Testing for competitiveness of markets for long distance telephone services: competition finally?

Review of Industrial Organization

Volumn 13, Issue 3, 1998, Pages 295-319

  MacAvoy, Paul W ^a  

^a YALE SCHOOL OF MANAGEMENT   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

COMPETITIVENESS; MARKET CONDITIONS; TELECOMMUNICATIONS;

USA;

EID: 0031682295     PISSN: 0889938X     EISSN: None     Source Type: Journal    
DOI: 10.1023/A:1007748417477     Document Type: Article

References (45)

1. The U.S. Department of Justice/Federal Trade Commission Horizontal Merger Guidelines (1992) concur with this conclusion, stating that HHIs above 0.18 represent highly concentrated markets.
2. See Abba Lerner, The Concept of Monopoly and the Measurement of Monopoly Power, Review of Economic Studies, 1 (June 1934), pp. 157-175;
3. Keith Cowling and Michael Waterson, Price Cost Margins and Market Structure, Economica (August 1976), pp. 277-287;
4. William Landes and Richard Posner, Market Power in Antitrust Cases, Harvard Law Review, 95 (August 1982), pp. 1857-1874;
5. Dennis Carlton and Jeffrey Perloff (1994), Modern Industrial Organization (New York: Harper Collins College Publishers, 2nd. ed.), pp. 352-354 and 360-366. Carlton and Perloff cite numerous empirical studies that have used the price-cost margin to measure monopoly power, the dual of "competitiveness".
6. A corollary is that: markets with more reduction in concentration should have lower price-cost margins, all else equal, since they should be more competitive than most.
7. i is the firm's market share. Summing this expression over all firms yields the industry-average price-cost margin as shown in the text.
8. Steven C. Salop (1986), Practices That Credibly Facilitate Collusion, in Joseph Stiglitz and G. Frank Mathewson, eds., New Developments in the Analysis of Market Structure, Cambridge, MA: The MIT Press.
9. In October 1996, however, the FCC in its Second Report and Order under the Telecom Act of 1996 eliminated tariff filing requirements for all interexchange carriers. This has been stayed by Court proceedings brought by a number of carriers objecting to the non-filing requirement. But the elimination of tariff-filing requirements would be a necessary but not sufficient condition for the breakdown of tacit collusion. It remains to be seen whether or not the FCC=S decision will have an effect on competitiveness.
10. David S. Sibley and Simon J. Wilke, A Repeated Game of Price Cap Regulation (University of Texas Working Paper, revised January 1996).
11. Id.
12. Shapiro, Carl (1989), Theories of Oligopoly Behavior, in Richard Schmalensee and Robert Willig (eds.), Handbook of Industrial Organization, Amsterdam: North Holland, pp. 329-414.
13. InterLATA, interstate prices were estimated by the application of software developed with HTL Telemanagement, Ltd. taking the assumed calling patterns and applying them to data files on tariffs that AT&T, MCI, and Sprint maintain at the FCC.
14. B. Douglas Bernheim and Robert A. Willig (October 25, 1996), The Scope of Competition in Telecommunications, working paper prepared for the American Enterprise Institute.
15. AT&T Long-Distance Traffic Volume Up (January 23, 1997), Dow Jones News Service Wall Street Journal Stories.
16. At the end of 1995, AT&T reportedly had approximately 30 million enrollees in its True discount plans, leaving approximately 37 million customers not on such plans. See Discount Calling Plans and Promotions (1996), Inside AT&T Corp., TPG Research and Reports, p. 122.
17. Yankee Group (August 1996).
18. Id.
19. Cf: LINK Resources Corp., 1993 Home Media Consumer Survey: Residential Telecommunications, p. 106.
20. The effect of the average length of a call on the average price per minute has become less important in recent years as long-distance carriers have lowered their charges for the first minute relative to charges for subsequent minutes.
21. An example of this procedure is the calculation of prices for AT&T's Reach Out America discount calling plan. Customers with monthly bills less than $50 were excluded from the analysis. Weights for the three categories $50, $75, and $100 are 47, 21, and 16 percent respectively. The last three categories were weighted 5 percent apiece.
22. According to AT&T's tariffs, these plans are technically called Block of Time - One Hour Plan with Evening & Day Option, True USA Promotion, and Simplified Calling Promotion "One Rate," respectively.
23. In Appendix 2 (p. 227) of The Failure of Antitrust and Regulation to Establish Competition in Long-Distance Telephone Services (P.W. MacAvoy) results are reported under the assumption that a customer switches plans immediately whenever a carrier offers a lower price plan, yielding an always minimum price index for each carrier. Minimum prices for AT&T and MCI were approximately $0.03 to $0.04 per minute lower for 1994 than those estimated for the MTS discount plans. Sprint's minimum prices were identical to the prices of its well-known discount plans until mid-1994.
24. Marginal costs are to be distinguished from long-run incremental costs, measured as the change in a firm's long-run total costs when an individual service is added (or removed) from a set of existing services. Long-run incremental costs do not include access charges.
25. Access charges could be measured in two ways: tariffed prices or average access charges per minute of service. I use tariffed access charges to be consistent with the use of tariffed prices for long-distance service.
26. See p. 106, note 2 and accompanying text in Paul W. MacAvoy (1996), The Failure of Antitrust and Regulation to Establish Competition in Long-Distance Telephone Service, Cambridge, MA: American Enterprise Institute and MIT Press.
27. Strategic Policy Research (September 23, 1996), A Bottom-Up Estimate of the TSLRIC of Long-Distance Calling.
28. LRIC per minute has likely decreased over the past decade as a result of productivity and technological gains. The one cent per minute estimate used here embodies the likelihood that technology and system productivity advances have been offset by inflationary increases. In any event, if LRIC per minute has declined over this time frame then the price-cost margins evaluated below would have an even greater increasing trend.
29. Direct Testimony of John Sumpter on Behalf of AT&T Communications of California, Inc., Application of AT&T Communications of California, Inc. (U 5002 C) for Authority to Provide Intrastate AT&T 800 READYLINE Service, June 18, 1990.
30. Wharton Econometric Forecasting Associates, Economic Impact of Eliminating the Line-of-Business Restrictions on the Bell Companies, pp. 20-21 (July 1993) (citing Bellcore data).
31. Price War Looms for Long-Distance Calls (June 25, 1996), Contra Costa Times, p. C-1.
32. Sprint's margin on its Residential Promo plan subsequently increased to 0.58 in late 1996.
33. See Hills, Michael T., (February 1994), Voice Networks Pricing Update, Business Communications Review, pp. 12-18.
34. In the Midwest, the use of outbound switched WATS service declined by a factor often during the period 1991-1995. This occurred because business customers substituted from switched WATS services to dedicated outbound services. Because of the small volume of minutes that remain in the outbound switched WATS category, no HHI was estimated.
35. HHI data are not available for Combined Services, preventing the comparison of changes in HHI with changes in margins.
36. When making comparisons between price-cost margins for switched and dedicated services, it should be recalled that the usage levels differ, being lower for switched than for dedicated. Since the fixed costs of dedicated service are higher than the fixed costs of switched service, a higher monthly usage level must be maintained by dedicated users to make the dedicated service economic relative to switched.
37. These margins are unambiguously above those previously obtained under 36 month contracts for outbound dedicated WATS services (which are not shown in Figure 8).
38. These margins are unambiguously above those previously obtained under 36 month contracts for outbound dedicated WATS services (which are not shown in Figure 8).
39. This equation formulation is presented in J. Brander and A. Zhang, Market Conduct in the Airline Industry: An Empirical Investigation, 21 Rand Journal of Economics 56 (1990). Demand elasticity is assumed to equal -0.70.
40. See Lester D. Taylor, Telecommunications Demand in Theory and Practice ch. 6 (Kluwer Academic Publishers 1994). Equivalently, the conjectural variation for a firm equals the partial derivative of all other firms' output with respect to a change in that firm's output.
41. See Stephen Martin (1993), Advanced Industrial Economics ch. 2 (Blackwell).
42. One variable indicates the conjectural variation for switched and dedicated outbound WATS while the other dummy is associated with switched and dedicated inbound WATS plans. An alternative specification using four dummy variables separating switched and dedicated did not significantly affect the results of the regression model.
43. FCC, Long-Distance Market Shares: First Quarter, 1996, Table 5.
44. FCC (December 1996), Telecommunications Industry Revenue: TRS Fund Worksheet Data, Figure 3.
45. The price-cost margins shown in Figures One through Eight have been constructed with prices based on stipulated calling patterns intended to be representative of customers in the Midwest (see Table I). These calling pattern assumptions have three dimensions: (1) time-of-day usage, (2) mileage distribution, and (3) monthly calling volume. One could ask whether the observed pattern of rising margins is specific to this calling pattern only. The sensitivity of prices can be tested by examining prices based on alternative time-of-day, mileage and calling pattern distributions. Each alternative scenario tested, results in price-cost margins that do not differ from those derived from the original set of assumptions. Price-cost margins under eight alternative scenarios increase over time and move in concert across the three interLATA providers. For an appendix demonstrating this pattern, please contact the author.