Mobile network operators have to license wireless spectrum from the government. A simple analogy for this would be radio stations, the kind that broadcast music and news over the air. Each one needs to broadcast on a different frequency (spectrum) so that when you tune in to one their signal isn’t overlapping with another.

The author of the Financial Post article attributed a hefty percentage of consumer wireless costs to Canada’s spectrum policies. To prove the point, the article included a calculation to show how much it currently costs to license spectrum, the cost of borrowing the money for spectrum, and how that translates to higher wireless rates when compared to other countries.

What’s not to believe? The article was written by a Senior Fellow at the Technology Policy Institute in Washington! The first clue that something’s not quite right is the disclosure that he is also currently working for TELUS as a consultant.

A flawed assumption breaks the whole analysis

When I read the article I noticed that everything started with the cost of capital (borrowing money, in essence) and that it was very high. The Weighted Average Cost of Capital (WACC) was assumed at 20% before taxes. In simple terms, this is essentially the interest rate on the loan and I’m fairly certain only loan sharks and credit cards would charge that much. Telecommunications companies have access to less expensive loans, closer to mortgage rates.

Given that the sensitivity of the telecommunications sector is low, these companies should be able to borrow at less than 5% interest. Their required equity returns cannot be more than 10%.

I double-checked my assumptions with the help of a seasoned analyst. He immediately agreed that the WACC should never even remotely approach 20%. Across most of the market, a more realistic rate would be between 6% and 7%.

So, how is it that the author of this Financial Post op-ed made an assumption so out of line with the reality of the market? It’s possible the author was given some bad advice, or that the assumption was made to drive a specific outcome.

Regardless of the motivation, the entire calculation is broken by this initial, flawed assumption about WACC.

The fact is that money for spectrum purchases is available at much lower rates. Spectrum is also non-depletable and newer technology improves its efficiency (for example, the minimum four-fold improvement we saw moving from 3.5G to 4G LTE). Plus, depreciation on these capital investments is currently accelerated. Combined, this greatly reduces the tax burden on profits, another fact that is completely ignored by the author.

Auction results show who really drives up the price of spectrum

Throughout the article, shots were fired at both the new entrants (Freedom, Videotron, etc) and the CRTC spectrum set-aside rules that favour them. The rules involve the government setting aside some spectrum for new companies so that they no longer have to outbid the big three.

The author’s position is that set-aside rules reduce spectrum supply for incumbents, increasing the costs for all of them. These same new entrants who benefit from the set-aside aren’t yet using everything that they licensed.

This fails to consider the actual results of spectrum auctions.

The auction rules use the second-highest bid to set the spectrum pricing. At least two telecommunications companies have to put an “inflated” value on the spectrum, and their bids surely reflect the returns they expect to generate by securing the spectrum.

No one is forcing bids to reach a certain level. The bids reflect the business case behind them, namely that very high profits could be made by purchasing spectrum at a price that keeps it out of competitors hands.

The set-aside rules were created because spectrum bids were already supranormal and prevented new competitors from entering the market. The cost of spectrum in other countries reflects the expected return on investment, which is lowered due to more intense competition (often driven by MVNOs).

The auction results also show that spectrum value is concentrated in urban centres where demand is high. It’s true that new entrants have not expanded quickly into rural areas and small towns, but spectrum prices in those areas are already low (supply and demand). Set-asides in these areas have not contributed greatly to spectrum costs or reduced supply in a meaningful way. Besides, incumbents have played a big part in making expansion more difficult for new entrants.

There are direct benefits to consumers as a result of the auction, as well. Government programs, like the broadband fund, are used to drive investment into markets otherwise uninteresting to big telecoms. These are often the same areas that the new entrants have yet to expand into. It's because of these funds that incumbents have expanded their networks, and the cash behind these funds has to come from somewhere.

At some point, the telcos need to accept that they’ve made their own bed and now it’s time to sleep in it.

It’s not spectrum costs, the telcos are driving inflated wireless rates all on their own

If there’s one thing I’ve learned working in the industry, it’s that telcos are good at spending money. They will hire a company to build the network, another company provides the equipment, technicians are outsourced, call centres are outsourced, marketing automation is run by an agency… the list goes on.

So in an environment where everyone is given a budget to spend, and spending it all gets you a bigger budget next year, these companies have become very good at spending lots of money.

When you’re spending money, you always need more of it. Last year, Average Revenue per User increased by almost 7%. When the coffers need to be filled the telcos tax their subscribers and earn more revenue, with average ARPU increasing 3 our the past 4 years according to the Communications Monitoring Report.

So how do we reduce Canadian wireless rates?

The most obvious opportunity is for telcos to spend less money. This isn’t the strong suit of these organizations - they are so large that they inherently have to be everything for everyone. This approach to the market forces Canadians to pick from homogeneous options, forcing many to choose between a rock and a hard place.

What if, instead, independent virtual operators (MVNOs) could focus on building products designed for specific types of Canadians with unique needs. These companies could shed the products, processes, and expenses that aren’t required to serve these needs. An optimized organization would incur less cost to deliver a better experience, reducing the price to the end-user in the process.

Furthermore, the efficient use of infrastructure through network sharing will reduce unnecessary redundancy and improve utilization. TELUS and Bell are already doing it, each only covering half the nation with their own radio access network. Virtual operators only will further tap into these efficiencies, to the benefit of Canadian consumers.

At dotmobile, we believe that many seniors, youth, students, new Canadians and visitors to the country find themselves without a solution that fits their needs. We have plans to meet their needs, but also understand that we won’t be able to build a product that works for each and every one of them. If we did, complexity and cost would grow to a point that we mirror the incumbents and no efficiency is gained.

We expect that other brands or Canadian entrepreneurs will find innovative ways to meet the needs of those that we are unable to. Together, independent MVNOs will find ways to optimize the use of existing network infrastructure for those who fall between the cracks today.