Bell has been using this as an excuse for decades now while barely providing any services to rural areas. More than 90 percent of Canadians live within 150 miles of the US border in areas that are networked. Apparently, this 90 percent is paying for infrastructure that does not exist.

http://canadianspectrumpolicyresearch.org/canada/inventory/c...

Considering things from the mobile perspective (but largely true for internet too):

> Yes, most of it is empty and unserviced

Well, that's the key, isn't it. Look at the highly optimistic coverage maps from the companies and you'll see how little of Canada is covered. Actual coverage is significantly poorer than those coverage maps indicate outside urban areas.

> but building a national network costs the same amount no matter how many users it has

These "national" networks are a few paltry routes across the country and in no way compare to the national networks of other countries.

The most obvious counter to the apologists' landmass argument is the provinces of Manitoba and Saskatchewan where mobile rates were half that of the rest of Canada despite having lower population densities than most of the other territories. They both had robust state-owned companies that competed with the big 3 (likely soon to be big 2) and all 4 profitably operated at these lower rates. In a very controversial move, the state-owned company in Manitoba (MTS) was privatised in 1996 and bought out by one of the big 3 in 2017. A few years later the rates in Manitoba rose to match the rest of the country. This highlights the real reason for expensive mobile in Canada.

The vast majority of Canadians are concentrated and live near the US border.

That's true, but Canada has suburban sprawl just like the U.S. where, in most cities, home internet is just as expensive if not more so. Think how much less wire, amplification/splitter/switching/etc. equipment, and drilling would be required to install new home internet infrastructure to serve the same number of people if the Greater Toronto Area was as dense as the (4x denser) Stockholm metro area. The one of the obstacles (in addition to monopolies and regulatory capture) may be that the built-up area near the U.S. border still aren't dense enough to have infrastructure efficiency comparable to European countries with cheaper and faster home internet like Sweden, Finland, France, etc.

This is not really true. Yes, there are economies of scale to be had, but if you're talking landmass of Canada, then the primary cost is laying fiber, which tends to be a one-time cost (some caveats there) almost entirely subsidized by the government.

Telcos can then expand networks in cities, towns, regions, based on expected utilization (some of which is also subsidized.) Most of the costs here are operational: management, maintenance, customer support, however there are also costs associated with customer premisis equipment, aggregation hardware, copper/fiber to the homes. (These are not associated with the total landmass of Canada.)

Telecom providers in Canada are extremely monopolistic, very aggressive in keeping competitors at bay (if anyone remembers the huge full-page ads that the big Canadian providers put out when Verizon was planning on entering Canada -- "Americans are going to take our jobs!!!".)

Canada has subsidized networks for rural users. We've, as a population, invested a lot of money in connection remote communities. And next generation internet is going to be powered by 5G, which will make it a lot cheaper to connect remote communities.