A study [ pdf, 368KB] by a New Zealand policy research firm is calling into question billions of dollars of government stimulus for upgrading broadband connections to fibre.

In a paper called The Need for Speed: Impacts of Internet Connectivity on Firm Productivity, researchers at Motu Economic and Policy Research claim to provide the first firm-level estimates of the degree of productivity gains sourced from upgraded internet access.

We use a large New Zealand micro-survey of firms linked to unit record firm financial data to determine the impact that differing types of internet access have on firm productivity. … Having matched firms, we examine the productivity impacts that arise when a firm adopts different types (speeds) of internet connectivity. Broadband adoption is found to boost productivity but we find no productivity differences across broadband type.

Translation? The study finds a productivity benefit associated with a firm being connected to an ‘always on’ internet connection, but it could not show a benefit associated with higher speeds.

The implication? Billions of dollars are being spent by some governments on a belief that upgraded infrastructure is necessary to improve national economic productivity. Some governments, perhaps impatient with the pace of natural evolution of network development, are imposing structural separation or even nationalization of access infrastructure.

This study appears to affirm the benefit of ensuring universal access to a broadband connection; the challenge is in demonstrating a benefit from government intervention in broadband network upgrades.

If the Motu study is really the first to study the impact of broadband upgrades, one would hope that a fraction of the billions of tax dollars being spent around the world will fund econometric research to make sure that it is a worthwhile investment.

The Globalive ownership Decision from the CRTC may have an error.

Frequently, the expression ‘regulatory symmetry’ refers to the even handed application of regulations across various technologies or industries.

In today’s context, I use the expression in its literary sense.

Last Thursday’s CRTC Decision that reviewed Globalive’s ownership was complex: 119 paragraphs, organized under about a dozen level 1 headings and a few of those sections had sub-sections.

It is a well organized and readable piece – and for those who are willing to take the time, the decision lays out the difficult arguments and issues that were considered by the Commission in reaching its conclusion.

The literary symmetry that I want to refer to is in paragraph 115 of the Decision. The language in the decision reads:

Globalive has made numerous significant changes to its corporate structure and documents in order to address many of the Commission’s concerns. In this decision, the Commission has identified additional changes that are necessary to address certain remaining concerns with respect to Orascom’s influence over Globalive. These changes relate to the composition of the boards of directors, liquidity rights, and the threshold for veto rights.

However, if you look through the body of the Decision, you will see that the term “should” shows up in paragraphs 45, 64 and 72. These are the final paragraphs in sub-sections titled: “Composition of the boards of directors”; “Eligible Purchasers”; and, “Veto rights”.

Contrast these three items with the last sentence in paragraph 115. That paragraph, quoted above, indeed refers to 3 changes being necessary, but the second item listed in paragraph 115 says “liquidity rights”, not “eligible purchasers”.

In fact, there is a section called “liquidity rights”, but paragraph 59 in that section simply observes that this is an “an indication of Orascom’s influence” without suggesting that a remedy is required (such as the use of the word “should” which appears in the other paragraphs). As such, the last sentence in paragraph 115 appears to have had a literary asymmetry – a dissonance that doesn’t align the concluding paragraphs with the body of the decision.

This is what leads me to believe that an erratum is needed to clarifies the summary and restores a kind of talmudic balance.

Like straightening out a painting on a wall – did that dissonance bother anyone else?

More importantly, does that dissonance lead to confusion on what is required of Globalive to cure the CRTC’s foreign control concerns?

Each year, the CRTC conducts an extensive data collection exercise, working together with Industry Canada, Statistics Canada, BBM Canada and BBM Nielsen Media Research, BBM Analytics for Media Technology Monitor, comScore, and Mediastats.

Despite this collaborative effort, it is quite possible that the Communications Monitoring Report undercounts business internet accesses in Canada.

The issue is partly due to definitions used in the data collection process.

The CRTC’s Form 253 is the primary source of information for internet subscriptions.

That form asks for information about internet subscriptions, not internet connections, broken out by residential, business and wholesale.

Look at a business with 5 locations each having internet access. How does Form 253 count this customer? Is it one subscription or is it 5 connections? Note that “subscription” is defined by the CRTC as:

Internet access subscriptions are measured by the number of unique IP addresses which may be used simultaneously by different parties.

Is the term “subscription” an adequate way to determine internet connections?

Note that Form 253 may also not consider internet connectivity enabled over integrated communications services access. The form asks about internet subscriptions.

Consider a business that buys a hosted telephone solution. In this case, the internet connection comes for “free” with integrated data pipe that connects the business to the service provider. For Form 253, is the internet connection for that business scored at all?

A further enhancement to the data collection process might be to identify fibre connectivity for businesses and residences. Already, the forms [252 and 253] seek to distinguish ‘fibre’ from DSL and other technologies, but this may not be sufficient resolution to enable meaningful international comparisons.

For example, if there is a fibre connection to an apartment building and then DSL or cable modem connections from the utility closet to the customer, should fibre to the premise accesses be captured separately?

Cogeco Data Services has announced a 10 year deal with the City of Toronto to expand the fibre network that already connects all schools and school board buildings (as we wrote about in March), under separate deals with the Toronto District School Board and Toronto Catholic District School Board.

Under the terms of the deal with the City announced earlier this week, Cogeco Data Services will provide a fibre optic-based wide area network to various departments, agencies, boards, and commissions, serving a majority of the City’s 50,000 employees.

The new sites will be installed over a three year period with the first coming on-line later this year. Cogeco Data Services expects 95 per cent of construction completed by 2011.

There is a lot of fibre being deployed in a variety of communities by telephone companies, cable companies and alternate providers such as Atria and Cogeco Data Services. Businesses and residences are being served with fibre to the premises solutions, often with multiple suppliers in a building.

How do we improve statistical data collection to improve our understanding of the state of advanced networks in Canada? We’ll at this question some more on Monday.

The CRTC issued its decision in the review of Globalive’s ownership. Reading some of the re-tweets and comments on various news sites and blogs, I see the distain for our regulator that might be expected from those who don’t have an understanding of Canada’s structural framework.

As Michael Geist wrote:

It is tempting to blame the CRTC or the incumbent telecom providers (who filed the complaint over the Globalive structure) for this mess, but the real culprit lies with outdated legislation that prioritizes Canadian ownership over a competitive Canadian marketplace.

I agree with that assessment. Last month, I wrote about the Telecom Policy Review Panel’s views on liberalization of foreign ownership.

One of the Globalive lawyers, Hank Intven, was a member of that panel that delivered its report three and a half years ago. Shortly thereafter, Hank was commissioned to develop a book detailing the specific legislative changes required to implement the recommendations of the report.

The Panel was created by the last Liberal Government and its response was delivered to a the new Conservative Minister of Industry. I remember asking a local Liberal MP if the bipartisan origins of the TPRP would permit a more constructive legislative approach to speed discussion and passage of the package of reforms.

Silly me.

So here we are. Where do we go from here? To those who wonder how Industry Canada could approve the ownership and then watch the CRTC deny approval, there are a number of reasons that we can discuss.

The CRTC set out the remedies required to gain approval in its decision. The fastest and most certain course would be for Globalive to bring its structure into conformance and receive authorization from the Commission.

Alternatively, Globalive can appeal to Cabinet, but with such a process, the Telecom Act has prescribed timetables to allow consultations with the provinces and the public. This would not be fast and I’ll ask lawyers to consult on whether Cabinet has the legal authority to simply ignore the foreign ownership restrictions entirely. It is possible that the path of least resistance would be for Globalive to ask Cabinet to order the CRTC to delete paragraph 118 of the Decision – the one dealing with Orascom holding the debt – perhaps using a justification of the exceptional circumstances of the current economic situation.

Alternatively, Globalive could sit on the spectrum asset, waiting for legislation to remove the foreign ownership restrictions. I’m not holding my breath.

One way or another, there will be new players in the Canadian mobile space. Public Mobile, DAVE Wireless and Videotron are also building networks.

At the end of the day, Globalive entered the spectrum auction fully aware of the ownership regulations and precedents for previously approved corporate structures. It acquired new entrant spectrum that had been set aside to enable smaller start-ups to offer increased wireless choices to Canadians and it outbid the others perhaps hoping that it could push the envelop a little further.

All along, Globalive has been in control of the timetable for approval. The next move is up them as well.