Comments about this article should be sent to mttlr@umich.edu.
{1} Resale is the ability of a firm to purchase
a service on a wholesale basis, for the purpose of reselling that same
service, either alone or in combination with other services or features,
to end users in direct competition with the original service provider.[1]
Resale of local telephone services as a telecommunications policy is currently
among the most important and contentious issues facing regulators, legislators,
and the industry.[2]
{2} The primary reason that regulatory agencies
and legislators pursue resale policies is to foster increased competition
in telecommunications markets where the current level of competition is
considered inadequate to serve the public interest. Unrestricted resale
of the incumbent local exchange carrier's (LEC's) local exchange services,
or the network functions and facilities that make the provision of resold
local exchange service possible, is viewed by policymakers and lawmakers
as an essential component of the transition to effective local exchange
competition. For example, it is an integral component of innovative plans
to foster local exchange competition, such as the Rochester "Open Market"
Plan,[3]
and it figures prominently in the recently enacted Telecommunications Act
of 1996.[4]
Unrestricted resale, it is argued, is necessary to "jump-start" local exchange
competition by allowing competitors to enter the local exchange market
expeditiously and on a ubiquitous basis.
{3} Similarly, it is argued that with a properly
designed resale policy, entry into local exchange markets could take place
with minimal investment, and without the delay of entrants having to deploy
their own facilities or assume the risk associated with facilities-based
entry into these markets. Thus, with a properly designed resale policy,
competitors could quickly establish a customer base at minimal cost, and
then begin selective deployment of facilities in locations where they have
sufficient customers to justify the investment.[5]
{4} The purpose of this Article is to evaluate proposed resale policies from an economic perspective. Specifically, this Article evaluates whether mandated resale can be expected to lead to the benefits ascribed to it by its proponents. In addition, this Article identifies issues which must be addressed before an economically sound local service resale policy may be put into place.
II.
Local Service Resale Issues
{5} Resale as a policy is not automatically in the
public interest, as legislators and regulatory agencies often assume. It
can only be an effective policy if: (1) the competitive process in the
retail market is impeded or forestalled somehow (with a resulting detriment
to consumers via high prices, etc.) due to a lack of alternatives to the
resold service; (2) direct regulation of retail prices is incapable of
remedying the situation; and (3) industry cost conditions make resale conducive
to welfare-improving competition. All three of these conditions are violated
when applied to local exchange resale.
{6} First, if alternatives to a resold service are
generally available, resale is not a necessary component of sound public
policy because there is no "problem" to correct in the first place. However,
even assuming that there are no alternatives available, it is not clear
that the competitive process in local exchange markets is impeded. One
does not observe entry into this market because prices are subsidized;
thus, while there are no alternative providers of local exchange service
for residence customers and many small business customers, they are not
paying overly high prices as a result. The reason that competition does
not develop is due to low subsidized prices, which means that most prospective
entrants to this market cannot expect to earn positive profits post-entry.
Further, even if resale policies increase the number of retail suppliers,
it does not necessarily follow that the competitive process in that retail
market has been enhanced.[6]
{7} Second, it is clear that the direct regulation
of prices in local exchange markets is possible. In fact, the low prices
that regulators set for these services cannot be circumvented by the LECs
and is probably the very reason that entrants are not observed. If more
renumerative and hence more efficient prices were set by regulators, these
would be the prices in evidence in these markets.
{8} Third, preliminary studies indicate that the
costs avoided by LECs by reselling instead of providing end-to-end service
directly to end-users are relatively small. This, combined with the extremely
low price elasticity of demand for local service, means that it is extremely
doubtful that resale policies can lead to significant changes in consumer
welfare. The "Catch-22" behind this characteristic of resale is discussed
in more detail in the next section.
{9} In summary, a regulatory authority should only mandate resale of local exchange services if resale will improve the economic performance of the retail market, and customers in the retail market will be made better off as a result. From an economic perspective, it is extremely doubtful that this scenario will occur. Section II(B), infra, discusses why resale is unlikely to lead to increases in economic efficiency.
B.
The Ability of Resale Policies to Result in Increased Economic Efficiency
{10} Local exchange service has an extremely low
price elasticity of demand.[7]
The lower the price elasticity of demand for local exchange service, the
greater the retail price decreases needed (as enabled by resale) to produce
significant increases in consumer benefits in the form of increased consumer
surplus.[8]
In other words, a large price reduction is necessary just to produce a
modest increase in the demand for local service.
{11} Yet the very nature of resale makes these
price decreases unlikely. Resale allows new entrants to purchase the component
of production of local service that they cannot provide, and requires them
to furnish only the components of production (e.g., retail sales
operations) that are generally available to all firms. For this
latter component of production, no single firm is likely to have a significant
cost advantage. However, if resale is to bring the types of price reductions
that can lead to significant increases in economic efficiency (via increases
in retail market consumer surplus), it is this area in which entrants must
have a significant cost advantage over the LEC, since these are the only
costs resellers can control.
{12} Somewhat of a "Catch-22" situation emerges-to
yield significant increases in consumer surplus, resale must enable competition
that leads to large decreases in local service prices; yet resale requires
entrants to provide (aside from the resold service) only that factor of
production for which no firm is likely to have a significant cost advantage.
Further, the costs avoided by the incumbent LEC, if it cedes the marketing
function to a reseller, are a relatively small component of the total cost
of providing local exchange service. Thus, even if there were significant
innovations (and resulting cost reductions) in this component of production,
the probability of reducing the overall cost to society of local
exchange service significantly is remote. Given this, resale is unlikely
to result in a significant welfare improvement for consumers.
{13} When regulators decide to mandate resale,
a significant dilemma can arise: establishing the optimal wholesale price
for the services that are to be resold. Several largely anecdotal methods
have been proposed for calculating a wholesale price for local service
resale. For example, a "tops-down" approach has been proposed, in which
the embedded costs of the retail elements of "bundled" services would be
estimated, and would then be subtracted from the prevailing bundled price.[9]
Alternatively, a "bottoms-up" approach has been proposed, in which the
LEC would identify the incremental costs of all services or components
offered on a wholesale basis and add a fixed percentage of contribution
to each service/component to establish the wholesale price.
{14} Regarding such wholesale pricing, the overarching
question for regulators is: Which of the proposed wholesale pricing methods
is optimal, i.e., what wholesale/retail price differential
will maximize economic efficiency in the retail market? The issues surrounding
determination of the optimal pricing method for resold services are not
new; they are in large part a variation of the problems addressed in the
economics literature on the pricing of access to so-called "essential facilities."[10]
{15} Economics dictates that the optimal price
for wholesale services purchased for resale is the current retail tariff
rate, minus the avoided incremental cost of retail marketing, plus the
incremental cost of wholesale marketing to the LEC.[11]
Resellers should also pay a fixed charge designed to recover the fixed
and per-firm set-up costs of making resale possible.[12]
This method of pricing the wholesale service is equivalent to the well-known
efficient component-pricing rule developed in the regulation and economics
literature for the pricing of intermediate productive inputs.[13]
The ECPR serves as an economic brightline defining the point at which wholesale
prices are "too high."
{16} If local exchange markets were not the recipients of subsidies used to foster universal service, this wholesale pricing rule could be implemented directly. However, as Section II(D), infra, discusses in more detail, there are additional considerations when retail local exchange markets are subsidized. The pricing rule discussed above makes sense only if existing LEC tariff rates are compensatory retail prices that cover all relevant costs and provide an appropriate contribution to the common costs of the LEC.[14]
{17} Local service rates have traditionally been
held artificially low by regulators to foster universal service. This was
initially achieved through residual pricing of local service,[15]
with long distance and other discretionary services priced relatively high
to help support low local service rates. In addition, local service rates
were geographically averaged over the entire service area to help keep
prices low in high cost areas. There are also many explicit funding mechanisms
used to support universal service objectives.[16]
Regulatory policy has thus resulted in rates for long distance services,
discretionary services, and basic exchange service to businesses in metropolitan
areas that are far above cost in order to hold rates for basic residential
service below cost.[17]
With increased competition, many states adopted price ceilings or outright
price freezes on basic local service rates. Thus, the prices a LEC charges
today for local telephone service do not necessarily reflect market-based
retail prices, and hence cannot serve as the starting point in developing
wholesale prices.
{18} Since retail prices in residential markets
have been set at below-cost levels to pursue universal service objectives,
it begs the question of why a regulatory agency would want to pursue resale
in the first place. The overarching economic reason for resale is to enhance
or enable the competitive process in the retail market, if the market
failure in the retail market is caused by the lack of availability of the
service to be resold (and if mandating resale can lead to increases in
economic efficiency). Resale makes sense as a policy only if prices in
the retail market are too high (due to the possession of market
power by the incumbent firm), if the competition that would result from
resale would curb that market power, and if the direct regulation of retail
prices is considered an ineffective means of correcting this problem.
{19} A market in which the retail price has been
set at levels below cost is not a market in which
prices are too high due to market failure. If retail prices have already
been set at levels below cost, then the direct regulation of downstream
prices is a presumptively effective policy tool, and the reason
that competition from alternate suppliers does not take place is that prices
have been set so low in the first place, with regulatory sanction. At the
subsidized retail price, entry is unlikely to occur anyway, and it is inappropriate
to apply resale policies.
{20} Entry into the subsidized residential and small business markets would not occur on the basis of profits from these markets alone. Such entry may be economically attractive only if it gives providers (the incumbent LEC and resellers alike) the first opportunity to obtain the business that is priced far above cost, such as vertical features and long distance services. In a "one-stop shopping" telecommunications environment, customers are likely to purchase most, if not all, of their telecommunications services from the same provider to which they subscribe for their basic local service. By requesting a discounted wholesale rate in a market that is already subsidized, prospective resellers are essentially demanding an equal opportunity at servicing the overpriced markets-where they could undercut the LEC's artificially inflated prices that help support below-cost local service-without having to bear any of the costs that justify that overpricing.[18] A wholesale rate which is below the actual cost of service will ultimately require the LEC's customers or stockholders to subsidize the reseller's customers. The remedy is obvious: prior to setting the wholesale price for resold local exchange services, it would be in the public interest for regulators to set the price of downstream retail services at an appropriate level exceeding cost (taking carrier-of-last-resort responsibility and related factors into consideration). Regulators could then observe that market to see if government intervention is necessary to result in a more efficient industry structure.
E.
Resale and Antitrust Concerns
{21} It is sometimes argued that the failure of
a LEC to offer services for resale, at prices allowing the reseller to
earn positive profits, is anticompetitive. However, whether the lack of
availability of services for resale is truly anticompetitive hinges on
one question: is access to services via resale an "essential facility"
in the antitrust sense? If the answer to this question is yes, then there
may be a case for mandated resale policies. However, qualitative economic
analysis indicates the most likely answer to this question is "no."
{22} Resale proposals implicitly treat LEC local
exchange services as de facto "essential facilities," in the antitrust
sense. That is, they assume that for true competition to take place, a
regulatory agency must mandate open access to the LEC's local exchange
services via resale. However, in economic terms, whether a so-called "essential
facility" exists in a wholesale telecommunications market depends on its
effect on the competitive process in the adjacent retail markets. The "essentiality"
of resold services under the economic efficiency criterion requires at
least four necessary (though not sufficient) conditions to hold
true: (1) the absolute requirement that an entrant have physical access
to the "essential" resold service to provide service at all; (2) a welfare-enhancing
competitive process could not operate properly in the retail market unless
efficient entrants have access to the resold services; (3) the "essential"
resold service is available only from a monopolist or consortium of firms
acting as a monopolist, and no other source; and (4) prospective
entrants can earn positive profits post-entry when paying the efficient
wholesale price for resold services, as discussed above. In economic terms,
the essentiality of resold services requires the denial of resale to result
in a von Weizsäcker entry barrier to the downstream retail market.[19]
{23} Under the criterion of economic efficiency,
the failure of a vertically integrated firm to make services available
via resale to firms requiring them (as inputs to their retail service)
is not a prima facie indicator that such services, as productive
inputs, are "essential." Further, under the efficiency criterion, the fact
that prospective entrants to the retail market can increase their profits
if resale policies occur is immaterial, since the existence of such entrants
may or may not enhance economic efficiency. Nor is it necessary to expect
the vertically integrated firm to increase its costs by engaging in resale
just to make a retail service offering possible by its downstream rivals
in the retail market.
{24} At this point in the resale debate, it is
clear that a lack of availability of resold services is not anticompetitive,
for several of the economic conditions required for essentiality are violated
in current telecommunications markets.
{25} First, there is the absolute requirement that
an entrant have physical access to the "essential" resold service to provide
service at all. Though this may be true of some prospective entrants to
the local exchange markets, it is not true for all of them; some prospective
entrants have the ability to provide facilities-based service. As Areeda
and Hovenkamp have argued, a resource is not essential if competitors can
operate effectively without it. For a resource to be essential, it must
be not just helpful, but vital to competitive survival.[20]
It is important to note that in economic terms and in the eyes of the courts,
an alternative to a productive input (e.g., in the manner that facilities-based
service is an alternative to resold services) is not necessarily infeasible
simply because it is more expensive.[21]
Perhaps more importantly, the fact that access to a facility (e.g.,
via resale) is merely more economical than other alternatives is not sufficient,
in economic terms, to demonstrate that a given facility is "essential;"[22]
nor is the fact that with resale, a competitor could achieve savings (and
hence increased profits) at the expense of the vertically integrated firm.[23]
{26} Second, there is the condition that a welfare-enhancing
competitive process could not operate properly in the retail market unless
efficient entrants have access to the resold services. This condition is
wholly inapplicable to local exchange markets, for they are subsidized
markets. It is not yet possible to know if efficient competition is foreclosed
in local exchange markets. This cannot be known until compensatory rates
are set for local service. Once this is done, it would then be possible
to observe the local exchange market and see if further government intervention
is needed to foster efficient competition. Until that time, it is not possible
to argue that competition is foreclosed in retail local exchange markets.
It simply doesn't exist yet because universal service policies have required
local service prices to be below cost, which makes it less likely that
entry will be observed in this market. Competition is not foreclosed due
to a lack of resold services-it simply doesn't exist yet in most markets
because prospective entrants cannot earn positive post-entry profits, due
to the low prices that have been set to meet universal service objectives.
{27} Third is the condition that the "essential"
resold service is available only from a monopolist or consortium of firms
acting as a monopolist, and no other source. The fact that facilities-based
competition is possible violates this condition.
{28} Fourth, if resold services are truly essential,
it must be true that prospective entrants can earn positive post-entry
profits when paying the efficient wholesale price for resold services,
as discussed in the previous section. This condition is necessary to ensure
that prospective entrants are capable of engaging in welfare-increasing
competition with the incumbent firm if all other impediments to entry are
relaxed. The proper wholesale price is extremely important in determining
if a given resale policy (or the lack of one) is anticompetitive.
{29} The efficient wholesale price of resold services
discussed above is an application of the efficient component-pricing rule
(ECPR). Prospective entrants into local exchange markets may argue that
overly "high" wholesale prices for resold services are a de facto
denial of resale. The ECPR, however, serves as an economic brightline defining
the point at which wholesale prices are "too high." It serves as a screen
against inefficient or opportunistic entrants seeking to gain entry to
a market at rates subsidized by the incumbent firm's stockholders. As Baumol
and Sidak have stated, "the [ECPR] offers the prospect of success to entrants
who can add efficiency to the supply of the final product, while it ensures
that inefficient entrants are not made profitable by an implicit cross-subsidy
extracted from the incumbent [firm]."[24]
If (but for the unavailability of resold services) retail local exchange
markets are not competitive, then prospective entrants capable of fostering
the competitive process in those markets should be capable of paying the
ECPR-based wholesale rate; if not, then a welfare-improving competitive
process has not been foreclosed in the retail market, and neither a denial
of resale nor a wholesale price for resold services at the ECPR level (or
less) are anticompetitive. Prospective entrants that cannot pay an ECPR-based
wholesale price, and earn positive post-entry profits, are irrelevant to
the competitive process. It is not anticompetitive if the lack of a resale
policy prevents the entry of such firms; nor is it anticompetitive if such
prospective entrants fail to earn positive post-entry profits at the ECPR-based
wholesale price.
{30} One way in which resale policies would, in fact, be anticompetitive is if the wholesale price of resold service is based on an arbitrary fixed percentage of the incumbent firm's retail price. This would needlessly eliminate the incumbent itself as a potentially efficient competitor. If the wholesale price is tied to the incumbent LEC's retail price, the incumbent LEC can never compete on the merits of its retail pricing; each time it lowers its retail price, its competitors' input prices also are lowered. Each time the incumbent firm attempts to compete on price, it faces competitors whose input prices have decreased by a proportion of its own retail price decrease. If the wholesale price is tied to the incumbent's retail price, the incumbent firm will eventually be forced to exit the retail market (if legally allowed to do so), and become only a wholesaler, ceding the sale of the retail service to its competitors. If the incumbent LEC cannot exit the market (due to carrier-of-last-resort responsibility or other legal reasons), it will needlessly run deficits, and its stockholders would be subsidizing the entry of resellers.
{32} More specifically, Fisher concluded that if
there are constant returns to scale in the marketing component of production,
"then the fully integrated case and the competitive dealers case both yield
the same same outputs, the same retail prices, and the same profits to
the manufacturer."[26]
If, however, there are decreasing returns to scale in marketing, the competitive
dealers case yields lower profits to the manufacturer and results
in a lower output and a higher retail price than does the
fully integrated case.[27]
Thus, Fisher concluded that in choosing between the fully integrated case
and the case of competitive dealers, the interests of the incumbent manufacturer
coincide with those of society.[28]
The relevance of this conclusion to resale policies in telecommunications
is that, within the assumptions of the Fisher model, there is no compelling
reason to mandate resale policies.
{33} Fisher's analysis also compares the case of competitive dealers with that of a single monopoly dealer, and concludes that the competitive case always results in a greater output and a lower retail price than does the case of a monopoly dealer.[29] A failure to distinguish properly between the integrated supplier case and the monopoly dealer case may be the root of confusion about the benefits of resale policies. The case of the telecommunications industry is not that of a monopoly dealer versus competitive dealers-it is a case of an integrated supplier (i.e., one that de facto has integrated forward from the manufacturing function into the marketing function) versus manufacture by one firm (with distribution and marketing handled by competitive dealers). In the latter situation (but not the former), the interests of a monopoly manufacturer coincide with those of society.
{34} In general, facilities-based competition is
likely to be more efficient than competition that is based solely on resale
policies. Though resale may spur innovation in the marketing of
local exchange services, it will not do so for the production of
the local exchange services themselves, as facilities-based competition
will. As this Article has already pointed out, the costs to market
local exchange services are relatively small in comparison with the corresponding
costs to produce it. Resale provides incentives to lower a very
small component of the costs of local exchange service, but unlike facilities-based
competition, it does not provide incentives to engage in cost-reducing
innovation where the reductions in costs are likely to be greatest: in
the production of the service itself.
{35} A poorly designed resale policy can lead to
inefficient incentives for both prospective entrants and incumbent LECs,
and in so doing, retard the growth of efficient facilities-based competition.
In this regard, two problems with resale immediately unfold.
{36} First, any prospective entrant knows that
if there are network components it needs, it can attempt to gain access
to them via resale policies in lieu of engaging in its own investment.
Entrants thus have reduced incentives to develop vertically integrated
production processes leading to the completion of a final product for sale
to end users. A poorly designed resale policy will give entrants the incentive
to engage only in the stages of production in which they can excel, but
not the incentive to innovate and develop all stages of production required
for completion of the final product.
{37} Second, LEC competitors may seek to gain access
to valuable network components at resale prices that do not reflect the
true social costs of the access. Such entrants may be able to enter the
market only if they are allowed access to these network components at advantageous
resale rates and terms (in lieu of engaging in their own investment). Unfortunately,
this is not what true competition is about, for entry that occurs in this
way does little or nothing to yield the benefits to consumers of facilities-based
competition.
{38} A regulatory authority may pursue a mandated
resale policy because it believes that by doing so, it can simulate the
results of a so-called "contestable" market downstream by removing what
it considers "barriers to entry" to the downstream retail market (i.e.,
by removing sunk costs of producing the retail service). Thus, the intent
of creating a "wholesale" local exchange service through resale policies
may be to reduce the initial start-up costs of entering the local exchange
market, and to circumvent the large investments and associated risk required
to build competing local distribution networks. If competition is based
on resale policies in lieu of facilities-based competition, LECs would
assume all the expense and risk of putting plant into the ground and of
maintaining and upgrading it, while resellers could use this plant at a
low price and walk away from it without losses if adequate consumer demand
did not develop. Facilities-based competition for local exchange telecommunications
services is far more likely to bring about the benefits of competition
compared to "competition" based on resale. The prospect of competition
has resulted in research, development, mergers, joint ventures, partnerships,
and the trialing of innovative new technologies and new applications by
diverse (and often non-traditional) companies, all in efforts to become
viable and successful competitors in the emerging competitive telecommunications
environment.
{39} Thus, resale may seriously reduce incentives
for LECs and other firms to engage in cost-reducing innovation or network
modernization in the future, since the LECs must expect that they may be
required to make components of their innovations available to competitors
on terms that may not allow recovery of the investment. This type of "competition"
cannot bring about real economic efficiencies and the type of competition
that lowers total industry costs. Because resale cannot spur true local
exchange competition (in the economist's sense), "competition" from resellers
will have no effect on the incumbent LEC's incentive to increase quality
or lower production costs. Since the non-facilities based reseller must
purchase its primary productive input from the existing LEC, any quality
improvement by the LEC will also be granted to the reseller. No competitive
advantage would be granted to the LEC by improving the quality of its service.
Similarly, the existing LEC will have no increased incentive to lower the
cost of its service. Any cost savings achieved by the LEC would also be
granted to the non-facilities based competitor through lower wholesale
rates. The incumbent carrier would not realize any competitive advantage
from lowering its costs.
{40} Thus, resale provides misguided incentives to both incumbent LECs and to prospective entrants: it rewards innovative marketing (a small component of the total cost of local service), but unlike facilities-based competition, it offers prospective entrants no incentive to engage in cost-reducing innovation.
{42} This Article arrives at the following conclusions:
(1) Resale as a policy is not automatically in the public interest, as legislators and regulatory agencies often assume. It can only be an effective policy if: (a) the competitive process in the retail market is impeded or forestalled somehow (with a resulting detriment to consumers via high prices, etc.) due to a lack of alternatives to the resold service; (b) direct regulation of retail prices is incapable of remedying the situation; and (c) industry cost conditions make resale conducive to welfare-improving competition.
(2) Resale is unlikely to result in a significant welfare improvement for consumers. For resale to yield significant increases in consumer surplus, it must enable competition that leads to large decreases in local service prices. Yet resale requires entrants to provide (aside from the resold service) only that factor of production for which no firm is likely to have a significant cost advantage. Further, if the costs avoided by the incumbent LEC (in ceding the marketing function to a reseller) are a very small component of the total cost of providing local exchange service, then even significant innovations (and resulting cost reductions) in this component of production has a low probability of reducing the overall cost to society of local exchange service significantly.
(3) The optimal price for wholesale services purchased for resale is the current retail tariff rate, minus the avoided incremental cost of retail marketing, plus the incremental cost of wholesale marketing to the LEC. Resellers should also pay a fixed charge designed to recover the fixed and per-firm set-up costs of making resale possible. This method of pricing the wholesale service is equivalent to the well-known efficient component-pricing rule developed in the regulation and economics literature for the pricing of intermediate productive inputs.
(4) Resale in subsidized markets is inappropriate. A market in which the retail price has been set at levels below cost is not a market in which prices are too high due to market failure, and hence which may require a resale policy.
(5) The failure of a vertically integrated firm to make services available via resale to firms requiring them (as inputs to their retail service) is not a prima facie indicator that such services, as productive inputs, are "essential facilities." Further, the fact that prospective entrants to the retail market can increase their profits if resale policies occur is immaterial, since the existence of such entrants may or may not enhance economic efficiency.
(6) Facilities-based local exchange competition is more likely to lead to efficient retail pricing than the competition spawned by resale policies. A poorly designed resale policy can lead to inefficient incentives for both prospective entrants and incumbent LECs, and in so doing, retard the growth of efficient facilities-based competition.