cursory review of the terms of service policies from AT&T, Comcast. Verizon and Time Warner – the big four ISPs in the U.S. that account for just over 60% of internet access revenues (Noam, 2012) – illustrates. While may offer a model of a free and open internet that maximizes its users’ privacy by minimizing data collection and retention, the fact of the matter is that with less than .05 percent of the US internet subscriber base, it is easily ignored.

Ultimately, the relevant measuring rod of communication rights is not corporate behaviour or the market, but legal and international norms. Social norms govern how we disclose personal information in complex, negotiated and contingent ways, as well (Nissenbaum, 2011). Internet companies’ terms of service policies and the Twitter – Wikileaks cases largely ignore these realities, and thus are out of touch. These issues as well as the fact that the vast majority of people do not even read online terms of service policies — and those who do more often than not do not fully understand them — were brought to the court’s attention, but brushed aside. The decision at least makes it clear that the hyper-commercialized ‘free lunch’ model of the Internet comes with a steep price: our privacy rights and an entire industrial arrangement poised to serve as the handmaiden of the national security state.

The Virtues and Vices of Twitter: It is against this backdrop that the significance of what Twitter has done is clear. It has not flouted the law, but has been hoisted upon its own petard on account of its “business model”. But this is an irreconcilable contradiction of capitalism, to use a marxist formulation, and won’t be solved by simply changing Twitter’s business model. Nonetheless, Twitter went beyond just complying with the law to afford as much respect for users’ rights as circumstances allowed. We might also ask if Twitter’s recent adoption of a “transparency report” chronicling government requests for user information and to take-down certain content reflect lessons learned by the company in the midst of the anti-Wikileaks campaign as well?

There is no need to pretend that Twitter is the epitome of virtue, because it is not. While Google, WordPress and others have all signed on to the broad statements of principles regarding privacy and online free speech rights in the Global Network Initiative, for example, Twitter, along with Facebook, has not. But in this area, pontificating is rife, and while Google preaches transparency and open information absolutism, it has said nothing direct or substantial about the U.S.’s treatment of Wikileaks, or even if it has been implicated in the campaign.

Deepening National Security Imperatives: The U.S. governments’ campaign against Wikileaks further entrenches the post 9-11 securitization of the telecom-internet infrastructure, in the U.S. and globally, given the reach of the most well-known US telecom and internet giants (Risen & Lichtblau, 2006; Calame, 2006). Some courts have condemned expansive claims of state secrets and unbound executive powers when it comes to national security matters, but others seem to grant the state a blank cheque (United States, 2006; United States, 2011b). When the law has not proved serviceable, as the earlier discussion suggests, important U.S. government figures have tip-toed around its edges, compliant private companies in tow, to get what it wants. Congress has also stepped in occasionally to make legal what before was not, as in the passage of the much revised and expanded Foreign Intelligence Services Act (FISA) in 2008, which is now up for renewal again and set to pass with little opposition in Congress (United States, 2008).

The Global Dimension: The campaign against Wikileaks cannot be kept to narrow confines and readily spills over into wide ranging areas, including diplomatic and global internet policy angles, too. Nation-States, and the US in particular, are flexing their muscles and attempting to assert their sovereignty over cyberspace – a point that defines the Wikileaks case. Scholars such as Lawrence Lessig, Ropald Diebert, Jonathan Zittrain, Jack Goldsmith and Timothy Wu, among many others, have long shown that cyberspace is no more immune to government intervention than you or are I are immune to the laws of gravity. Struggles over the Internet Corporation for the Assignment of Names and Numbers (ICANN), the rift between Google and China, and the United State’s campaign against Wikileaks clearly expose that fallacy for what it is. Legitimate criticisms of U.S. dominance of critical internet resources has been a staple of global internet politics since the ITU’s tussles with ICANN in the late 1990s, through WSIS I & II (2001 – 2005), to the creation of the Internet Governance Forum (2005), and back again to the ITU in 2012 (Mueller, 2010). The Wikileaks case offers a rational basis for such concerns. Criticisms of the U.S. in the Wikileaks case by EU parliamentarians, for instance, are of this kind.  The Guardian newspaper in the UK made the same point, too, by choosing Jónsdóttir, Assange, Applebaum and Twitter’s chief legal counsel, Alex MacGillvray, for its list of twenty “champions of the open internet” in April 2012 (Ball, 2012). Many of the awards bestowed upon Wikileaks by respectable human rights and free press organizations before and after the organization’s Collateral Murder video, war logs and Embassy Cables trilogy in 2010 are of a similar kind.

The problem, however, is that legitimate criticism are often mangled when mixed with attempts by strong states in authoritarian countries to use them as a Trojan Horse to smuggle in even less appealing attempts to dominate their own sovereign slices of the internet. A balkanized collection of Web 3.0, nationally-integrated internet media spaces is the result.  To the extent that the anti-Wikileaks campaign feeds such a pretext and fuels the ‘clash of sovereigns’ on the internet, it is unhelpful.

At the opposite end of the spectrum, the Twitter-Wikileaks rulings may serve the U.S. government’s bid to drive Wikileaks out of business well, but they have also lit a fire in the belly of hactivist groups like Anonymous and LulzSec, for whom such things are their raison d’etre. It may not be too much to suggest that the whiff of the anti-Wikileaks campaign fresh in the air helped to bring about the demise of recent attempts to strengthen national and international copyright laws – e.g. SOPA, PIPA and ACTA — given that, like the campaign against Wikileaks, each sought to leverage critical internet resources to control content and further restrict what people can do with their internet connections. If that, in fact, is the case, perhaps the battering of Wikileaks may have unintentionally served a noble cause.

Perhaps we can take solace in that and the fact that the distributed nature of the Internet means complete copies of Wikileaks files have been scattered across the planet, beyond the reach of any single state, no matter how powerful: the ultimate free speech trump card in a way. Yet, the fact that Wikileaks is now floundering, one of its founding figures on the lamb, and its funding down to a tenth of what it once was means that we ought not be so sanguine in our views. Happy stories about digital democracy should not deter us from the harsh reality that important open media principles have already been badly compromised, and more are at stake yet. Indeed, the deep ecology of the Internet is at stake, and so too is how we will conduct our lives in this highly contested place.

Tales from New Zealand — the Ultrafast Broadband Internet: Digital Public Works for the 21st Century vs. Incumbent Interests?

Last month I visited Auckland, New Zealand to give a talk at the The Future with High Speed Broadband Conference organized by the Competition Commission (full paper here).

The aim was to assess the factors that might encourage or hobble the country’s plan to make ultrafast, broadband internet service available to all. The Ultrafast Broadband Initiative, and its counterpart for rural areas, looks like a digital public works project for the 21st century, with the government forcing a restructuring of the country’s backwards incumbent telecoms players and investing nearly $2 billion in rolling out a combination of fibre-optic and wireless connections to over 90 percent of New Zealanders in the next six- to eight years. However, as I discovered, there are several factors that significantly stand in the way of such ambitions.

New Zealand shares several things in common with Canada that could turn this project into a big ‘white elephant’. The most important similarities are (1) extremely high levels of media concentration (higher in NZ than Canada); (2) powerful and recalcitrant incumbents; and (3) being two of just four countries worldwide where bandwidth caps are nearly universal and set at exceeding low levels (Iceland and Australia are the other two, see OECD).

As luck would have it, the British comedian and actor Steven Fry was in New Zealand working on Peter Jackson’s new film The Hobbit at the same time I was in town. He set things up perfectly a day before my talk by lambasting the pathetic state of New Zealand’s internet service after his attempts to upload recently completed film footage were throttled and thwarted by the ridiculously low bandwidth caps of between 2 to 5 GB per month that come standard with most telecom-ISP plans.

New Zealand has “probably the worst broadband I’ve ever encountered”, Fry railed on Twitter. Turns itself off, slows to a crawl. Pathetic!”

Media, ministers, Telecom NZ spokespeople and the island was abuzz with a basic fact of life in the country that everybody knows, but which takes an outsider to cast a bright light on in none-too-polite terms: the country’s internet service sucks. That was my role too: the outsider who can say things that local industry-regulatory-political insiders cannot.

With attention on high alert, the attendance at my talk was likely higher than it might have been. Media coverage was good from day one, too, with ComputerWorld catching the gist of my talk as follows: “Get real on data caps, peering and Sky TV dominance, says Canadian professor”. A video of the talk is below.

The Minister responded the next day by trying to squelch any idea that things were as bad as I painted, or that new approaches to regulation are needed. The New Zealand Herald and ComputerWorld, however, have drawn directly on my paper since then to counter such a do nothing attitude.  We’ll know better next month how all this will play out when the Competition Commission publishes its much anticipated final report on the matter.

The New Zealand situation is interesting and important beyond its own inhabitants for several reasons. For one, for much of the last quarter-of-a-century, it has been the outpost of a ‘free market fantasy’. To supporters of such a view, deregulation would liberate telcos from the heavy hand of government intervention and competition, innovation and lower prices for better service would flourish for all as a result.

That never happened. The “free market fantasy” years were nothing short of a disaster.

More recently, however, the country has embarked on a series of seemingly forceful steps that would leave the free market fantasy years behind in favour of something altogether different. The four key steps in this process include the development of telecoms specific regulation and functional separation at Telecom NZ in 2006, followed by the launch of the Ultrafast Broadband (UFB) initiative by the right-of-centre, conservative National government after its election in 2008, and finally what some call the ‘nuclear option’ in telecoms regulation — structural separation — earlier this year.

According to many observers, such steps and a strong regulator are necessary to counter incumbents intent on thwarting the rise of real competition and open networks. The UK regulatory, Ofcom, for instance, argues that only once it stiffened its spine and required British Telecom to break itself into two parts —  one for wholesale, and one for retail — under the Openreach framework did telecoms and Internet development significantly improve in that country.

The new regime led to a huge influx of service-based competition, new investment, cheaper broadband prices and more internet providers, while broadband use increased significantly as a result. Prices for residential broadband services fell 16% per year between 2005 and 2007.

With an eye on the UK experience, New Zealand followed suit in 2006. The results to many observers have been impressive. As the Berkman study (2010) concludes, “in the two earliest instances where functional separation was introduced [UK and New Zealand], it had rapid effects on competitive entry, penetration, prices, and/or speeds” (Benkler, et. al, 2010, p. 84).

New Zealand’s Ultrafast Broadband (UFB) initiative charts new ground as well, both as a way out of a legacy of a muddling market and as a forceful response to the financial crisis of 2007-8. The scale and hefty investment in commercial and state-owned companies involved is unique, but several other countries have also begun to follow suit, not just for legacy networks, but for ‘next generation access’ (NGA) networks based on a combination of fibre optics and wireless, too, including: Australia, Italy and Sweden (see the OECD’s study on Next Generation Access Networks).

The state of telecom and internet development in New Zealand has indeed improved since these changes were implemented. There is a modest increase in competition in some telecom markets and an improved regulatory environment.

Nonetheless, the country still sits at the bottom of the pack when it comes to broadband internet development. Moreover, its rank has actually fallen relative to other countries. Thus, where the Berkman study ranked New Zealand 22nd out of 30 countries based on 2008 data, my ranking puts it at 28th out of 34 OECD countries based on 2010 data. The following table shows the results.

Table 1: Country Ranks Based on Weighted Averages for Broadband Penetration, Price and Speed (2010 Data)





Overall Weighted Avg Rank

1 Sweden





2 Japan





3 Finland





4 Korea





5 Denmark





6 France





7 Netherlands





8 Norway





9 UK





10 Estonia





11 Slovak Rep





12 Australia





13 Iceland





14 Austria





15 Italy





16 Germany





17 Switz.





18 Belgium





19 Portugal





20 Slovenia





21 Czech Rep.





22 Poland





23 US





24 Hungary





25 Canada





26 Greece





27 Israel





28 New Zealand





29 Lux





30 Ireland





31 Spain





32 Turkey





33 Chile





34 Mexico





Note: Prices in USD PPP and include line charges (where applicable). Penetration is a composite of fixed broadband subs/100, households and mobile broadband; Speed is based on average advertised download speed plus fastest speed; Price on low, mid and high-end offerings. Source: OECD (2011). Broadband Portal.

The significant decline in New Zealand’s broadband conditions relative to other countries, slipping from 22nd to 28th, reflects the fact that other countries are also pushing similar initiatives, too, seemingly faster than New Zealand.

Ultimately, the UFB could be the great national digital public works project of the 21stcentury, but if New Zealanders are to realize its full benefits, they must confront several realities head on: (1) high levels of media concentration; (2) restrictive bandwidth caps, (3) low levels of media and internet use, (4) a regulator that is not yet accepted as an essential element in an open, competitive and pluralistic media environment, and (5) strong incumbents intent on bending new technologies and possibilities to their ends.

The last point is particularly important because whilst all the policies and indeed the UFB initiative itself is based on an open Internet model that places as much of the capabilities and resources of these networks at the ends of the network and on to the desktops and into the hands of as many online service providers and users as possible (Saltzer, Reed, & Clark, 1981; Isenberg, 1996; Benkler, 2006), New Zealand’s telecom and broadcasting industry incumbents are hell-bent on creating a supplier-driven walled garden model, circa 1999, where power and resources reside in the core of the network, owned and controlled by network operators and their business partners.

A series of unregulated deals struck between Sky TV — the local monopoly provider of pay-tv services delivered by satellite and local arm of the global behemoth, News Corp. — and all of the key telecom and ISP players (except Orcon) since late-2009 reveal a full-court press by the incumbents to simply graft the ultrafast broadband Internet now being built out mostly at government expense onto their current business models. If they succeed, the media economy will not expand as much as it could; nor will it be as pluralistic. Media and internet use will also likely remain low, because uninspired. Competition, diversity and an open Internet, in short, will be crushed in the name of preserving incumbent interests.

For those who are fans of structural separation, the New Zealand case shows how even that form of regulation can be subverted. It also shows that strong measures are needed to enhance not just network neutrality but also to deter alignments between network and content providers alike designed to throttle competitors and to maintain their own position at the centre of the network media universe.

This is particularly problematic in the context of the UFB because, left unaddressed, one gets the impression that the government is financing the roll-out of a state-of-the-art broadband Internet for incumbents in the telecoms and media industries. Perhaps that’s how we can square the  decidedly right-of-centre government’s decision to publicly-fund such a project to begin with?

%d bloggers like this: