How Come My ISP Won't Increase Internet Speed and Lower My Bill, Like They Do in Sweden?

We Americans are getting addicted to our high-speed broadband connections. Unfortunately, they are often slower and more expensive than the Internet hook-ups you can grab in many other developed nations.

For example, my brother pays $40 a month for his 100-megabit broadband connection in Karlstad, Sweden. He can take his pick from 19 different service providers, all using a common last-mile infrastructure and competing on price and features. For $70 a month, he could upgrade to a full gigabit.

Me, I'm stuck paying $83 a month for a 50-megabit connection. Moreover, my upload connection rarely goes past half the speed Verizon (nyse: vz) promised. So I'm paying more and getting less, and if I wanted Verizon's fastest available FiOS connection, I'd be paying $300 a month (plus taxes!) for half a gigabit.

It's pretty clear that the Swedish model delivers more value and lower prices than the American way. And it's all made possible by a robust collection of municipal networks.

So why isn't America following the municipal path to high-speed bliss?

... it's complicated
Here's what happened in Sweden, in a nutshell.

As the 1990s drew to a close and the Internet started to become a big deal, Swedish consumers and businesses started thirsting for faster and more reliable connections. An industry emerged to serve this unmet need, much like anywhere else.

But there was a problem. Formerly government-owned telecom giant Telia had already installed a nationwide fiber-optic network and was unwilling to share network access with other operators. One by one, county and municipality leaders started drawing up their own plans for local high-speed networking -- and reaching out to one another to connect the separate networks into a nationwide loop.

This began with the larger cities, before spreading to suburbs and downright rural areas. Encouraged by the success of local leadership projects, the Swedish government started pumping additional funds into this expansion -- and regulating the newborn industry.

Take government funds, and you'll have to commit to opening up the network for others to use. This is crucial.

The fiber expansion is still going on, and slowing down as the return on investment shrinks when the fiber stretches into thinly populated markets. But slowing down is not the same thing as stopping development altogether.

Today, 57% of Swedish households have access to 100-megabit broadband services. Nearly 200 of Sweden's 290 municipalities offer a so-called city network, under public or private ownership but conforming to the same open-access principles.

The government aims to reach 90% of the population with 100-megabit broadband by 2020 and has approved $500 million of funds to support rural expansion over the next six years. Moreover, each one of the nation's 21 counties has a separate budget to coordinate these broadband expansion efforts.

City networks: Orange = municipal, black = private, gray = missing. Source: SSNF.org

The Swedish model today
That's the history and a snapshot of the current situation. The end result is that a majority of Swedes have direct access to a broadband infrastructure that was developed, funded, built, and supported by local and national government bodies. In a few places, the local electric power utility has played some or all of these roles -- but these tend to be run by the government anyhow.

This last-mile infrastructure and the connecting fiber loops are accessible to a variety of telecoms, cable service providers, and Internet specialists. All of them have equal access to the consumer-facing and core networks, on "equal and non-discriminatory terms."

The companies and government agencies that built these networks to begin with often sold consumer services at first. Today, that's incredibly rare, as third-party service operators have come to dominate the end-user market. Intense competition between them forces prices way down and gives the service providers incentive to develop new selling points for the same basic network access.

But again, all of these benefits spring from open access policies, where anybody can compete on equal terms.

And therein lies the rub.

This is why we can't have nice things!
The American broadband infrastructure grew from a very different place.

First, Sweden is about the size of California, with the population of North Carolina. With about 24 citizens per square kilometer, compared with America's 35, the population density is much thinner. Kind of like a scaled-up version of Mississippi.

The economic challenges of criss-crossing Sweden with fiber-optic broadband are very different from covering the New York cityscape or the big-sky vastness of Wyoming. But that's just a huge math problem, and not the biggest difference.

The real game-changers are political in nature. Funding and building a coast-to-coast broadband network would meet massive resistance in the tax-averse American market. And if tax dollars went into building the darn thing, why would the government then turn the network over to others, rather than drawing a profit from the huge investment?

Now, the early days of the Internet itself saw two American backbone networks develop much like the Swedish municipal networks are doing much later. The early ARPANET and NSFNET networks were built and operated by government agencies, before merging and then getting privatized in 1995. The commercial core network operators that emerged then still run the show today as Tier 1 networks, alongside a few later additions.

But things were different back then. Way different.

When commercial forces took control of America's Internet backbone, about 16 million global Internet users surfed 19,000 websites on primitive browsers like NCSA Mosaic, Netscape Navigator, or the text-based Lynx.

Today, 3 billion Internet users worldwide use smartphones and modern browsers to access over 1 billion websites -- and a ton of content available only to specialized apps.

The core network is no longer serving up a trickle of text and simple graphics to a microscopic niche market. The Internet has gone mainstream, including bandwidth hogs like online video and digital health services.

So the game has changed, while the fundamental business model hasn't. Commercial interests have built out the backbone networks we use today, and they rightly expect to see a decent long-term return on their investments. It's not about serving the community with low-cost, high-quality services. Serving investors sets up a very different incentive structure.

It all comes down to this:

The Swedish model is using public funds to create an Internet service that provides consumers and local businesses with affordable and powerful services. Turning a profit is a secondary consideration, if at all. This is true on several levels, from the nationwide backbone to local last-mile services.

The American model is powered by private, for-profit organizations. On the next level down, consumers are facing a Balkanized patchwork of cable, fiber, and DSL services with minimal competition and zero infrastructure sharing. Flooding or overriding this system with government support would be politically impossible, so we're stuck with this framework. That means focusing on profits over service quality, and there is no incentive at all to lower the cost to consumers.

Final words
The American status quo could still change, but not easily. Google (GOOG -4.37%) (GOOGL -4.48%) Fiber is throwing a wrench in the monopolistic pricing tendencies, as local telecoms and cable companies tend to introduce cheaper and faster Internet services wherever Google's services pop up.

Increased competition brings price pressure. Sweden has a lot of it, thanks to an open infrastructure and low profit expectations. American consumers have very little, for a variety of political and historical reasons. Google Fiber and similar upstarts can narrow the gap a little bit, but never the twain shall meet.