Is the government a learning organization?

One of the strengths of the modern military is that they have determined to be and remain “learning organizations”: tactics are obsessively analyzed, and information shared among officers, so that battles are won, casualties reduced, and what needs to be changed, is changed. The opposite seems to be true in telecom policy.

The government says that it is pursuing a policy of inducing more competition into telecommunications services. There are two ways to do this. One is to engage in spectrum set-asides that have the effect of lowering entry costs for new players, and the second is mandated resale of underlying capacity held by incumbents. I dismiss the third idea, of reducing foreign ownership limits, because that would lead to Verizon purchasing Bell and some other behemoth buying Rogers and, absent rule changes, nothing else would change. So we are back to the two methods that might conceivably induce more competition into Canada’s comfortable oligopoly.

Spectrum set asides have been pursued by Liberal and Conservative governments alike, with equal zeal and equal futility. Emily Jackson’s article today in the National Post may have been composed with the reporter’s tongue in her cheek. It reads like a satire.

Citing the need to increase competition and affordability, ISED will set aside 43 per cent of the 70 MHz of available spectrum for potential new market entrants and regional competitors.

A few paragraphs later, the report says:

It’s a similar framework to the 2008 spectrum auction that resulted in the birth of Videotron, Mobilicity (eventually bought by Rogers), Public Mobile (now owned by Telus) and Wind Mobile (now Shaw’s Freedom). Startups got spectrum for cheaper than they would have in an open auction.

Thus three out of four start-ups were bought by incumbents: chomp, chomp. And this is not the first time this has been attempted. There was a spectrum set aside in the late 1980s that led to the same pattern of start-up and buy-out. The name Microcell comes to mind. Spectrum set asides lead to entrepreneurs getting in, taking a bath, and being bought out for the value of their spectrum.

Ah! But never let a failed attack reduce the belief that another assault in the same place and the same way will also fail to succeed.

I think the problem is memory, or the lack of it. I can remember these failures. More than half the people who could recall telecom policy in 1995 were employed by the incumbents, and they would see the failure of set asides as a win for their teams. But even if one had a civil service of exceptional duration, it would be difficult to find a way to recall that something that failed in 1995 would also fail 13 years later. People move on, they are promoted and retire. No historians, official or informal, have written the history of telecom policy in Canada since 1980. No one has written a manual or monograph for the Telecom Policy Branch of Industry Canada or whatever it is called  this week, and if one were, it would have to be scrapped and rewritten to accommodate intersectionality or some such flavour of the year.

And thus we come to the second line of attack, the one the CRTC declined to make last week, again. The Canadian Chapter of the Internet Society has expressed its disappointment in civilized and reasonable terms. From my perspective, the only policy that correctly analyzes the problem of oligopoly is one that sees competition in telecommunications in terms derived from the Internet.

Competition can exist in the applications layer that cannot feasibly exist in the material or physical layers. The cost characteristics of the tubes, pipes, towers and huts, and backhaul remain intractable barriers, and why duplicate infrastructure when you do not have to?

An MVNO like Ting can produce customers for Sprint in the US, for cheaper than Sprint can produce  them for itself, and operate out of Canada. Clearly, the right level and form of disentanglement has been found between the MVNO and the incumbent.

Arguments that the level of investment in infrastructure will be harmed by allowing mandated resale have merit if the wholesale price is miscalculated. That alone seems to me to be a serious concern.

The fetishism of investment in physical infrastructure as the sole criterion of a successful telecom policy reminds me of the “one big tractor factory in the Urals”. We know where that idea of the economy led.

But the CRTC, or its leadership, is persuaded of an erroneous conception of what competition means in an Internet era, and further study and reflection – which it promises – will not change minds that were not influenced by rational argument in the first place, or people who were not bold enough to realize that risks must be undertaken to get to desirable policy outcomes. A safe and comfortable life awaits the cautious appeaser of the incumbents. In the meantime, Canadian prices and rates of adoption of new applications remain unacceptably high and slow.



I am grateful to Len St.Aubin, the former Director General of Telecom Policy Branch, Industry Canada, for the following recital of facts in relation to spectrum set asides:


Government measures to promote wireless competition began in 1983 when the former Department of Communications decided to establish a national duopoly by dividing the available spectrum in two (a form of set-aside). Cantel, a company owned by Ted Rogers and other investors was awarded a licence on December 14, 1983; at which time the Telecom Canada incumbent local telephone companies were invited to submit applications for sub-band B. All licensees were authorized to begin offering service on July 1, 1985. The new competitor was incorporated in 1984 as Rogers Cantel Mobile Communications Inc. Over the next four years, Rogers bought out his partners to become the sole owner of Cantel, and the wireless operator was eventually folded into the company known today as Rogers Communications.

In 1995, measures were again taken to foster competition with the release of spectrum for digital “Personal Communications Services” (PCS). Of the 80 MHz available, the Department awarded two national 30 MHz PCS licences, one each to new entrants Clearnet PCS Inc. and Microcell Networks Inc.; awarded a national 10 MHz licence to Rogers, as well as a 10 MHz licence to each of the 11 regional shareholders of Mobility Canada for use in their respective operating territories. At that time, the governmentalso implemented a spectrum “cap”, which placed an aggregation limit on the amount of spectrum that any one entity or its affiliates could hold. The spectrum cap was set at 40 MHz and applied to frequencies for PCS at 2 GHz, cellular radiotelephony, and similar public high-mobility radiotelephony services in the 800 MHz range. In 1999, the cap was raised from 40 MHz to 55 MHz, when it was announced that 40 MHz of additional radio spectrum for PCS would be licensed by auction. The auction took place in 2001.

This cap was subsequently removed in August 2004. 

Next came the 2008 AWS spectrum auctions aside… [discussed in the article above].

To top