By Michael H. Lee and William A. RosquistThe Vanishing LATA:
Pricing Chasms
and Clashing Markets
for Toll Service
Wide disparities can occur in toll rates within some states (em a legacy of multiple LATAs. Now, with
barriers falling, where will prices go?
When local access and transport areas (LATAs) were introduced a decade ago,1 separate geographic markets emerged for intra- and interLATA long-distance (toll) service. Montana, like many other states, has more than one LATA. Now, Congress has enacted legislation that allows regional Bell operating companies (RBOCs) to enter the interLATA market.2 The two LATA markets will converge; toll rates will change. A new long-distance price equilibrium will emerge, possibly at an unexpectedly low level.
How competitors in this new market will react to the presence of other competitors is as yet unknown. However, their revenues will be affected by converging toll-service rates. Market shares for toll and carrier access services will also be affected.
The Problem: Unsustainable Rate Disparities
The interLATA toll rates are similar for the major Montana interexchange carriers (IXCs) (em AT&T, Sprint, and MCI. However, these toll rates differ from the intraLATA toll rates charged by U S WEST Communications, Inc., sometimes significantly. Table 1 shows the disparities in residential message toll service (MTS) rates between U S WEST and AT&T. Significant differences exist in certain night/ weekend MTS mileage bands, such as the 11 to 16 mile band, where AT&T's rate exceeds U S WEST's by 233 percent.
Such rate disparities probably will not survive once the intra- and interLATA markets become one. Unrestricted competition between suppliers of similar commodities will cause currently disparate prices to converge. A competitive market, or rivalry in noncompetitive markets, cannot sustain the rate disparities we now see.
If the market cannot sustain these disparities, then where will competitive prices settle? Will AT&T's higher rates fall, or will U S WEST see an increase in its
relatively low prices? Unless regulation maintains the rate disparities, a new equilibrium will evolve. The factors that will determine the new price equilibrium include interactions between firms, the market power or dominance of any carrier(s), technological change, scale and scope economies, tolerance for complex rates, customer loyalty, marketing strategies, and regulatory policies. These factors will determine whether the existing time-differentiated and mileage-sensitive rate designs will give way to either a simplified
single rate, or a myriad of customized opaque tariffs.3
In the meantime, U S WEST and AT&T face unequal opportunities. While it is technically possible for AT&T to compete in the intraLATA toll market, U S WEST enjoys a 1+ advantage. And, while currently prohibited from entering the in-region interLATA market, U S WEST can enter and compete in out-of-region markets. The Telecommunications Act of 1996 should create economic parity within three years, however. Once free of LATA boundary constraints, U S WEST, AT&T, and the other IXCs will compete in both local and toll markets to provide end-to-end service.
Simple Case: One Carrier Matches the Other's Rate
As described in more detail in the sidebar ("Model and Parameters"), Table 2 shows what we found when we studied the effects on U S WEST and AT&T if the residential toll disparity disappeared, and one or the other carrier's toll prices dictated the market and determined the new price equilibrium, without accounting for elasticities or other feedback effects.
The benchmark case displayed in Table 2 estimates the total residential toll revenues collected by U S WEST and AT&T, respectively, in 1995. The combined benchmark revenues of $28 million are assumed to include intrastate residential MTS sales in all mileage bands for three time periods (day, evening, and night/weekend).
In this example, if prices at U S WEST increased to equal AT&T's and nothing else changed, the figure for combined revenues would rise to $34 million. Conversely, if AT&T lowered its prices to the levels for U S WEST and nothing else changed, the combined annual revenue figure would decline to $25 million
These results permit one to estimate the hypothetical cost to U S WEST of implementing 1+ intraLATA equal access. If AT&T captured the entire U S WEST residential market, U S WEST could lose $17 million (leaving aside any offsetting increases in carrier access-charge revenues that would substantially diminish this revenue loss).
Thus, we could conclude that total revenues do not vary significantly in relation to price changes. If AT&T's prices declined to equal those for U S WEST, AT&T would lose about $3 million in revenues, and U S WEST would enjoy no benefits. However, because these conclusions do not reflect market realities, we must include demand elasticities and carrier access charge (CAC) impacts in our analysis.
Complex Case: Consider Elasticities
To inject market realities (again, see sidebar, "Model and Parameters"), we conducted a second-stage analysis to consider how price elasticities and other "feedback" effects might influence the eventual equilibrium price and carrier revenues if U S WEST and AT&T modified prices in the residential toll market to eliminate price disparities. We show the results in Table 3.
Our second-stage analysis uses two alternative own-price elasticities of demand: -0.5 (low) and -2.0 (high). These two values likely define the boundaries of the unknown actual demand elasticity.4 Given the looming market changes, we find the often-used demand elasticity range of -.5 to -1.0 on the low side.
Elasticities take many forms, two of which are "own" and "cross." These two elasticities are easily understood in the context of energy markets; they are less intuitive with telecommunications.
For example, if the price of natural gas falls by 50 percent, one normally expects consumption of natural gas to increase as customers luxuriate in warmer homes. This behavioral response is captured in own-price elasticities: As the absolute price of gas declines, the demand for gas increases. Yet other customers may switch from electric space heat to natural gas. This behavioral response is captured in cross-price elasticities: As the price of gas declines relative to electric prices, the demand for natural gas increases as it is substituted for electricity.
While electricity and natural gas can be considered substitute goods, intra- and interLATA toll services may display both substitute and complementary characteristics as the markets merge. If U S WEST's prices remain attractive relative to AT&T's, customers will substitute MTS service from U S WEST for AT&T's service. And if AT&T's price level declines to match the U S WEST price, then customers will also increase their demand for what was strictly intraLATA toll service by U S WEST. Methods of modeling these effects are poorly developed; therefore, we do not model the demand impacts on one carrier due to another's price changes.
Table 3 compares the low- and high-demand elasticity change cases to the base case. In the base case, AT&T's average rate (18.7¢) exceeds the U S WEST rate (13.5¢) by about 40 percent. In both change cases, we lower AT&T's prices to those of U S WEST and, in turn, estimate the change in AT&T's toll revenues. Whether AT&T's revenues rise or fall depends on the own-price elasticity of demand. Since AT&T's increased toll demand produced higher demand for carrier access, U S WEST will benefit from AT&T's price decline regardless of the demand elasticity, because of higher CAC revenues.5
(Of course, one can question these results, given that cross-price elasticities are ignored. One could also argue that AT&T's price declines would make AT&T more attractive and entice some
customers to leave U S WEST. Since we do not permit AT&T's prices to decline below those for U S WEST, and assume away creative marketing instruments, customers would only shift to AT&T if they perceived AT&T's product as a better deal. We also assume that customers are informed (em a stretch of imagination in this industry.)
We lowered AT&T's relatively high MTS toll prices down to the price levels for U S WEST for several reasons. First, U S WEST has asserted (rightly or wrongly) that its residential access rates are cross-subsidized by toll rates. In other words, assuming that U S WEST remains price regulated, eliminating the subsidy would require U S WEST to reduce its toll rates and increase its residential local rates. We assume that U S WEST would not exacerbate the alleged cross-subsidy by now increasing toll rates (unless, of course, it has market power and pays to do so). That is, U S WEST should not seek to increase MTS prices, given its past cross-subsidy allegations.
Toward a New Equilibrium
Pressure to lower toll prices will emerge from interstate price competition. One such source: Sprint's simple and understandable "Dime-a-Minute" rate. When the LATA boundary lifts, Sprint's rate will exert a magnetic effect. Sprint's 10¢ rate is about 35 percent below the average MTS rate for U S WEST, and even though this rate does not apply to daytime calling, we suspect its effect will eventually apply to all time periods.
Further, we suspect that technological change may eventually render a 10¢ rate conservative. Although in its infancy, technology permits the use of the Internet for voice communication, and Internet access can be purchased at rates as low as 3¢ per minute.
Finally, other opportunities may present themselves via creatively packaged services. A customer whose separate intra- and interLATA use did not qualify for price discounts from either carrier may benefit from aggregating intrastate toll demand and, in turn, use a single IXC. Residential customers with home computers will likely be offered package deals tailored to their combined voice and data consumption. These, however, are not purely empirical factors; they depend on pricing strategies.
The results in Table 3 suggest that AT&T faces asymmetric risks and rewards from reduced toll price. When we reduce AT&T's average price level to that of U S WEST, revenue impacts depend on the assumed demand elasticity. With a demand elasticity of -0.5, AT&T's toll revenues decline by about $1.5 million annually. If demand elasticity is -2.0, AT&T stands to gross about $4 million annually.
In contrast to AT&T, U S WEST stands to gain increased CAC revenues, regardless of demand elasticity, when AT&T's prices decline. A low demand elasticity would garner $1 million annually for U S WEST; a higher demand elasticity of -2.0 would collect an additional $5 million per year.
As suggested earlier, we believe the ultimate equilibrium price level will be, minimally, in the 10¢ range. That is, the above analysis, if flawed, assumes a conservatively high price equilibrium. And although the rate at which equilibrium will arrive as well as its
duration are unpredictable, we doubt that IXCs will sit on their laurels for long.
Policy Implications
The policy implications appear most significant for those states with substantial toll-rate disparities. Since we had no evidence of the cost of providing toll services, other than access charges, different toll cost functions will significantly alter our findings.
Suppose AT&T were to assert that its toll rates are cost based and that it would suffer financial loss from lower toll prices? How, then, could U S WEST further reduce its already lower toll rates without a subsidy? If the two firms have different cost functions, how can the firm with higher costs remain competitive when the two markets merge? If the firms have the same cost functions, and if U S WEST can cut its toll rates more, yet hold prices above cost, are interLATA toll markets really competitive? What keeps AT&T from further reducing residential MTS rates? Could interLATA toll rates be subsidizing interstate rates?
Our estimated impacts (see Table 3) are consistent with prior testimony and assertions made by U S WEST and AT&T. The strategic motives of LECs and IXCs to affect the direction of change for toll and local rates also clearly support past testimony. U S WEST consistently asserts that its toll rates subsidize local (residential) rates; AT&T consistently counters that assertion. Both positions are predictable and strategic.
Our analysis also reveals the degree to which revenues remain sensitive to the assumed decline in toll rates. The impacts, however, are asymmetric; they depend on the assumed elasticity of demand. U S WEST benefits from AT&T's declining prices regardless of demand elasticity, at least down to its own average price level. This outcome reflects increased carrier access sales to AT&T, combined with our assumption that AT&T's prices merely fall to a par with the average price level for U S WEST. Reducing prices below the average U S WEST price level may produce a different conclusion.
In contrast, lower toll rates are anathema to AT&T. Whether AT&T benefits from declining toll prices depends critically on the assumed elasticity of demand. Low demand elasticity (-0.5) causes AT&T's revenues to decline. Conversely, high demand elasticity (-2.0) lifts AT&T toll revenues. Thus, AT&T's argument that local rates cross-subsidize toll rates comes as no surprise. These results are consistent with economic axioms.
However, for U S WEST we must consider impacts before as well as after the 1996 Telecommunications Act. Before the Act, U S WEST would have stood at risk, since there was no interLATA entry quid pro quo for RBOCs when IXCs received 1+ intraLATA dialing parity. Nevertheless, U S WEST would see an increase in net revenues in the lower mileage bands if it lost sales to competitors (the 8.8¢ CAC rate collected by U S WEST exceeds its toll rates at lower mileage bands).
Now that U S WEST's entry into the interLATA market (in-region) is linked under the Act with 1+ intraLATA dialing parity, a price-adjustment period will ensue. Both U S WEST and the IXCs offer volume discounts to customers with relatively high usage. Similarly, customers with relatively small intraLATA and interLATA usage, purchased separately from U S WEST and AT&T, may exhibit a combined usage that now qualifies for a discount plan. Whether, and to what extent, the merged markets will benefit customers with inconsequential aggregate toll demand remains unclear.
Finally, our analysis reinforces the importance of cost-based pricing for U S WEST's toll and local rate. Although night/weekend, MTS, additional-minute rates for U S WEST do not exceed the CAC, its aggregate MTS revenues do exceed the aggregate of CACs. However, if rate rebalancing leads to lower CACs, reductions to some or all toll rates may follow.
We suggest two questions for further study. First, as discussed above, do RBOCs and IXCs face fundamentally different toll-cost functions? Second, and beyond our resource capability, how will the pending merger of intra- and interLATA markets impact demand elasticities? t
Michael Lee is bureau chief of rate design and economics at the Montana Public Service Commission. He has an MA in economics from the University of Montana. William Rosquist is a utility rate analyst with the PSC. The opinions expressed in this paper are those of the authors and do not necessarily represent the views of the Montana PSC or any individual Commissioner.
1. With the AT&T breakup, the divested Bell carriers could serve only "within" a LATA (em not "across" a LATA boundary. Remaining traffic, the so-called "interLATA" market, fell into the domain of AT&T, MCI, and other interexchange carriers.
2. See, section 271, the Telecommunications Act of 1996. To gain entry to the interLATA market prior to 1999, U S WEST must offer equal access in the intraLATA market.
3. Rivalrous (not effective) competition, if allowed, could lie veiled in repackaged toll services.
4. Dr. Lester Taylor suggests that price elasticities range from -.4 to greater than -1.0. See, Taylor Telecommunications Demand in Theory and Practice, Kluwer Academic Press, 1994.
5. The assumed carrier access cost is 8.8 cents/BMOU (as per memo from staff to Montana PSC, April 23, 1996).
Articles found on this page are available to Internet subscribers only. For more information about obtaining a username and password, please call our Customer Service Department at 1-800-368-5001.