Friday, April 6, 2018

How Much Cost Can Muni Broadband Cut Out of the Business?

Supporters of municipal broadband services always cite the high cost and low quality of telco or cable TV internet access services as the rationale for building competitive networks. Perhaps one key question is how much room big municipal networks might have to actually cut prices.

So far, virtually all of the independent municipal broadband efforts have taken place in smaller markets where one or both of the incumbent providers appear to be capital-starved, or have chosen not to invest more in access speed, as a deliberate business policy.

One can criticize such strategies, but sometimes, in a declining market, all a business leader can do is harvest returns until some disposition of the assets is made. That was the case for telecom giant AT&T, when it faced a long-term decline in long distance revenues that were the core of its business.

Other smaller telcos have tried cutting some costs (small firms typically have little overhead to trim), launching growth initiatives (outside region business-focused services being most common) or selling assets entirely. Sometimes all three happen, in sequence (cut costs, invest in new lines of business, but then when that does not work, sell the asset).

So one big issue for proposed municipal broadband networks in tier-one U.S. cities is whether enough cost can be cut from the business to make the business model work.

In other words, can a large municipal broadband network operate at costs so much lower than a telco or cable TV company that end user prices are far lower?

So far, it does not appear that major savings, if any, are possible on actual construction of a fiber-to-home network.


A consultant studying the cost of building a municipal broadband network for San Francisco concluded that the actual network would cost $2,050 per passing. That is arguably typical for a large urban network, so it is hard to see any potential end user savings based on lower network cost.

But that is just the cost of the network. To activate a customer, more than that much is required, per customer.


The bottom line is that the proposed San Francisco network would not have lower construction costs than any private operator building a fiber to home network in the city.

The other areas where a cost delta might be obtained are marketing and operating costs. In principle, small ISPs, in selected smaller markets, might be able to operate with lower overhead, lower marketing cost and other fulfillment costs.

In the U.S. market, mobile subscriber acquisition costs might run between $350 and $500 per new account. Some believe a municipal broadband network might spend just $200 per subscriber in marketing costs, for example.

Acquisition costs often are higher in the declining subscription TV business price discounts and promotions frequently are offered.

So if a municipal network in a tier-one city was to offer a price advantage, it would be because its own operating and marketing costs were substantially lower than incurred by telco or cable competitors.

The consultants believe the proposed network would be sustainable at consumer prices of $51 a month (with a range between $26 and $67 a month) and business access costs of about $73 a month (with a range of $38 to $97 a month).

Those figures are roughly in line with what some small ISPs have been offering, but below the level of other municipal networks with less scale. The proposed municipal network for Fort Collins, Colo. projects $70 per month retail fees for gigabit consumer internet access, and $50 a month for access at 50 Mbps.

At least one study suggests such networks do offer lower prices, at least in terms of posted tariffs. It is not clear whether the typical plans chosen by most customers actually mean that the typical buyer in such markets saves money, though that is a logical assumption. There is little incentive for a customer to buy from a municipal provider rather than a commercial provider unless there is a price savings, a value advantage, or both.

Some might argue that another issue is financial risk, in the form of borrowing by entities to build and operate such networks. That might be a substantial issue, for a big network in a major city.

The bottom line is that the business model might hinge on take rates and operating costs, however. History suggests breakeven at about 33 percent take rates, a figure that seems to hold both for municipal and for-profit internet service providers.

One might argue that, in volume, FTTH materials costs have dropped over the last couple of decades, though construction arguably has risen. But even some older estimates have used far-lower materials, construction and connection costs than the proposed San Francisco municipal network consultants have used.

In fact, the cost of the network is higher by two orders of magnitude (100 times), while customer connection costs aer 10 times higher than what some believed was possible in 2003.


The bottom line is that a mature internet access business, with strong competitors (even if some do not believe that), will increasingly be cost-sensitive, as revenue will be challenged.

It is difficult to model precisely what weaker demand does to every ISP business model, but that is a strong possibility, going forward. At the very least, cost containment will be essential. And that, many would argue, is where the risk lies for big municipal networks.

No comments:

Many Winners and Losers from Generative AI

Perhaps there is no contradiction between low historical total factor annual productivity gains and high expected generative artificial inte...