The Economics of Telco Deregulation: Califronia Dreaming, Economic Realities, and the “Reverse Ramsey” Pricing Model

This article in the LA Times on the impact of telco price deregulation in California is a good illustration of the complex nature of the economics of competition and deregulation, and why it’s so friggin’ important for regulators and the public to understand this stuff. In 2006, the California PUC decided that voice service faced sufficient competition to phase out price regulation. In theory, competition would lead to lower costs and increased services and would remove the invariably stultifying impacts of regulation.

The result has been an increase in the availability of services and an overall decrease in the cost of service, but not in the way that ordinary folks understand or that regulators professed to expect from deregulation. Most customers have, in fact, increased the amount they pay for telecommunications services overall. But because they buy larger bundles of services that profess to discount the price of each element in the bundle, the average cost per service is lower although the amount of money paid has gone up. That might seem a good value trade if it were driven strictly by consumer choice. But consumer choice is driven by the decision of telcos to increase the cost of stand alone services. So people not looking to bundle do so because it is “cheaper” while poor people who cannot afford the higher price for the bundle get a real price hike with no value added.

Example: Feldco the Telco raises the price of basic local voice from $10 to $20, and raises the price of additional services taken a la carte from $5 to $10, but I offer a package of basic voice and five additional services for $30 (which I tell you charging $5 for voice and $ 5 for each additional feature). Any customer that can afford to upgrade to my bundled package will do so, because the “value” of the bundle (at my new prices) is $70 and you are getting it for $30. So even though you upgraded and are paying me more, the cost of basic voice (calculated as part of the package) just dropped by $5. What a savings! of course, the customers who cannot afford the additional $10 a month for the bundle experience a real price increase of $10.

Basically, the problem of wealth inequity that we have seen in every other sector of the economy — where the highest earners have enjoyed the greatest increases — is now mirrored in California’s telecommunication service market. How did this happen? Do we care? And what does this tell us about the future of the metered internet, wireless competition, and the ever popular video competition?

Answers below . . . .

We must begin, as just about all regulators, legislators, and members of the public do, with our old high school economic theory. In a competitive market, rival businesses will seek to offer services at the lowest price to attract customers. In addition, they will innovate to capture and keep customers. Therefore, competition provides the best means of increasing social utility. By contrast, regulation inherently distortes market choices because it constrains the behavior of competing firms, can be manipulated by incumbents to the detriment of new entrants, delays the introduction of new services, and adds to overall costs. As a result regulation should be regarded with intense suspicion and used sparingly. Like poisons used in chemotherapy to eliminate a cancerous tumor, regulation should only be used in cases of demonstrated market failure, lest the “cure prove worse than the disease.”

The trouble with this theory is that it has as much to do with actual reality as basic high school physics, which eliminates messy concepts like “friction.” The difference is that legislators don’t tell NASA “when you design and build rockets, assume no friction. You can save a ton of money on heat shields that way.” Markets are extremely complex, increasingly so as “convergence” makes things more interrelated. In addition, we often have policy concerns about fundamental infrastructure and basic services (like telecom) that don’t pop up in the basic model.

With this in mind, we get back to California. Most folks who look at the basic model and the emergence of competition usually fail to appreciate a basic truth of network cost structure: basic services have the least profit margin. The bulk of the cost of the network is in building the physical infrastructure of the network. Layering on additional features, such as caller ID is relatively cheap, and thus has a high profit margin. In addition, as the cable folks discovered when they upgraded to hybrid-fiber-coax, the costs associated with offering different sets of features (such as digital cable and broadband) can be shared between the services, so that the profit margin on the bundle is higher than the profit on the stand alone service.

In addition, operators have other incentives to drive consumers to bundled products. Not only does it increase overall profit, it decreases the likelihood that a consumer will switch to a rival service. The larger and more differentiated the bundle, the harder the bundle is to replicate by a rival, and the more hassle from the consumers perspective from switching (which would require learning new features, reentering information into defaults, and other forms of “switching cost”). There is also what Dilbert creator Scott Adams refers to as the “confusopoly” effect. While the sophisticated user, with much effort, can keep track of the myriad of charges, the average user cannot. The more line items, the easier it is to raise prices on individual components and fees. In addition, the inclusion of additional service for a free trial period that require an opt out (e.g., “three free months of text messaging”) help capture consumers who do not closely monitor their services.

Finally, the more products in the bundle, the easier for the provider to charge a higher price overall — especially if the provider raises the cost of the basic service and of the individual components well above cost. If “basic” phone service costs $50 and “naked” DSL costs $29.99, but the bundle costs a mere $69.99, that appears as a tremendous savings. But the overall effect is to increase what consumers pay, even if the price of the service (when considered as part of the bundle) drops.

The good part of this is that it gives providers lots of incentives to develop new services and differentiated bundles. The bad part is that it also drives providers to raise prices on the basic service and to seek to get rid of “marginal” customers. Translated to English, people still end up paying more on their monthly bill. And, so that telecom service can mirror the general problem of wealth inequity in this country, the high-end consumers (the richest) will receive a lot more value for what they pay on a dollar for dollar basis than the poorest customers.

I like to call this “Reverse Ramsey Pricing.” Ramsey Pricing is a theory that says in the case of monopoly providers, regulators should structure prices to maximize social welfare by letting those who can afford to pay more do so. Thus, re regulate basic service, and expect the monopolist to charge more for any additional services because only those able (and willing) to pay for the additional services will do so.

But as we see in California, deregulation of the telco system creates a Reverse Ramsey effect. The cost of basic service as a stand alone product is driven up because it is in the incentive of the provider to drive as many people as possible to bundles. The cost of individual services in the bundle likewise rises, because the provider has incentive to maximize the size of the bundle.

In theory, competitors should solve this problem by offering lower prices (which they can still do profitably) and thus capturing the additional profits. But they don’t, because no one wants the marginal customers. They are a pain in the ass for very minimal return, and there will never be enough of them to balance the gain in profit from driving non-marginal customers to my bundles. And because all network operators in the market share the same general cost structure, all my rivals will offer similar bundles and pricing structures and will absolutely NOT compete for marginal customers. (I invite folks to check out Joel Waldfogel’s Tyranny of the Market for a more complete explanation of the problem of the customers that no one will ever want to serve, even if one could make a profit by serving them.)

As a result, Verizon and advocates of deregulation speak truth when they say that prices for phone service have dropped (as they do in the article) and that the value and variety of services available to consumers has increased — but only when considered as part of the bundle and taken for the market as a whole (which is being driven to select bundles). But it is also true that most people are paying more for their combined telecom service overall, and that those who are poorest and can only afford the most basic package experience the highest increase in price for the least return on value.

Which creates the conflict between the California Dream of deregulation and the economic reality of who wins and who loses. I think if you ask regulators who voted for deregulation, they will tell you they did not expect that prices for the most marginal customers would rise and that overall customer bills would increase. They understood the promise that prices would drop in the plain English sense of “everyone will pay the phone company less money,” not “everyone will pay more money, but some folks will get a lot more value on a per-dollar basis.”

California regulators could have learned the Reverse Ramsey lesson from a number of relevant markets. We have seen a similar dynamic in cable, where operators have raised cost faster than inflation and increased profit-per-subscriber, and justified this on the grounds that the increase in channels and services means that rates have actually dropped on a per-channel basis. We see the same dynamic in cellular telephony, where prices have risen in the last several years but price-per-minute has fallen when considered as part of an overall plan that includes text messaging and other features that increase the overall bill. And we may see the same dynamic emerge again as we shift broadband access to metered pricing, especially if we permit providers to bundle services.

There are valid arguments to be made in favor of the Reverse Ramsey approach. It does encourage certain kinds of innovation and does increase value for certain customers — especially sophisticated ones able to bargain effectively among multiple providers. But if that’s going to be our policy, then we need to say so in plain English and live with the economic consequences. Regulators and legislators that favor such an approach should stop pretending that they expect prices to go down for poor people, small businesses, and other marginal customers. They won’t. On the other hand, the largest customers and wealthiest consumers will almost certainly see an enormous gain in value on a per-dollar basis, resulting in statistics that demonstrate an overall drop in the cost of basic services.

At least it’s consistent with the economic policies of the last 7 years.

Stay tuned . . . .


  1. Were the FCC smart, oxymoronic I know, they would require of broadband providers a minimum of a 2 level split. One for merely the transport, than all the other ancillary transport on top of it. Yes I can do that today with TW my provider as basic internet and yes the individual price is $15 more than if I consider it as part of a bundle of internet/TV/phone. Just like you say.

    But not exactly what I am after. Even if I request only the basic internet I am paying for a whole level of services on it as well. Like the RoadRunner splash page, WebMail, and all the Britneyesque feeds. None of which I use by the way. In fact based on my ad hoc research, only about 40% of those I know of use what I call the front loaded garbage from the ISP. Fact when I show them the Wiki frontend on my laptop they all want it. Done in an hour, but I digress.

    So why can’t we Americans get ‘transport only pricing?’. It should be an available option at a really attractive price. There are economy of scale factors that also come into play here that you did not touch on in the network infrastructure. But a decent broadband connection, transport only, no support, etc @ $25/mo should be a profitable reality. Its a min-mix problem and ironically like most, having only bundle customers as the only mix does not maximize the profit margin of the overall product suite.

  2. Harold,

    Brilliant, as usual.

    I read an analysis recently about why college costs have gone up so much over the last 20 years or so. If I remember correctly, it was in this article by Andrew Hacker in the New York Review of Books:

    It was his contention that it was all the “extra” consumerist stuff, the counselors, the nutritionists, the fitness trainers, etc, that have driven up the costs–that and the salaries of celebrity faculty — and not the costs of basic instruction, even including expensive subjects like the sciences. Many students (and their parents) would be happy to attend a college without all those consumerist perks, in exchange for a much lower bill. But they’re not available anymore. All colleges are competing for the rich students who can afford to pay the full-boat “tuition and fees”, and those kids are not going to go to a school that does not pamper them. (At least, that’s how I remember the argument, without having the article in front of me.)

Comments are closed