More on “Unbundling”


The precise definition of “unbundling” is often unclear. Some statutes identify it as “separately identifying and charging” for different services. Other textbooks define it as “making services separately available”. Understanding the concept is crucial to the process of figuring out how to de-integrate a vertically integrated utility:  “Unbundling” competitive functions from noncompetitive functions.

The term is often used without reference to the market structure in which unbundling will occur.  Some market structures allow the incumbent, once it unbundles its monopoly services from its competitive services, to remain in both markets.  In other markets, policymakers have required the utility to sell off its businesses in the now-competitive market and cease providing unbundled services.  Either approach only begins the process of ensuring evenhanded competition post-unbundling, because immature markets are prone to early concentration.  Further, unbundling itself must be accompanied by (a) nondiscriminatory access for competitors to the monopoly service; (b) “cost-unbundling” to ensure that the incumbent’s competitive service costs are not recovered in the monopoly service rates; and (c) enforcement of an arms-length relationship between the incumbent’s competitive and noncompetitive services (if the regulator has allowed the incumbent to engage in both activities).

Unbundling must be accompanied by monitoring.  For if unbundling does not produce a competitive market, the regulator needs to find out why. It could be an economies of scale problem (the services are provided more efficiently as a package), an entry barrier problem (new entrants have a cost disadvantage relative to the incumbent), or a customer inertia problem (customers don’t bother changing suppliers unless there’s a clear advantage to the effort).  Sometimes benefits are possible but long in coming, requiring patience by customers and regulators.  The question is whether policymakers (and consumers) should accept the short-term inconveniences and costs (a loss of static efficiency) to gain a long-term increase in dynamic efficiency (and possibly in static efficiency).


Eric Witte

The central concept for understanding unbundling is “externality”: There are advantages to structuring markets to give people choices – provided the people making the choices bear nearly all the costs of their choices. But if my choice will end up imposing uncompensated burdens on you, then arguably the regulators shouldn’t give me that choice. To anticipate whether my choice will impose an uncompensated burden on you does not depend on philosophy; it depends on Industrial Organization – a deep knowledge of the market in question. Thus, beware of people who profess to like or dislike unbundling in the abstract; the devils and the angels are in the details – it’s all about the details.

No doubt, centralized, command-and-control, you-can-have-any-color-you-want-so-long-as-it’s-black regulation is simpler. Calculating average costs rather than marginal costs is simpler. But it sacrifices efficiency and autonomy. Yet on the other hand, the complexity of telecommunications cost models and interconnection agreements, or regional transmission organization tariffs and charge lists, testify to the challenge of unbundling. They reflect the effort to identify an element that we can permit individuals to choose or not, regardless of the choices of other individuals, and the effort to identify an appropriate price for that element. Define the element badly, or set the price wrong, and the advantages of unbundling are lost. Again, it’s all about the details.

Finally, “unbundling,” just like “deregulation” and “deaveraging,” reflects a direction on a continuum, not a destination. A municipality may say it is “unbundling” bills when its bills begin distinguishing between property taxes and charges for municipal electric service. Or when it begins charging customers different amounts depending on how much energy a customer consumes. Or when it starts distinguishing among charges for demand, energy consumption, and customer accounts. Or when it begins distinguishing between firm service and interruptible service. Or when it begins distinguishing between energy consumed on- and off-peak. Or when it begins distinguishing between energy from renewable sources and energy from non-renewable sources. Arguably each of these changes represents a form of unbundling relative to some prior form of regulation. And arguably none of these represents a “fully unbundled” system. Conceptually, unbundling is limited only by our powers to identify useful distinctions – and that power may be limitless.

Post new comment