Back to Index

In the Matter of AMERICAN TELEPHONE & TELEGRAPH CO. AND THE ASSOCIATED BELL SYSTEM COMPANIES

Charges for Interstate and Foreign Communication Service;

In the Matter of AMERICAN TELEPHONE & TELEGRAPH CO. Charges, Practices, Classifications,

and Regulations for and in Connection With Teletypewriter Exchange Service

Docket No. 16258; Docket No. 15011

FEDERAL COMMUNICATIONS COMMISSION

9 F.C.C.2d 30

RELEASE-NUMBER: FCC 67-776

July 5, 1967 Adopted

INTERIM DECISION AND ORDER n1

---

n1 This decision disposes of only a portion of the proceeding, dealing primarily with jurisdictional separations and overall rate of return. The balance of the case has been reserved for later disposition (see pars. 6, 7, 8, and 9, infra).

JUDGES:

   BY THE COMMISSION: COMMISSIONERS LOEVINGER AND JOHNSON CONCURRING WITH STATEMENTS AND COMMISSIONER WADSWORTH CONCURRING.

OPINION:

    [*32] 1. GENERAL STATEMENT

   A. Procedural Matters

   1. This proceeding is an investigation instituted by our order of October 27, 1965 (2 FCC 2d 871), pursuant to sections 201(b), 202(a), [*33] 205, and 403 of the Communications Act of 1934, as amended, into the lawfulness of the charges of the American Telephone & Telegraph Co. and the Associated Bell System companies n2 (hereinafter referred to as respondents or Bell System) for interstate and foreign communication services and other related matters.

n2 The Bell System companies are hereinafter described at par. 16, infra.

   2. In that order, we directed that the investigation include inquiry into the following matters:

   1. What are the revenue requirements of the Bell System companies applicable to their interstate and foreign communication services and the basis upon which such revenue requirements are to be determined, including

   (a) The reasonableness and propriety of the procedures employed for separating and allocating plant investment, operating expenses, taxes, and reserves between the intrastate and interstate operations of such companies;

   (b) The amounts properly includible as the net investment of the abovementioned companies in property and plant used and useful for providing interstate and foreign communication service;

   (c) The amounts properly includible as expenses and taxes incurred by the above-mentioned companies in the provision of interstate and foreign communication service;

   (d) The reasonableness of the prices paid by the Bell System companies for equipment, supplies, and services purchased by such companies from the Western Electric Co. and the adjustments, if any, that should be made in the plant investments, expenses, and taxes as reflected by the amounts determined pursuant to items (b) and (c) above to the extent that any prices are found to be unreasonable. In this connection, consideration shall be given, among other things, to whether Western Electric requires a greater rate of return on investment related to its sales to the Bell System companies than the rate of return determined to be required with respect to the interstate operations of the Bell System companies;

   (e) The fair rate of return required by the Bell System on the amounts of net investment determined pursuant to the foregoing;

   (f) The amount of operating revenues that are or may reasonably be expected in the foreseeable future to accrue from interstate and foreign communication services rendered by use of the plant and facilities, the costs of which are included in the net investment to be determined herein;

   2. In the light of our determinations with respect to the aforementioned issues, whether the overall charges of the Bell System companies are just and reasonable within the meaning of section 201(b) of the Communications Act of 1934, as amended.

   3. Whether the charges for (1) message toll telephone service, (2) WATS, (3) private line telephone grade service, (4) private line telegraph grade service, and (5) all other service, except TWX and Telpak (as to which separate proceedings are now pending in dockets Nos. 15011 and 14251, respectively), are or will be just and reasonable within the meaning of section 201(b) of the Communications Act of 1934, as amended.

   4. Whether the specific charges for the above-mentioned services will subject any person or class of persons to unjust or unreasonable discrimination, or give any undue or unreasonable preference or advantage to any person, class of persons, or locality, or subject any person, class of persons, or locality to any undue or unreasonable prejudice or disadvantage within the meaning of section 202(a) of the Communications Act of 1934, as amended.

   5. Whether the Commission should prescribe just and reasonable charges or the maximum or minimum or maximum and minimum charges to be hereafter followed with respect to any or all of the above-mentioned services placed at issue by this order, and, if so, what charges should be prescribed. By order of July 20, 1966 (4 FCC 2d 545), this proceeding was consolidated with docket No. 15011, involving consideration of the charges, [*34] practices, classifications, and regulations for and in connection with American Telephone & Telegraph Co.'s Teletypewriter Exchange Service (TWX). By order of November 9, 1966 (FCC 66-1005), in docket No. 14251, the issue as to justness and reasonableness of rates for Telpak C and Telpak D private line services was also added to the issues in this proceeding and the record in docket No. 14251 was incorporated herein by reference. (Reconsideration denied 6 FCC 2d 177.)

   3. Our order designated the American Telephone & Telegraph Co. and each of the Associated Bell System companies as respondents herein (sometimes hereinafter called respondents, Bell System, or Bell). In addition, leave to intervene was granted to any connecting or concurring carrier, the Western Union

Telegraph Co. (WU), United States Independent Telephone Association (USITA), General Telephone & Electronics Corp. (G.T. & E.), the National Association of Railroad & Utilities Commissioners (NARUC), and every State regulatory body. All of the named parties and 35 State Commissions intervened. Additionally, 33 other parties were granted leave to intervene, and 31 parties were added by the consolidation of the TWX proceeding, resulting in a total of 103 parties to this proceeding, in addition to respondents. n3 In the course of the hearing, 1 day was devoted to the receipt of testimony of persons who had indicated a desire to appear as nonparties, under section 1.225 of our rules and regulations.

   n3 The additional intervenors represent independent telephone companies, other communications common carriers, government and civilian users of

respondents' services, municipalities, labor unions, and a supplier of communications equipment. Not all the parties participated in this first phase of the case, however, since many of them are interested primarily in the matters to be taken up in the next phase (see pars. 6, 7, 8, and 9, infra).

   4. By our memorandum opinion and order of December 22, 1965 (2 FCC 2d 142), we designated the Telephone Committee (Chairman Hyde and Commissioners Bartley and Cox) as a panel of Commissioners to preside at the hearing, with the assistance of an examiner to sit with them and preside in their absence. Subsequently, Hearing Examiner Arthur A. Gladstone was so designated. n4 At the invitation of the Commission, and in accordance with the plan of cooperative procedure set forth in appendix A to the Commission's general rules of practice and procedure, a panel of three State commissioners was designated by NARUC to sit with the Commission's presiding officers at the hearings. The cooperating State commissioners so designated, who later submitted comments and recommendations to the Commission, are: Commissioner Wallace R. Burke of the Connecticut Public Utilities Commission, Chairman Paul A. Rasmussen of the Minnesota Railroad & Warehouse Commission, and Commissioner Jesse W. Dillon of

the Virginia State Corporation Commission.

   n4 Memorandum opinion and order, Jan. 19, 1966, FCC 66-55.

   5. Our December 22, 1965, memorandum opinion and order established the procedures to be followed in this investigation to enable us to determine the issues "in the most expeditious and orderly fashion." To facilitate that determination, we directed that the investigation proceed in two phases, specifying the questions and evidence to be considered in each phase.

   6. As originally planned, we proposed to examine, in phase 1, respondents' [*35] total revenue requirements applicable to their interstate and foreign communication services, as well as the relevant ratemaking principles and factors controlling the distribution of such revenue requirements among respondents' principal rate classifications. Accordingly, respondents were required to present for examination their total interstate and foreign operating results, based on the most recent 12-month period, "adjusted for all known factors affecting revenues, investment, reserves, expenses, and taxes which have occurred during that period, and which may be reasonably anticipated to occur in the foreseeable future." Respondents were also directed to make an appropriate showing, with full justification, as to the rate of return required by them with respect to their total interstate and foreign operations. Phase 2, also as originally scheduled, was to include the presentation of evidence with respect to the remaining issues, such as the jurisdictional separation of plant and expenses between interstate and intrastate services, reasonableness of Western Electric prices, the amounts properly includible as respondents' net investment, expenses, and taxes, and other matters. As detailed below, however, some modifications of this plan were adopted as the proceeding developed.

   7. The first modification was made by orders of June 17, 1966 (FCC 66M-850), and July 13, 1966 (FCC 66M-960). In those orders, respondents were directed by the Telephone Committee to submit their complete justification for inclusion in their rate base of amounts of telephone plant under construction, cash working capital, and material and supplies claimed in their presentation with respect to phase 1.

   8. Further modification occurred when the Telephone Committee, in its ruling of August 12, 1966 (read into the record by the presiding examiner at Tr., pp. 3462-3464), directed respondents to present evidence describing the police considerations underlying the nonuse by the Bell System of the provisions in the Internal Revenue Code (sec. 167) permitting the use of liberalized depreciation for income tax purposes, and to estimate what the past and future dollar effects on income taxes and reserves for taxes would have been had the Bell System taken liberalized depreciation. The Telephone Committee ruled that evidence in this area would initially be considered only as it affected the issues in phase 1, but might be the subject of further testimony and consideration in phase 2. n5

n5 Memorandum opinion and order of Telephone Committee, Dec. 12, 1966, FCC 66M-1686.

   9. A further basic change in procedure resulted from our order of December 7, 1966 (5 FCC 2d 844), which modified the order of December 22, 1965 (2 FCC 2d 142), by including, among the issues in phase 1, the reasonableness and propriety of the jurisdictional separations or procedures utilized for the separation and allocation of plant investment, operating expenses, taxes, and reserves between the intrastate and interstate operations of respondents. The December 7, 1966 order also temporarily deferred consideration of ratemaking principles and factors.

   10. The order of December 7, 1966, also found that due and timely execution of our duties required us to dispense with the conventional two-step procedure of an initial or recommended decision followed by a final decision. Accordingly, the order provided that the Telephone [*36] Committee, without preparing any recommended decision, should certify the record to the Commission for decision. By its order of April 11, 1967 (FCC 67M-601), the Telephone Committee certified to us the record compiled through February 16, 1967. This record encompasses 3 days of prehearing conferences and 71 days of oral hearing (comprising 10,094 transcript pages and 119 exhibits totaling 3.485 additional pages of material), and covers the testimony of 66 witnesses. Provision was made for the filing of briefs and proposed findings of fact and conclusions, and a number of the parties availed themselves of this right. n6

   n6 Specifically, these included: American Broadcasting Co., Columbia Broadcasting System, Inc., and National Broadcasting Co., Inc.; Bell System, respondents; California Public Utilities Commission; executive agencies of the United States; Hawaii Public Utilities Commission and State of Hawaii; Independent Telephone Group; National Association of Broadcasters; National Association of Railroad & Utilities Commissioners; New Jersey Board of Public Utility Commissioners; Public Service Commission of West Virginia; and the Western Union Telegraph Co.

   11. Oral argument before the Commission, en banc, was held April 14 and 18, 1967. n7 Our order of April 7, 1967 (FCC 67-421), which assigned the dates for argument, specified a number of questions to which the parties were requested to direct answers in their presentations. These matters will be discussed later as they relate to our decision.

n7 The parties participating in the oral argument on separations were: Bell System respondents, people of the State of California and the Public Utilities Commission of the State of California, executive agencies of the United States, Hawaii Public Utilities Commission, Independent Telephone Group, National Association of Railroad & Utilities Commissioners, New Jersey Board of Public Utility Commissioners, and Western Union Telegraph Co. Those participating in the oral argument on rate of return were: American Broadcasting Co., Columbia Broadcasting System, Inc., and National Broadcasting Co., Inc.; Bell System respondents: people of the State of California and the Public Utilities Commission of the State of California; executive agencies of the United States; Independent Telephone Group; National Association of Broadcasters; New Jersey Board of Public Utility Commissioners; and Western Union Telegraph Co.

   12. On May 9, 1967, pursuant to our order of March 15, 1967 (FCC 67-341), the cooperating State commissioners filed their comments and recommendations. Two State commissioners joined in a majority view, and the third concurred in part and dissented in part in a separate statement. Comments upon the cooperating State commissioners' recommendations were filed by respondents; people of the State of California and the Public Utilities Commission of the State of California; American Broadcasting Co., Inc., Columbia Broadcasting System, Inc., and National Broadcasting Co., Inc.; Public Service Board of the State of Vermont; public utility commissioner of the State of Oregon; USITA and Public Utilities Commission of Hawaii.

   B. Collateral Procedural Matters

   13. Respondents, in a petition for modification of procedures, dated January 12, 1966, directed against the Commission's order of December 22, 1965 (2 FCC 2d 142), requested, inter alia, that the Chief of the Common Carrier Bureau and his staff not be treated as decisionmaking personnel, as provided in section 1.1209(d) of our rules and regulations. Instead, respondents urged that such personnel be subject to the same procedural requirements in this proceeding as participants representing the parties. Similar petitions were filed by G.T. & E. and USITA.

    [*37] 14. The Commission denied those petitions on March 2, 1966 (2 FCC2d 877). In their brief on rate of return and related issues (pp. 3, 4), respondents have reasserted their position that the role of the staff of the Common Carrier Bureau is "inherently unlawful and prejudicial to respondents." Our disposition of the matter, in the March 2, 1966, order, is in accord with the provisions of the Communications Act and the Administrative Procedure Act for ratemaking proceedings. The specific objection has twice been litigated and our procedures in this respect have been upheld. (Wilson & Co. v. United States, 335 F. 2d 788, 796, cert. denied, 380 U.S. 951 (1965); American Trucking Assoc., Inc. v. F.C.C., slip opinion, Sept. 15, 1966,     Fed. 2d     (1966), cert. denied, 386 U.S. 943.) The fact that the Commission, with agreement of the respondents, n8 and in accordance with law (sec. 8(a), Administrative Procedure Act), dispensed with a tentative or recommended decision does not affect the propriety of the procedures herein adopted, as contended by respondents.

n8 Respondents' counsel, on oral argument, advised that respondents "have not requested a tentative decision and we do not do so now. We are interested as the Commission is in an expedited final decision." (Tr., p. 10102.)

   C. Background for the Proceeding

   15. In our orders of October 27, 1965 (2 FCC 2d 871), and December 22, 1965 (2 FCC 2d 173), we described the four closely related problems, each of which formed the basis for one or more designated issues in this proceeding. The four matters may be paraphrased as follows:

   (a) On September 10, 1965, in connection with the Telegraph Inquiry, docket 14650, Bell submitted to the Commission a fully allocated cost study for a 1-year period ending August 31, 1964, showing investment, expenses, revenues, and rate of return for each of seven different classes of communications services making up, in sum, Bell's total interstate operations. This study indicated a wide disparity in the rates of earning for different classes of service, with the lowest rates of return tending to occur for services offered in direct competition with similar services offered by other carriers. n9 On the other hand, two noncompetitive services, message toll telephone and Wide Area Telephone Service -- comprising 85 percent of Bell's interstate business -- were shown as earning at a rate of 10 percent or higher. These facts, our order stated, warranted "a thorough examination by the Commission of the interstate rate structure of the Bell System to determine the lawfulness of the rate levels and rate relationships within that structure." (Par. 1 of order of Oct. 27, 1965.)

n9 Private line telegraph, 1.4 percent; Telpak, 0.3 percent; TWX, 2.9 percent.

   (b) Allocation or separation of common plant and expenses between interstate and intrastate services has heretofore been made in accordance with a separations manual published by the National Association of Railroad & Utilities Commissioners (NARUC). This manual is the product of informal cooperative studies involving NARUC, the Commission, the Bell System, and the independent segment of the telephone industry. The Commission interposed no objection to the use, on an interim basis, of a major revision in these separation procedures, employing important new principles of allocation. Final approval of the revised separation procedures was deferred, however, and the Commission decided that the matter should be considered formally, for the first time, in the context of this proceeding.

   (c) In its reports to the Commission, Bell has traditionally included, in its rate base, items associated with telephone plant under construction, cash working capital, and materials and supplies. In the Private Line Case, [*38] 34 FCC 217 (1963), the Commission excluded such costs. Since we have never formally determined that this exclusion shall apply generally, and since a determination as to the proper treatment of all cost items is necessary in a ratemaking proceeding, this question was also included in the investigation.

   (d) In the Private Line Case, supra (and informally thereafter), the Commission raised questions as to the reasonableness of the prices and profits of the Western Electric Co. with respect to its sales of equipment, supplies, and services to Bell System affiliates. The ultimate resolution of these questions is, of course, highly relevant to a determination of the revenue requirements of the Bell System companies. Since these questions have not thus far been resolved, they were included in this investigation.

   D. Description of the Bell System

   16. The Bell Telephone System is comprised of the American Telephone & Telegraph Co. (A.T. & T.), the parent company, 21 principal telephone operating subsidiaries, and two telephone operating companies in which A.T. & T. has a minority interest. These 23 companies are generally referred to as the Associated Bell Telephone companies. Additionally, there are a number of nonoperating subsidiaries, the most important of which are Western Electric Co., Inc., and Bell Telephone Laboratories, Inc. (see table 1 following par. 21). The Bell System is in the business of furnishing domestic and foreign communication services. It provides about 85 percent of all telephone service in the United States, the balance being provided by independent companies.

   17. Each of the Associated Bell System telephone companies furnishes communications services and facilities within the geographical regions in which they operate. These regions usually embrace all, or a portion of, several States. The companies operate in the District of Columbia and in every State except Alaska and Hawaii. The Bell System interconnects with some 2,300 independent telephone companies in the United States.

   18. Through its Long Lines Department, the parent company, A.T. & T., operates a network of cable, wire and radio circuits, and related equipment, that interconnects the territories of the regional companies. In addition, Long Lines provides service over submarine cable, satellite, and radiotelephone systems to more than 190 countries and territories throughout the world.

   19. Interstate and foreign communication services are, in general, offered jointly, under a nationwide uniform schedules of rates and charges, by respondents and the independent companies over a network of interconnected facilities. The revenues from such services, after payouts have been made to the independents as compensation for their participation, are treated each month by respondents as a common pool from which each is reimbursed for its expenses and taxes related to its participation in these services. The balance remaining in the pool is then apportioned among all respondents in proportion to their net plant in service and certain other investments assigned in that month to these services. Thus, each respondent derives the same rate of return on its investment allocated to interstate and foreign services.

    [*39] 20. The foregoing investments, expenses, and taxes are computed monthly by each respondent in accordance with separation or allocation

procedures which are considered in part IV of this decision. Unlike the respondent associated companies which engage in intrastate, as well as interstate and foreign, services, A.T. & T.'s operations, through its Long Lines Department, are entirely of an interstate and foreign nature. Hence, it requires no monthly separations of its investments and expenses.

   21. In terms of plant investment and revenues, the interstate and foreign services of the Bell System respondents account for about 25 percent of their total operations. The foreign services, in turn, are relatively small in comparison to the interstate operations of respondents. Under the above-described division of revenues arrangements among respondents, the foreign operations are treated no differently than interstate operations. Accordingly, for purposes of phase 1 of this proceeding, the foreign services of respondents are treated as if they were interstate services, and all uses herein of the term "interstate" shall be regarded as meaning "interstate and foreign."

   TABLE 1

   (Bell System Companies -- December 31, 1966)

   [Companies included in consolidated statements; American Telephone & Telegraph Co.; principal telephone subsidiaries]

Capital stocks owned by A.T. & T.

Percent Telephones

New England Telephone & Telegraph Co 69.5 4,285,954

New York Telephone Co 100.0 10,065,966

New Jersey Bell Telephone Co 100.0 3,870,954

Bell Telephone Co. of Pennsylvania 100.0 5,270,701

Diamond State Telephone Co 100.0 298,690

Chesapeake & Potomac Telephone Co 100.0 785,522

Chesapeake & Potomac

Telephone Co. of Maryland 100.0 1,905,451

Chesapeake & Potomac Telephone

Co. of Virginia 100.0 1,508,012

Chesapeake & Potomac Telephone

Co. of West Virginia 100.0 576,555

Southern Bell Telephone & Telegraph Co 100.0 10,212,325

Ohio Bell Telephone Co 100.0 3,441,061

Michigan Bell Telephone Co 100.0 3,838,940

Indiana Bell Telephone Co., Inc 100.0 1,304,839

Wisconsin Telephone Co 100.0 1,465,803

Illinois Bell Telephone Co 99.3 5,243,213

Northwestern Bell Telephone Co 100.0 3,243,870

Southwestern Bell Telephone Co 100.0 8,328,353

Mountain States Telephone &

Telegraph Co 86.8 3,280,976

Pacific Northwest Bell Telephone

Co 89.1 2,026,666

Pacific Telephone & Telegraph Co 89.7 8,948,660

Bell Telephone Co. of Nevada n1

Total 79,902,511

Subsidiaries not consolidated:

Bell Telephone Laboratories, Inc n2 50.0

Western Electric Co., Inc 100.0

195 Broadway Corp 100.0

Other companies:

Southern New England Telephone Co 18.1 1,682,130

Cincinnati & Suburban Bell Telephone Co 28.0 727,840

Bell Telephone Co. of Canada 2.4 4,868,392

Total 7,278,362

n1 Whollyowned subsidiary of Pacific Telephone & Telegraph Co.

 n2 Remainder owned by Western Electric Co.

    [*40] II. RATE BASE

   A. General

   22. The Commission, in its order, adopted December 22, 1965, 2 FCC 2d 142, stated that for the purpose of interim action (phase 1) it would accept the respondents' claimed net investment and expenses as derived from their books without adjustment. Later, however, by orders, dated June 17, 1966 (FCC 66M-850), and July 13, 1966 (FCC 66M-960), respondents were directed to submit justification for inclusion in the rate base of amounts claimed for telephone plant under construction, cash working capital, and material and supplies. The average net investment claimed by respondents for the rendition of their interstate and foreign communication services for the year 1966, as derived from their books, is as follows:

Thousands of dollars

Telephone plant:

Telephone plant in service 10,205,307

Telephone plant under construction 544,044

Property held for future telephone use 16,867

Telephone plant acquisition adjustment 219

Total 10,766,257

Other investment:

Cash working capital 141,701

Material and supplies 56,731

Investment in affiliated companies 24,498

Total 222,930

Gross investment 10,989,187

Less: Depreciation and amortization, reserves 2,382,978

Net investment 8,606,209

   23. In view of the limitations established by our orders with respect to the matters to be considered in this interim action, which limitations are stressed in respondents' brief, we do not now consider certain adjustments in the net investment and expenses derived from their books, as suggested by respondents, which would have the effect of increasing the claimed net investment by $21,003,000 and increasing claimed expenses and operating taxes by $13,344,000. We will give consideration to all of the items in the rate base, revenue, and expenses, as well as suggested adjustments, in phase 2 of this proceeding. As indicated by our orders, for the purpose of this interim decision, our critical review of rate base items is confined to telephone plant under construction, cash working capital, and material and supplies.

   B. Telephone Plant Under Construction

   24. Respondents include, as part of their claimed net investment, $544,044,000, which is that portion of the average investment recorded in account 100.2, "Telephone Plant Under Construction," n10 that has been allocated to interstate and foreign services. Respondents follow [*41] the practice of adding interest during construction, at a rate of 5 percent per annum, to the investment in telephone plant under construction and this is included in the book cost of such plant when it is transferred to account 100.1, "Telephone Plant in Service." Contracredit entries are made to "other income" as interest is accrued. Respondents have included $24,828,000 in "Revenues and Other Income" representing the income from interest during construction assigned to interstate and foreign operations during the year 1966.

n10 The Commission's rules and regulations, pt. 31, "Uniform System of Accounts for Class A and Class B Telephone Companies."

   25. The Commission's rules n11 require that reasonable amounts for interest during construction be included as a construction cost and added to the amounts in account 100.2, "Telephone Plant Under Construction." The rules do not specify the rate, but only that the amounts be reasonable. Respondents, having selected the 5-percent rate for interest during construction, contend that it is not a fair rate of return for the investment in plant under construction. They state that they selected this lower rate "* * * sufficiently below the rates of return used by regulatory authorities to avoid controversy and the accounting confusion that would result from any attempt to keep the capitalization rate precisely at the rate of return." Respondents contend that investors have supplied the funds being used for construction and that such funds are necessary for the operation of the business.

n11 47 CFR 31.100.2 and 31.2-22b(10)(i).

   26. Therefore, it is said, respondents and the investors are entitled to a full rate of return on the portion of the capital that is invested in plant under construction. In support of this contention, it is urged that plant must be completed and ready for service at the time a customer places an order for service. Respondents stress that there is very little distinction between present ratepayers and the future ratepayers who will use the telephone plant now under construction because most of the plant under construction will be used to provide expanded service to present customers rather than to provide service to new customers. Respondents further state that service to present customers would deteriorate if construction should be cut off and that the present customers benefit from construction which will be used to provide service to new customers.

   27. The California Public Utilities Commission and the executive agencies of the U.S. Government both offered expert witnesses who took issue with respondents' position with respect to inclusion of plant under construction in net investment. n12 As we reach the same ultimate conclusion as those two witnesses in this issue, we will not set forth their contentions in detail.

n12 The California Commission also urges a deduction from the rate base in the amount of $56 million representing the unpaid-for portion of purchases from Western Electric which were capitalized in accounts 100.1 and 100.2. Since the Commission's orders specify that the company's booked figures representing plant in service will be accepted for this point of the proceeding, this item will be considered in the later phases of the proceeding.

   28. Funds invested in telephone plant under construction are generated from several sources, internal and external. Since there are constant changes occurring in the proportionate contributions from those different sources, there are fluctuations in the composite costs [*42] of such funds. In view of respondents' continued use of a 5-percent rate of interest over an extended period of years, it must be presumed that they regard such rate of interest as providing reasonable and adequate compensation as to these funds. Furthermore, the practice of capitalizing interest during construction provides a return, not only on the capital invested in plant while it is in the process of construction, but also provides the basis for a continuing return on the capitalized interest throughout the life of the plant after it goes into service. Similar considerations led us to the conclusion, in the Private Line Case, n13 that telephone plant under construction should be excluded from the rate base.

n13 34 FCC 217, 219, 220; 34 FCC 244, 263, 264.

   29. In the final decision in that proceeding, we stated:

   "Where the computations of capitalized interest are available to us and the rate of capitalized interest is not shown by the record to be unreasonable, as is the case for the telephone companies in this record, we adhere to the capitalized-interest technique in preference to recognizing construction work in progress as part of the rate base." 34 FCC 217, 220. Our conclusion in that case, that the investment in plant under construction should not be included in the rate base when interest during construction is capitalized, was in accordance with the weight of authority as reflected by decisions of other regulatory agencies. n14 A further review of such decisions since our Private Line decision reaffirms that this position continues to be in accordance with the weight of authority. n15 The record in this proceeding offers no new or material evidence which would warrant a finding different than that in the Private Line Case. Accordingly, $554,044,000 for telephone plant under construction claimed by respondents in their net investment is disallowed and the related interest charged to construction ($24,828,000) is excluded from "Other Income."

n14 Los Angeles v. Southern California Tel. Co. (1936), 14 PUR NS 252; Upstate Telephone Corp. of N.Y. (1936), 13 PUR NS 134; Ohio Bell Tel. Co. v. PUC (1936), 15 PUR NS 443; Southern Bell Tel. & Tel. Co. (Alabama, 1948), 75 PUR NS 298; Chesapeake & Potomac Tel. Co. (District of Columbia, 1953), 99 PUR NS 314; Southern Bell Telephone & Telegraph Co. (Florida, 1952), 92 PUR NS 335; Southern Bell Telephone & Telegraph Co. (Georgia, 1951), 91 PUR NS 97; Citizens Telephone Co. v. PSC (Kentucky, 1952), 94 PUR NS 383; New England Telephone & Telegraph Co. (Massachusetts, 1949), 83 PUR NS 238; New England Telephone & Telegraph Co. (New Hampshire, 1949), 78 PUR NS 67; New Jersey Bell Telephone Co. (1949), 78 PUR NS 97; Southwestern Public Service Co. (New Mexico, 1951), 90 PUR NS 449; Carolina Telephone & Telegraph Co. (North Carolina, 1952), 95 PUR NS 500; and Southern Bell Telephone & Telegraph Co. (Tennessee, 1953), 100 PUR NS 33.

   n15 Pacific Tel. & Tel. Co. (1964), 53 PUR 3d 513; New Jersey Water Co. (1963), 51 PUR 3d 224; Chesapeake & Potomac Tel. Co. (1964), 57 PUR 3d 1. Bell contends that all three items of rate base discussed herein represent investor-supplied funds and hence are entitled to a full, fair return. It cites in support of its position a statement from the minority opinion in Missouri ex rel. Southwestern Bell Tel. Co. v. Missouri Pub. Serv. Commission, 262 U.S. 276, 290, PUR 1923 C 193, 201, that capital embarked in the public utility enterprise is guaranteed a fair return by the Federal Constitution. No authority is cited by respondents supporting the specific proposition that plant under construction must be added to the rate base. We do not question that such plant under construction is necessary for the provision of service, as contended by Bell. We hold merely that plant under construction, in the circumstances of record and for the reasons stated above, should not be included in the rate base.

C. Working Capital

(1) General

   30. Working capital includes the funds representing necessary investment in materials and supplies plus the cash required to pay operating expenses incurred for services rendered prior to the time [*43] when revenues for such services are available to the utility. Such funds should be included in the rate base only to the extent that they are supplied by investors.

   31. The purpose of allowing cash working capital in the rate base is to compensate investors for funds required to be provided by them for the purpose of paying operating expenses in advance of receipt of revenues from customers and in order to maintain minimum bank balances. n16 There is a distinction between the working capital that is supplied by investors and is necessary for the day-to-day operation of the business, and other funds which are held for any other purposes management may desire. This distinction has been expressed as follows:

n16 34 FCC 244, 271.

   Working capital consists of that capital, above the investment in fixed assets and intangibles, which is necessary for the economical and satisfactory conduct of the enterprise. Working capital, in the technical sense in which it is here employed, does not include the total liquid funds with which the business is conducted. It is not the property which the business has; that is, it is not the excess of current assets over current liabilities. Working capital, rather, is an allowance for the sum which the company needs to supply from its own funds for the purpose of enabling it to meet its current obligations as they arise and to operate economically and efficiently. n17

n17 Irston R. Barnes, "The Economics of Public Utility Regulation," F. S. Crofts & Co., 1942 (p. 495).

   32. Respondents claimed as part of their interstate investment rate base in 1966, $141,701,000 representing cash working capital and $56,731,000 representing materials and supplies. In support of these amounts, respondents presented evidence regarding: (1) An allocation of the total average cash balances on hand and the results of a "lag study"; (2) a "balance sheet analysis"; and (3) its "current ratio" requirements. Parenthetically, it is noted that the balance sheet approach is used by respondents to support working capital, and also the entire claimed net investment, including telephone plant under construction and material and supplies.

(2) Cash Working Capital

   33. The respondents' claim for cash working capital was derived from an allocation of the funds in their cash accounts and a "lag study" made to determine the operating costs paid in advance or in arrears of receipt of revenues. This resulted in a claim for cash working capital of $141,701,000,

composed of the following:

Millions

Cash and temporary cash investments $325.3

Payment of expenses in advance of receipt of revenues 94.0

Revenue received in advance of payment (247.2)

Federal excise tax collected in advance of payment (30.4)

Total cash working capital 141.7

(a) Allocation of cash on hand

   34. The cash and temporary cash investments of $325.3 million represent an allocation to interstate operations of the average amounts [*44] of cash and temporary cash investments maintained by respondents during the year 1966, as follows:

Millions

Long lines $5.7

Associated companies 23.3

General department 296.3

Total 325.3

Respondents also include $13,451,000 in "Other Income," representing the portion of interest on temporary cash investments allocated to interstate operations.

   35. The most significant amount of cash is that maintained by the General Department of A.T. & T., representing over 90 percent of the total cash on hand in 1966. The remainder, $29 million, represents cash on hand of the associated companies and Long Lines Department of A.T. & T. and is comprised of $24.6 million in demand deposits, $2.5 million in working funds, and $1.9 million in temporary cash investment of the associated companies. The principal purpose of the General Department funds, which are maintained in a central "pool of funds," is to make advances to the associated companies for use in day-to-day operations and for the day-to-day operations of the A.T. & T. Co., including its Long Lines Department. The principal purpose of the funds maintained by the associated companies and Long Lines is to satisfy bank requirements for the maintenance of minimum bank balances and working funds which are discussed later. The activities which are considered by respondents to be the "day-to-day operations" of A.T. & T. include advances to the associated companies for construction, purchase of the stock of associated companies, payments of dividends to stockholders, interest payment to bondholders, and various other corporate purposes for which the General Department is responsible. The sources, or inflow, of funds into the General Department are principally from dividends and interest received from the associated companies, Long Lines Department earnings, license contract fees paid by the associated companies, the sale of A.T. & T. stocks and bonds, and the repayment of advances by the associated companies and the Long Lines Department.

   36. The amount of the pool of funds is determined by respondents' stated policy to maintain a 1-to-1 ratio of current assets to current liabilities. As the other elements of current assets, i.e., accounts receivable and material and supplies, are relatively inflexible, it is necessary to increase or decrease cash funds periodically. It is this policy that dictates the amount of cash to be held in the pool of funds, rather than the requirement of respondents' day-to-day operations. An indication of the amount of cash required solely for the purpose of maintaining this ratio is given by the fact that the minimum balance in the pool of funds was $1,215 million in 1965. During the first 6 months of 1966, the minimum balance was $669 million and the ratio dropped below 1-to-1 during this period.

   37. Respondents assert that the funds collected from others to be paid at a future date to Federal, State, and local governments should be treated as a form of "trust fund" and should be restricted in use. It [*45] submitted a letter from outside legal counsel, dated December 5, 1966, stating an opinion to the effect that taxes collected by respondents to be paid to the Federal Government may be commingled with company funds, but that there is an "implied" condition that such funds must not be utilized in the operation of the business and that nonsegregated tax funds must be treated in such a way as to effectively remove these funds from working capital. No authority for this viewpoint is cited, which, as we note below, is contrary to decisions of regulatory authority. On the other hand, a Bell witness asserted, with respect to these tax funds:

   The Government doesn't care what you do with them as long as you pay them over to the Government on the day they are due (Tr., p. 4033).

   38. It has been a widespread practice, for a long period of years, for regulatory bodies to treat taxes collected in advance of the time when payment is due as being available to operate the business. n18 To our knowledge, no governmental agency has asserted that this treatment of taxes is prohibited. The Government will invoke restrictions and penalties only in the event payments are not made when due. The Internal Revenue Code requirement for segregation of tax funds into a special fund applies only after there has been a failure to remit taxes when due. n19 The respondents have not been subjected to this restriction or penalty in the past and it is unreasonable to assume that they will be in the future.

n18 Pittsburgh v. Pennsylvania PUC (1952), 94 PUR NS 353 (Bell Telephone of Pennsylvania); Southern Bell Tel. Co. (Alabama, 1954), 4 PUR 3d 195; Diamond State Tel. Co. (Delaware, 1958), 21 PUR 3d 417 and cases cited therein; Southern New England Tel. Co. (Connecticut, 1962), 42 PUR 3d 310; Randolph Tel. Co., Inc. (North Carolina, 1962), 46 PUR 3d 30; New Jersey Water Co. (New Jersey, 1963), 51 PUR 3d 224; the Pacific Tel. & Tel. Co. (California, 1965), 58 PUR 3d 229.

n19 Internal Revenue Code of 1954, sec. 7512.

   39. To the extent that the General Department pool of funds is used for future construction expenditures, accruals for future dividend declarations and payment, and future interest payments to bondholders, no need or justification for inclusion of amounts for these purposes in the working capital requirements had been shown and, in fact, it is inconsistent with the concept of working capital requirements to do so, since such items are not required for the day-to-day operations of the business. n20

n20 34 FCC 244, 271.

   40. The Commission, therefore, disallows the claim for the General Department funds which, for the year 1966, as recorded, amounted to $296,300,000

and excludes from "Other Income" the related interest from temporary cash investments ($13,451,000).

   41. In addition, the $1.9 million in temporary cash investments of the associated companies represents cash in excess of the associated companies' requirement and falls in the same category as the General Department pool of funds.

   42. For the same reasons the General Department funds were disallowed, the Commission disallows respondents' claim of $1.9 million in temporary cash investment of the associated companies.

   43. There remains $27,100,000 in demand deposits and working funds to be considered. While respondents have given no specific justification for amounts required for working funds and minimum [*46] bank balances, it appears that the $27.1 million is related to these purposes. Respondents' claim of $2.5 million for working funds required by respondents' employees for incidental expenses does not appear unreasonable in view of respondents' widespread operations and number of employees. Of the $24.6 million in demand deposits, about 50 percent was in the form of float; i.e., the amount represented by the lag between the date checks are deposited and the date collection is made.

   44. The record shows that respondents have accounts in about 4,600 banks throughout the country. It is recognized that each of these accounts must maintain a balance in order to avoid bank charges. Further, since there are fluctuations in the day-to-day needs for payments of operating expenses, additional amounts in the cash accounts are required to meet peak requirements. Therefore, while respondents have presented no specific justification for the amounts required for minimum bank balances, we will allow for purposes of this phase of the proceeding the $24.6 million for this purpose. However, it is expected that respondents will produce evidence justifying the amount for required minimum bank balances in the next phase of this proceeding, when we will again review this matter. Accordingly, we allow at this time $27,100,000, the sum of demand deposits and working funds.

(b) Lay study

   45. In developing cash working capital, operating cost paid in advance or in arrears of receipt of revenues was computed by respondents from data determined by a "lag" study of total interstate services. From this study, respondents determined the net average number of lag days associated with expenses, Federal taxes on income, and other taxes. Based on these lag data, daily expenses, and tax data, it was determined that, for its interstate and foreign services for the year ending December 31, 1966, as recorded, respondents required, on the average, $94 million for payment of expenses in advance of receipt of revenues. On the other hand, receipts in advance of the payment of various taxes, primarily Federal income taxes, amounted to $247,200,000 and the average cash available from Federal excise tax collections amounted to $30,400,000. The net result of the foregoing shows negative cash working capital requirements of $183,600,000. This last figure is reduced by our previous allowance of $27,100,000 for demand deposits and working funds, leaving a negative working cash requirement of $156,500,000.

   46. The California Commission contends that the revenue lag effect of depreciation and amortization expenses should be included in the lag study. It was stated that the current accrual for depreciation is deducted in arriving at the net investment in advance of the time when the funds to reimburse respondents for the depreciation are still in the hands of the subscribers. The effect of including this item, according to the witness and based on 1965 data, would be to increase the cash working capital requirement by $44,300,000. Depreciation and amortization expenses do not represent an outlay of cash on the part of respondents and have no bearing in a determination of working [*47] cash requirements. The Commission, therefore, rejects the inclusion of any amounts related to depreciation and amortization expenses.

(3) Materials and Supplies

   47. The amount representing material and supplies, claimed by the respondents, is $56,700,000. This represents the portion of the average investment in material and supplies used in telephone repair and construction, which has been allocated to interstate operations. However, about 20 percent of the amount represents material and supplies purchased from Western Electric for which payment has not been made. The effect of including the unpaid-for portion related to the amounts of materials and supplies charged to expenses has been considered by respondents in the lag study, but they have not considered that portion which will be capitalized as plant in service. According to the record, the portion of material and supplies that will be capitalized is about 75 percent of the total, but, of that amount, about 20 percent has not been paid for and should not be included in this item. Inasmuch as the unpaid-for portion does not represent funds supplied by the investors, it is appropriate to reduce the claimed amount by $8,505,000. The Commission, therefore, allows $48,195,000 for material and supplies as a component of the working capital required by respondents for their interstate operations.

(4) Balance Sheet Analysis

   48. Respondents submitted a "balance sheet analysis," which was supported by USITA, as additional justification for its claimed rate base, and primarily for the amount claimed for cash working capital. The objective of this approach is to demonstrate that the amount of funds supplied by investors is as great or greater than the entire amount of the claimed rate base. It follows, therefore, according to respondents, that this method of analysis demonstrates the correctness of including the claimed amounts for telephone plant under construction, cash working capital, and material and supplies in the rate base, as parts of the total funds furnished by investors. This approach consisted of developing a combined balance sheet for respondents, including the Cincinnati & Suburban Bell Telephone Co. and the Southern New England Telephone Co.

   49. Respondents eliminated from this combined balance sheet the investment items which they conclude were not directly related to the furnishing of interstate and foreign services, together with the off-setting items on the liability side of the balance sheet. It was then determined which of the items on the liability side of the balance sheet, such as accounts payable, taxes accrued, and advance billings, represented funds obtained from sources other than investors. It was concluded that the remainder of the items on the liability side of the balance sheet indicated the funds that were supplied by investors. By this process, the balances shown as common stock equity and long-term debt, which make up the usual classes of permanent capital, were classified as funds supplied by investors. These items total less than the total claimed rate base. In addition to these two classes of capital, the current liabilities of notes payable, interest and dividends accrued, and [*48] premium of long-term debt also were classified as funds supplied by investors. In addition, the deferred credits of unamortized investment credit, insurance and provident reserve, and matured interest and dividends were attributed to investors. The inclusion of these current liabilities and deferred credit items as a part of investor-supplied funds results in an aggregate of investor-supplied funds greater than the rate base.

   50. The respondents have never advanced this theory before and their witness knew of no jurisdiction where it ever has been accepted. n21 No actual showing was made that these current liabilities and deferred credits do, in fact, represent funds supplied by investors. Instead, they relied upon the premise that, even though these current liabilities are to be paid for by funds collected from customers, respondents are entitled to a full return because the funds have come into their custody and are thus considered as investor supplied. We find no logic in this contention and, therefore, reject the balance sheet approach as support for the inclusion of these items in the rate base.

   n21 See City of Pittsburgh v. Pennsylvania PUC (1952), PUR NS 353, 357, 358, "* * * we think that the superior court placed undue emphasis upon the balance sheet position of Bell either at a particular time or over a period of time as a factor in determining the need for cash working capital. Fluctuations in the cash and current assets position of a company are controlled by managerial policy. Such position has little to do with cash working capital requirements, which basically depend upon such factors as, for instance, whether the company follows a policy of paying out dividends relatively soon after they are earned or accumulating a large surplus. The balance sheet position per se, if a material factor, is one of the least important factors in determining the need for cash working capital." See also Northern Natural Gas Co. (1952), 11 FPC 375, affirmed 206 F. 2d 690, cert. denied 346 U.S. 922 (1954).

(5) Current Ratio Requirements

   51. Another method advanced by respondents in corroboration of their claimed cash working capital requirements was a determination of the appropriate level of current assets in relation to current liabilities, designated the "current ratio." The current ratio is derived by dividing the current assets by the current liabilities. Respondents state that it is their longstanding company policy to maintain a current ratio of about 1 to 1; i.e., $1 of current

assets for each $1 of current liabilities (commonly referred to as a ratio of 1). They claim this policy is required so as to have sufficient liquidity at any time to pay all current liabilities and to maintain financial stability. In support of this policy, respondents submitted evidence relating to the current ratios of selected groups of companies in other regulated areas; e.g., electric, gas, etc., and of five major companies each representative of a nonregulated industry. The ratios are the average for the 10-year period ending 1965 for each of these groups. The average current ratio for A.T. & T. (consolidated) was also determined for the same 10-year period. We have, therefore, a comparison, on the one hand, of A.T. & T. with an average current ratio of u.1, and, on the other hand, the average current ratios for these groups ranging from 1 to 2.3. In addition, it was shown that the minimum-maximum range of all the companies, including respondents, was 0.7 to 3. Based on these data, respondents contend the ratio of 1 is reasonable for them.

   52. Respondents did not evaluate the characteristics peculiar to the various companies and industries studied with respect to the composition of current assets, such as relative amounts of cash, inventory, and [*49] accounts receivable. The amount in each of these components would be dictated by the requirements of the particular industry or company. For example, although the electric industry as a group has an average current ratio close to that of respondents, cash and temporary cash investment for electrics are 33 percent of total current assets, whereas for respondents they are 50 percent. Stated differently, the cash and temporary cash investments of the electrics equaled 1.9 percent of their gross utility plant investment, while that of respondents

equaled 6 percent of their gross telephone plant investment.

   53. Although the electrics have almost twice the plant investment of respondents and about the same growth in plant, the electrics' cash and temporary cash investment was about 40 percent less than that of respondents. A significant difference in the current assets of respondents and the electrics occurs in the amount maintained in material and supplies. For respondents this represents about 5 percent and for the electrics about 25 percent. Moreover, respondents did not evaluate the different accounting policies within the various companies which could produce computed current ratios that are not meaningfully comparable. It is, therefore, apparent that a mere comparison of current ratios does not afford a reasonable or convincing basis for determining the required cash working capital for the respondents.

   54. In further support of their position, a second ratio was developed by the respondents in a similar manner as the current ratio, using the averages of the same groups of companies and the same time period. The ratio was designated as the "current funds/current applications" ratio and is determined by dividing current assets plus annual internal sources of funds by current liabilities plus annual application of funds. The ratio is considered by respondents as an expansion of the current ratio, in that net income and depreciation accruals are added to the current assets and capital expenditures and dividends are added to the current liabilities. The purported usefulness of the ratio with respect to cash working capital requirements is, thus, stated by respondents:

   To the extent that internal sources of funds are significantly less than applications of funds for forthcoming capital expenditures and dividends, the current ratio tends to overstate the apparent liquidity of the balance sheet. The current funds/current applications ratio tends to correct the overstatement. (Bell exhibit No. 32, p. 6.) The average ratio determined by this method was 0.8 for A.T. & T. and the other industry groups of companies ranged from 0.7 to 1.7, with the minimum-maximum range for the individual companies being 0.6 to 2. The comparison of these ratios has all the defects associated with the current ratio comparisons previously discussed. Moreover, these comparisons give no consideration to such factors as dividend policy and the construction policy which could have a significant effect on the results of the comparisons.

   55. USITA supports respondents' position with respect to the appropriateness of respondents maintaining a current ratio of at least 1. The USITA witness had made no study of respondents' cash working capital requirements, but compared the current ratios of respondents [*50] and those of four large independent telephone companies for the 5-year period 1961-65. The average of the ratios for the 5-year period ranged from 1.2 for respondents to 1.9 for G.T. & E. n22 Three of the four companies are holding companies and the fourth is the Hawaiian Telephone Co., which, of course, has operating peculiarities related to its location. n23 The 1.9 average current ratio for G.T. & E. includes its manufacturing affiliates, but respondents' ratio does not include the Western Electric Co. The record shows that, for its combined telephone subsidiaries, at December 31, 1965, G.T. & E. had an average ratio of 0.6 (0.8 if "Notes Payable" are excluded). It is also shown that the composition of current assets of these four independents are substantially different than respondents; e.g., material and supplies are from four to seven times greater than that shown for Bell. Respondents have approximately four times as much cash and temporary cash investment in relation to total assets as to all the independent telephone companies reporting to this Commission. n24

n22 Notes payable to banks were excluded from current liabilities of each company on the grounds that they would eventually be converted into some type of permanent financing.

n23 Hawaiian Telephone Co., operating as it does in a group of islands and between widely spaced urban centers, has operating problems unlike maintained companies.

n24 "Statistics of Communication Common Carriers, FCC," year ended Dec. 31, 1965.

   56. The USITA witness made it clear that "each utility's needs for working capital furnished by its investors are based on its own situation * * *." We agree.

   57. The current ratio comparisons advanced by respondents and USITA do not aid the Commission in determining respondents' cash working capital requirements. Further, arguments advanced by respondents with respect to the use of current assets and current liabilities have not been accepted by other regulatory bodies. n25 Accordingly, the Commission does not accept the current ratio as a method of determining cash working capital

requirements.

   n25 City of Pittsburgh v. Pennsylvania PUC (1952), 94 PUR NS 353.

(6) Conclusions

   58. The Commission concludes that the working capital requirements, based on the year 1966 data, as recorded, for the respondents' interstate and foreign

services are as follows:

Cash on hand $27,100,000

Payment of expenses in advance

of receipt of revenues 94,000,000

Revenues received in advance of payment (247,200,000)

Federal excise tax collected in advance

of payment (30,400,000)

Cash working capital (156,500,000)

Material and supplies 48,195,000

Total working capital (108,305,000)

   59. The findings above show that funds which have been furnished by other than investors, primarily by respondents' customers, are $108,305,000 in excess of the amount required to meet the payment of operating expenses as they fall due. It is advocated by California that this amount should be shown as negative working capital and operate as a reduction of the recorded rate base or average net investment. Respondents maintain that the showing of a negative working capital amount may not, in any event, be extended beyond a zero allowance for [*51] working capital. To do otherwise, they contend, would go beyond the issues of phase 1, in that the amount for telephone plant in service would, thereby, be reduced. Without passing on the merits of including the negative requirement for cash working capital in the rate base, we conclude that a determination in this matter should be made during phase 2 of this proceeding.

   D. Allowed Rate Base

   60. Based on the evidence in the record, and for the reasons heretofore stated, we conclude, for purposes of this interim action, the proper rate base for respondents' interstate and foreign services for the test year 1966 is summarized as follows:

Telephone plant:

Telephone plant in service $10,205,307,000

Property held for future telephone use 16,687,000

Telephone plant acquisition adjustment 219,000

Total 10,222,213,000

Other investments:

Investment in affiliated companies 24,498,000

10,246,711,000

Gross investment:

Less depreciation and amortization

reserve 2,382,978,000

Net investment 7,863,733,000

   61. We note parenthetically that respondents offered witnesses who discussed the effects of inflation on the rate base and suggested increases up to 14 percent in the net investment and up to 18 percent in depreciation charges to compensate for such inflation. Respondents subsequently made it clear, however, that they recognize that this Commission is committed to regulation on original cost rate base. Respondents requested that the inflation testimony be considered in connection with the fixing of a rate of return. Accordingly, we have given no consideration to this testimony in respect to rate base.

   III. RATE OF RETURN

   A. General

   62. The determination of the proper rate of return to be allowed respondents is one of the major issues in this proceeding. Much of the testimony and exhibits of respondents, as well as intervenors and Commission-sponsored witnesses, were addressed to the subject.

   63. Rate of return in simplest terms is a percentage expression of the cost of capital. It is just as real a cost as that paid for labor, material, and supplies, or any other item necessary for the conduct of business. A difficulty arises in determining, in the context of a ratemaking proceeding, what the purported cost of capital is to respondents, and whether this has been prudently incurred, or whether some other cost should be established by the regulatory authority in accordance with established legal standards. Accordingly, respondents' [*52] claimed rate of return must meet the test of reasonableness. A return which is too low could impair the ability of the respondents to raise additional needed capital and also imperil the integrity of existing investment, with adverse effects on the quality of their service. A return which is too high results in charges to the public which would be above the just and reasonable level.

   64. Respondents have historically raised the capital necessary for the conduct of their business through the issuance of both debt and equity securities. The rate of return to which they are entitled is the weighted average of the cost of the debt (interest) and the earnings or profit they require on the invested equity capital. The cost of debt, which is a prior charge against revenues before taxes, is lower than that of equity which is entitled to the profit or net remaining after all other costs, including interest and taxes. So, for example, while it takes only $1 of net earnings before taxes to pay $1 of interest on a bond, it takes $1.92 before taxes to earn $1 net on equity after payment of 48 percent Federal corporate income tax.

   65. Accordingly, the overall rate of return is affected by the capital structure in respect to the proportion of debt and equity in the total capitalization of the company. Respondents have the obligation to fix this proportion in such way as to raise the required capital at the lowest possible cost consistent with their overall responsibility to provide modern, efficient service at reasonable rates and to maintain the financial integrity of the enterprise.

   66. While respondents in the first instance determine their capital structure as part of their managerial function, their judgment in this respect, as in other areas, is subject to regulatory review. Capital structure with too large a proportion of debt may adversely affect the ability of the company to raise funds, and thereby increase the overall cost of capital and impair its financial integrity. Capital structure with too little debt may not only require the ratepayers to pay more than would otherwise be required for service, but also may adversely affect equity owners by reducing their earnings and dividends, as well as depressing the price of their stock. A balance must be struck which, on the one hand, obviates the risk inherent in too much debt, and, on the other hand, avoids the unduly high charges to the public and adverse affects upon shareholders from too little debt.

   67. No specific standards are established in the Communications Act for computing the rate of return. It is subsumed in the statutory standard that rates must be "just and reasonable" (secs. 201(b) and 205(a)) and in the stated purpose of the act that there be "adequate facilities at reasonable charges" (sec. 1). The term "just and reasonable" was brought into the Communications Act from the Interstate Commerce Act and also appear in many State regulatory statutes. This standard, therefore, has been the subject of considerable regulatory attention and of many court decisions.

   68. As early as 1923, the U.S. Supreme Court set forth certain basic guidelines for reaching a determination as to what is a fair or proper rate of

return. The Court stated:

 A public utility is entitled to such rates as will permit it to earn a return on the value of the property which it employs for the convenience of the public [*53] equal to that generally being made at the same time and in the same general part of the country on investments in other business undertakings which are attended by corresponding risks and uncertainties; but it has no constitutional right to profits such as are realized or anticipated in highly profitable enterprises or speculative ventures. n26

n26 Bluefield Water Works & Improv. Co. v. West Virginia Pub. Service Commission (1923), 262 U.S. 679, 692, 693; PUR 1923D 11, 20, 21; 67 L. Ed. 1176; 43 S. Ct. 675.

   69. In a second landmark case, the Supreme Court reiterated its views in the following language:

   From the investor or company point of view it is important that there be enough revenue not only for operating expenses but also for the capital costs of the business. These include service on the debt and dividends on the stock. Cf. Chicago & Grand Trunk Ry. Co. v. Wellman, 143 U.S. 339, 345-346. By that standard the return to the equity owner should be commensurate with returns on investments in other enterprises having corresponding risks. That return, moreover, should be sufficient to assure confidence in the financial integrity of the enterprise, so as to maintain its credit and to attract capital. See Missouri ex rel. Southwestern Bell Tel. Co. v. Public Service Commission, 262 U.S. 276, 291 (Mr. Justice Brandeis concurring). n27

n27 Federal Power Commission v. Hope Natural Gas Co. (1944), 320 U.S. 591; 51 PUR NS 193, 200, 201.

   70. State and Federal regulatory agencies have followed the principles enunciated by the Supreme Court in Bluefield and reaffirmed in Hope; e.g., Re Area Rate Proceeding for Permian Basin (FPC 1965), 59 PUR 3d 417, 469; Re Southwestern Bell Telephone Company (1960), 34 PUR 3d 257, 315 (Kansas State Corp. Commission 1960); affirmed Southwestern Bell Telephone Co. v. State Corp. Comm., 51 PUR 3d 113; 386 P. 2d 515 (Kansas Supreme Court, 1963).

   71. Controversies arise, however, in the application of these enunciated principles, in specific proceedings in which rates of return are determined. The dispute centers generally, as it has in this proceeding, on the application of the principle that the utility is entitled to a return equal to that of businesses of "corresponding risks."

   72. Turning to the substantive matters before us, we note that there is no dispute as to the embedded cost of debt for the Bell System. At the outset of the hearings early in 1966, testimony of Bell System witnesses fixed the embedded debt cost of 3.93 percent. By the time the record in this phase of the proceeding was closed about a year later, testimony showed that the embedded cost of Bell System debt had increased to about 4 percent. Interest costs to respondents have been trending upward and currently are in the range of 5 1/2 to 6 percent. It is expected by respondents that the secular trend of interest rates will be upward and, although they may not stay at current levels, they are not expected to go below 4 1/2 to 5 percent.

   73. There are, therefore, two significant areas of controversy which must be resolved in order to make a determination as to the allowable rate of return: The required return on equity and the proper capital structure.

   74. Our objective in fixing an allowable rate of return must be to meet the requirements of the public interest in just and reasonable rates consistent with the generally accepted judicial guidelines to which we have referred earlier. In determining that rate of return which will sustain the financial integrity of respondents' business and enable them to attract capital, we must emphasize that it would be inappropriate [*54] for us to give weight to the ambitions of the speculator who seeks the quick or unusual profit on his investment. Certainly, it would not be in the public interest for public utility rate

regulation to encourage or to facilitate extraordinary or speculative profits.

B. Summary of Positions

(1) Comparable Earnings

   75. It is respondents' position that they need a rate of return of at least 8 percent. They conclude, in their rate of return brief, at page 7, that as: "* * * the record stands * * * there is no reliable evidence to support a return lower than 8 percent." Respondents then assert:

   In determining that 8 percent is a minimum fair return, the critical considerations are: (1) How much should A.T. & T. earn on its common stock? And (2) has the Bell management been prudent in its policies regarding capital structure, i.e., the debt equity ratio? We note from the record that no regulatory agency have ever, in a formal proceeding, approved a rate of return at or even approximating this level for any electric or telephone utility.

   76. Respondents' basis for determining the minimum fair rate of return requirement is set forth in their proposed findings, R.R. 176-178, as follows:

The cost of capital for the Bell System's interstate business is composed of 35 percent debt at a cost of 4 percent and 65 percent equity at a cost of 10 to 11 percent. The fair rate of return is a range of 7 1/2 to 8 1/2 percent.

   Corroboration of this range is given by Scanlon's study of comparable overall earnings of the 50 largest manufacturing companies in the period 1925-65. The Bell System requires overall earnings of at least 8 percent in order to maintain the same relationship to these companies as existed prior to World War II.

   The overall return of the Bell System should be in the upper end of the 7 ½ to 8 1/2 percent range in view of the higher economic performance which will be required of A.T. & T. in the environment of the future and because of the rising profit levels of unregulated industry, with which the Bell System must compete for capital. [Emphasis supplied.]

   77. Thus, the basic position of respondents is that they are entitled to "comparable earnings" to those of unregulated industry. They further assert that this is required by the language of the Hope and Bluefield cases quoted above.

   78. At the oral argument, respondents summed up their position in the following terms:

   The Hope case decided in 1944 set the standard that the common stockholder is entitled to a return "commensurate with returns on investments in other enterprises having corresponding risks."

   The test is not limited to identical companies. It is not limited to regulated companies nor to companies in the same industry.

   In the Hope case, comparisons were made with a broad group of manufacturing companies. They were, in fact, the 400 Standard & Poor manufacturing companies, which is about the same group we have used here except today there are more companies than there were then. (Tr., pp. 10305-10306.)

   79. In evaluating respondents' contentions as to the meaning of the Hope case, we note, first, that nowhere have they cited specific authority for their contention that they must be allowed earnings comparable [*55] to nonregulated concerns. Our own review of applicable authorities also does not reveal any such support, but, to the contrary, shows many precedents against this contention, including later cases decided by the U.S. Supreme Court.

   80. In the Hope case, the Federal Power Commission did not use the standards by which respondents say we must be bound. Instead, it stated: That evidence reveals unmistakably that, compared to industrial and railroad enterprises, the utility business has relatively greater stability * * *. Cleveland and Akron v. Hope Natural Gas Co., 44 PUR NS 1, 32 (1942).

   81. The U.S. Supreme Court, likewise, merely noted that:

   It (the FPC) considered the financial history of Hope and a vast array of data bearing on the natural gas industry, related businesses, and general economic conditions * * *. FPC v. Hope Natural Gas Co., supra, 320 U.S. 591, 51 PUR NS 193, 201 (1944).

   82. Furthermore, the U.S. Supreme Court, in a later decision in 1963, State of Wisconsin et al. v. Federal Power Commission, 373 U.S. 294, 48 PUR 3d 273, 282, 283 (1963), stated:

But to declare that a particular method of rate regulation is so sanctified as to make it highly unlikely that any other method could be sustained would be wholly out of keeping with this Court's consistent and clearly articulated approach to the question of the Commission's power to regulate rates. It has repeatedly been stated that no single method must be followed by the Commission in considering the justness and reasonableness of rates, Federal Power Commission v. Natural Gas Pipeline Co. (1942), 315 U.S. 575, 42 PUR NS 129, 86 L. ed. 1037, 62 S. Ct. 736; Federal Power Commission v. Hope Nat. Gas Co. (1944), 320 U.S. 591, 51 PUR NS 193, 88 L. ed. 333, 64 S. Ct. 281; Colorado Interstate Gas Co. v. Federal Power Commission (1945), 324 U.S. 581, 58 PUR NS 65, 89 L. ed. 1206, 65 S. Ct. 829, and we reaffirm that principle today. As the Court said in Hope:

"We held in Federal Power Commission v. Natural Gas Pipeline Co., supra, that the Commission was not bound to the use of any single formula or combination of formulae in determining rates. Its ratemaking function, moreover, involves the making of 'pragmatic adjustments." Id. at page 586. And when the Commission's order is challenged in the courts, the question is whether that order 'viewed in its entirety' meets the requirements of the act. Id. at page 586. Under the statutory standard of 'just and reasonable' it is the result reached not the method employed which is controlling." 320 U.S. at page 602, 51 PUR NS at page 200.

   83. Furthermore, State commissions and courts have repeatedly and specifically rejected respondents' comparable earnings contention. For example, the Kansas State Corporation Commission, in dealing with the identical contention made here, stated:

   It is clear that the applicant is not entitled either in fact or in law to compare its earnings with that of industrials * * *.

After discussing the Bluefield and Hope cases, that commission further said:

   Clearly the applicant is not entitled to a rate of return synonymous with the earnings produced by highly unstable industrial companies, which are not only dissimilar to the applicant but are also dissimilar to each other. Re Southwestern Bell Telephone Co. (1960), 34 PUR 3d 257, 315. The decision of that commission was affirmed by the Kansas Supreme [*56] Court, Southwestern Bell Telephone Co. v. State Corp. Commission, 386 P.2d 515, 51 PUR 3d 113. n28

n28 See also Re Michigan Bell Telephone Co., 32 PUR 3d 395 (1960); Re Northwestern Bell Telephone Co., 92 PUR NS 65; Re Southern Bell Telephone & Telegraph Co. (Florida), 92 PUR NS 335 (1952); Re Pacific Telephone & Telegraph Co. (California), 53 PUR 3d 513 (1964).

   84. We find that respondents are in error in their interpretation of the holding in the Hope case. The comparable earnings test as applied by respondents, particularly with respect to unregulated industrial enterprises, is neither the principal nor the only test which the Supreme Court has directed the Commission to apply. The courts have made it clear that the result reached, not the method or formula used, is the controlling factor. FPC v. Hope Natural Gas, supra; FPC v. Natural Gas Pipeline Co., supra; State of Wisconsin, et al. v. Federal Power Commission, supra.

   85. Our rejection of the contentions of respondents in this respect should not be construed as a determination that their testimony on the subject is

considered irrelevant or that we have ignored it in reaching our determination herein. To the contrary, we have considered and evaluated it in the context of the entire record. We merely find here that it is not the mandatory standard which must be applied in fixing a rate of return.

(2) Capital Structure

   86. A further basic contention is presented by respondents with respect to the matter of capital structure or debt ratio, and repeated with respect to the issue of accelerated depreciation. On oral argument, respondents' counsel stated:

   As I said, I think the Commission's function here is to examine a debt policy that we follow, examine the reasons for it, but unless you find that we have abused our discretion or have been imprudent, I don't believe you should disturb it. (Tr. 10365.)

   87. Counsel further contended there was insufficient testimony to prove that respondents' policies had been imprudent, apparently inferring that this Commission is prohibited from exercising its judgment on this issue but is limited to the judgment expressed in the record by witnesses. On brief, essentially this same position is asserted, and, on the basis thereof, respondents conclude:

   88. And unless the Commission can find imprudence, it would be wrong both as a matter of law and sound policy for the FCC to penalize A.T. & T. shareholders by reducing the allowable return. (Bell brief on rate of return, p. 61.)

   89. We agree that this Commission is not the manager of respondents' business. It is neither our obligation nor duty to dictate the business policies and practices to be followed by management. On the other hand, we have the statutory responsibility for the establishment and maintenance of just and reasonable rates and, in the context of this proceeding, to fix such rates for the future. If we are to discharge this responsibility in a meaningful manner, we must be free to examine fully all matters affecting the future level of rates and to make judgments within a framework prescribed by the public interest rather than management policies and predilection. In other words, we are not limited to acting in situations in which we have first found abuse, [*57] imprudence, or indiscretion on the part of management in the past. A recent decision of the Louisiana Supreme Court, n29 for example, points out that following the Hope decision, 17 State regulatory commissions, in addition to Louisiana, have adopted hypothetical debt ratios in determining a proper rate of return.

n29 Southern Bell Teleph. & Teleg. Co. v. P.S.C. (Louisiana, 1960), 32 PUR 3d 1.

   90. We, therefore, reject respondents' contention that we are limited in our judgments to those expressed in this record, or prevented, as a matter of law,

from considering the effect of the existing capital structure on the rate of return we are required to establish. As stated by the U.S. Supreme Court, the ratemaking function involves pragmatic adjustments by the Commission seeking to determine "just and reasonable" rates. Federal Power Commission v. Hope Nat. Gas Co., supra.

C. Detailed Positions

(1) Requested Rate of Return

   91. John J. Scanlon, vice president and treasurer of A.T. & T., was its chief spokesman on the issue of fair rate of return. He is generally responsible for financing, and for the company's relations with investment bankers, investors, and security analysts. Scanlon's testimony is summarized by respondents in their brief on rate of return, pages 8-9, as follows:

   Bell's principal rate of return witness was John J. Scanlon (Bell exhibit 20). In order to ascertain the earnings level needed to maintain A.T. & T. stock as a comparable investment alternative, Scanlon analyzed the earnings on equity of 528 manufacturing companies over the periods 1946-65, 1952-65, and 1959-65. The study showed that equity earnings of these companies manifested strong central tendencies of about 10 to 12 percent for the three periods

studied. The average for all companies was even higher. As a further check, Scanion made a similar study of the equity earnings of 128 electric utility companies and found their equity earnings to be reasonably comparable to nonregulated companies. Any difference in business risk between manufacturing companies, telephone companies, and electric utilities tended to be equalized by differences in capital structure typical in the respective industries, so that the investment risk of the equity owner in each case was quite comparable. Finally, in order to allow for possible residual differences of equity risk. Scanlon said that Bell should be allowed to earn between 10 and 11 percent on its equity -- the lower end of the central range of manufacturing earnings.

   92. Scanlon's conclusion was supported by Dr. Walter A. Morton, who appeared as a witness for Bell, and by Dr. J. Rhoads Foster, who appeared for the USITA. These witnesses considered the proper test of a fair return to be primarily the earnings achieved by industrial manufacturing concerns, and, secondarily, by regulated electric utilities. The maintenance of credit, financial integrity of the enterprise, and ability to attract capital were also related by them to the results of the comparable earnings approach. Morton and Foster characterized their approach as the determining of the "opportunity cost" of capital. Each of them stresses the role of judgment in reaching their conclusions.

   93. Morton contended his opportunity cost of capital conforms to all of the standards of the Hope case, but the further contended it would [*58] be unreasonable to require comparison of A.T. & T., which he considers unique, with any other enterprise. He, accordingly, determined relative risk of the Bell System in relation to the industrial group in his study by recourse to market price measurements. The ratio of market price to book equity, he said, is a measure of the market value placed upon a given dollar of net book equity assets and that relative market to book ratios indicate the investor judgment. His study for the period 1952-65 indicated A.T. & T. had an earnings to book ratio of 9.1 percent and a market to book ratio of 151 percent. Moody's 125 industrials, on the other hand, had an earnings to book ratio of 13.1 percent and a market to book of 200 percent. He then found the ratio of these two market to book ratios (151 and 200 percent), or 1.44, which he multiplied by the earnings to book ratio of the industrials, or 13.1 percent, and arrived at 12 percent, which he designated as the gross earnings equivalent. Deducting A.T. & T. earnings to book of 9.1 percent from this 12 percent, he arrived at his incremental adjustment of 2.9 percent. He then made the judgment determination that the actual adjustment necessary was two-thirds of the incremental adjustment, and reduced this 2.9 to 1 percent, which he described as the net earnings equivalent. Deducting this from the gross earnings equivalent of 12 percent, he arrives at 11 percent as the earnings equivalent. As this is 2.1 percent below the industrials earnings to book ratio, he utilizes this figure as the required differential. After adding the 2.1 percent to A.T. & T. earnings to book ratio of 9.1 percent, he derived 11.2 percent as the required earnings on equity. He finds this to be "in the area of 10 percent" and adopted the latter figure as the required return on equity for A.T. & T. as a matter of opinion.

   94. Dr. Irwin Friend, a witness for respondents, provided a study of the cost of equity capital to A.T. & T. by determinatives of investment capitalization rates, sometimes characterized as the "discounted cash-flow" method. This method attempts to equate investors' expected dividends in the future, plus the market price they ultimately expect for their shares, to the present market price. It necessarily assumes a constant earnings-price or price-earnings ratio, and the growth rate utilized, according to Friend, cannot be ascertained with precision. Further, although the A.T. & T. dividends are currently at a payout ratio of about 60 percent, he assumed investors anticipate a 65-percent ratio. He concluded that a reasonable expectation of investors, based on an average price-earnings ratio of 18 1/2 between 1958 and 1965, is for a price-earnings multiple of 18. He further found an annual average growth in earnings per share of 5 percent. Combining these factors with a 65-percent payout, he concluded a capitalization rate of 8.5 percent is anticipated. With proceeds of A.T. & T. shares sold at an estimated 10 percent below market, Friend concluded that the cost to the company was 9 percent. Applying judgment, he utilized 8-9 percent for the range of the cost of equity to A.T. & T. He did not attempt to determine an overall rate of return. He subsequently indicated the 8-9 percent was related to market values, and that, in relation to book value, it would be something over 10 percent.

   95. Foster, on behalf of USITA, discussed the market value method of analysis and the comparative earnings method of determining cost [*59] of

equity capital and found that neither could be taken as an exclusive and reliable formula. He, accordingly, applied the tests of each and found they produced consistent results. While not advocating explicit recognition of inflation, he found that a return on equity of 10 to 10 1/2 percent on book cost

to be consistent both with the fact of past inflation and its effects on earnings and to returns on alternative equity investment opportunities. With a 10-percent return required on equity, and with a capitalization consisting of two-thirds equity and one-third debt (at 4 percent), he determined that an 8-percent rate of return is reasonable and that respondents should be permitted to earn within a range of 7.5 to 8.5 percent. He urged that determination of the debt ratio be left to management. He also advocated that the interstate operations, because of participation therein by independents, should be permitted a return in the range of 7.75 to 8.75 percent. Respondents reject this contention on the grounds that the earnings on both interstate and intrastate business should be the same; i.e., from 7 1/2 to 8 ½ percent return advocated by them.

   96. Telephone Shareholders, Inc., participating as a nonparty under section 1.225 of our rules, offered Frank D. Chutter, of the staff of Massachusetts Investors Trust, as a witness, Chutter advocated a present rate of return of 8 to 8.3 percent. He further proposed that this rate of return should be allowed to increase annually by an unspecified amount. He determined his proposed return by reference to the earnings of the electric utilities in the MIT portfolio.

(2) Economic Outlook

   97. Respondents tendered broad economic support for their position through Dr. Paul W. McCracken, Robert R. Nathan, and Alexander Sachs; Dr. Leon H. Keyserling, who was tendered by USITA to testify in connection with another matter, also presented evidence in this area; Dr. Sydney Weintraub, a consultant called by this Commission, testified as to the general economic outlook and its relation to respondents.

   98. These economic witnesses were in general agreement that the Bell System could expect continued future growth and expanded markets consistent with its experienced results in the growing economy since the end of World War II.

   99. Respondents' position regarding their participation in the post-World War II growth economy is best delineated by their summary:

   In the postwar period the system grew at nearly double the pace of the economy in general and in 1965 nearly 2 percent of the total gross national product originated within the Bell System. A basic element in the system's postwar performance has been the incorporation of technological innovation in a heavy and continuous investment program. Total new investment in plant and equipment in the years 1950-65 was nearly $35 billion, which is 7 percent of all corporate investment. The system's capital expenditures have been equal to about 36 cents for each dollar of GNP originating in the company, which is double the national average, and the system's stock of capital assets represents about 4 percent of the total productive facilities employed by American business today. Bell System research and development exceeded $2 billion in the 1950-65 period and reached merely $300 million in 1965 alone (Bell's proposed findings, R.R. 18).

    [*60] 100. Respondents expect a future economy growing at an even more rapid rate than the substantial postwar growth already achieved.

According to them, communications will necessarily be importantly involved in this growth, and many demands are expected to be made upon the Bell System to meet these needs. The independents expect the general level of profitability of the economy to be higher in the future than it is today.

   101. Respondents emphasize that they do not ask regulation to protect them from risk or to assure their earnings. Neither, according to respondents, do they now seek to raise their level of earnings by a rate increase. They do seek to improve their earnings through innovation and improvement in operating efficiencies. In this way, they expect to reach the higher level of earnings they claim they must have to participate fairly in our changing economy.

(3) Investors' Expectations

   102. Respondents allege that their obligation to meet the growth needs of an expected expanding economy will require them to raise large sums of money. In doing so, respondents claim they will be in competition with other enterprises for the investor's dollar. To meet this competition, they urge that they will have to have earnings, or the opportunity to earn on the book value of their common stock equity, at a rate equal to that of other companies.

   103. Respondents offered testimony of six active or retired officials of financial institutions as to the earnings requirements of A.T. & T. from the standpoint of the expectations of institutional and other investors. They support, in general terms, respondents' requested return. These witnesses were: John H. Moller, of Merrill Lynch, Pierce, Fenner & Smith; Gustave Levy, of Goldman, Sachs & Co.; Charles W. Buek, of United States Trust Co.; F. J.

McDiarmid, Investment Department of the Lincoln Metropolitan Life Insurance Co.; Adrian M. Massie, Trust Committee of the Chemical Bank-New York Trust Co.; Alexander Sachs, Lehman Corp.

   104. These financial witnesses offered opinion testimony of a general nature. Five of the institutions represented have a close relationship with the Bell System and thus the witnesses cannot be considered as disinterested. These relationships include substantial ownership of A.T. & T. stock, officials and directors holding positions with both Bell companies and the financial institutions, management of Bell pension funds, and extensive participation in

Bell System financing.

   105. These financial witnesses are uniformly of the view that, in the post-World War II period and until about 1958, investors were more concerned with stability of earnings and dividends than they are today. The witnesses differ sharply, however, as to the popularity of A.T. & T. stock during that period. For example, one witness considered A.T. & T. stock almost like a bond during that period and stated that his firm purchased more A.T. & T. stock for its clients during that period than it has in recent years. Another considered A.T. & T. stock had fared badly during that period. J. H. Moller, from Merrill Lynch, Pierce, Fenner & Smith, the institution that has the largest single block of A.T. & T. stock in its name, over 7 million shares, was critical of [*61] A.T. & T.'s standing in the financial community during this period. This testimony, however, must be received in light of the statements made in the published reports of his firm during this entire period. Its regular publication consistently rated A.T. & T. stock as a desirable investment producing liberal income. Moreover, the A.T. & T. stock held in the name of his firm has increased every year since 1945.

   106. A Merrill Lynch wireflash downgrading A.T. & T. stock and recommending against additional purchase was issued immediately upon the announcement of the FCC investigation but without its having seen the Commission's order. Moller did not know of any instance where a similar wireflash was sent. Moller admitted that this wireflash could have had a depressing effect on the market price. While this flash has not been rescinded, it is pertinent to note that the number of shares held in their name for customers has continued to rise. Moller, like other witnesses of this group, does not consider assurance of dividend and price stability to be important investment considerations.

   107. The witnesses were unanimous in the view that since the later years of the 1950's investors, particularly institutions, have become more concerned with obtaining investments that show growth in earnings per share than with stability. They express the view that A.T. & T. must have growth in earnings comparable to that of other companies, in order to compete for new equity capital. We note that the shares of common stock held by institutional investors have been growing slowly, but even now represent only about 15 percent of the market value of all common stock.

   108. Respondents argue that the experience they have had with investor reaction confirms these judgments. Respondents allege that, in the period 1946-57, their earnings, dividends, and market price per share lagged behind manufacturing and other public utility stocks and that investors were disillusioned by this record of comparatively low earnings and lack of dividend growth on A.T. & T. stock.

   109. During 1946-49, A.T. & T. earnings were about 7 percent on equity, which respondents characterize as "quite poor." This, they say, prevented recourse to straight equity financing. A.T. & T., therefore, raised $1.1 billion through convertible debentures which, while bearing the low interest rates characteristic of the period, were convertible into common stock generally below market price and, as to one issue, below the book value per share. From 1951-55, $2.1 billion more of debt was offered, convertible into common stock below both market price and book value. These latter issues, when converted at a premium, produced over $3 billion, at an average of about $5.50 per share less than book value. Thus, during a period of what respondents' witness Gustave Levy characterized as "woefully low earnings" A.T. & T. raised almost $5 billion; $3.2 billion in cash from sale of debentures, $1 billion of cash premiums from the conversion of debentures into stock, and over $600 million from sales of stock to employees. During the same period (1946-55), over $3 billion in straight debt was also raised, for a total of about $8 billion of new capital.

   110. Today, respondents allege, they face a situation in which "supplies of cheap debt money no longer exist"; fixed-dividend, static-earning stocks no longer appeal to the investor; and investors, having [*62] had experience with inflation, demand protection through growth in earnings, dividends, and market price. In the face of this, they say, it is necessary to look at comparable investment alternatives to determine the level or earnings which respondents should have.

(4) Capital Structure

   111. We have already set out, in paragraph 91, supra, a summary of the method used by Scanlon to arrive at the comparative earnings figure which respondents urge we adopt in this proceeding. "Despite the obvious differences in risk * * *" between electrics and manufacturing, respondents note that they show comparable earnings on their respective equities. This equalizing of earnings is attributed by respondents to the leveling effect which is brought about by the difference among them in capital structure. The allegedly higher risk manufacturers have low debt, averaging 15 percent. As a result, respondents say, their investment risks turn out to be comparable. Respondents argue that their risks are less than those of manufacturers but greater than those of electrics, and because of the Bell System's 30-40 percent debt ratio, which falls between the ratio of the manufacturers and that of the electrics, Bell's investment risks on equity are close to those of the average of the central tendencies of 578 manufacturers and 128 electric utilities used by Scanlon in his analysis. Respondents further contend that if they had a higher debt ratio their relative risk as an equity investment opportunity would increase and they would, therefore, need a higher return on equity to be competitive with the other comparable investment opportunities.

   112. Respondents state that under a policy established in the early 1920's the traditional capital structure for the Bell System consolidated is made up of 30-40 percent debt and 60-70 percent equity. As of 1965, the ratio was 32 percent and respondents decided after this proceeding was under way to move their debt ratio to the upper limit of this range, i.e., 40 percent debt. It is expected to take until 1970 to reach the 40 percent debt level. At that time, they intend again to consider whether the ratio should go higher. For the purpose of this proceeding, they have used the hypothetical ratio of 35 percent debt as the basis for computing their required rate of return.

   113. In 1945, the consolidated system had a 31-percent debt ratio. It rose to 54 percent by 1949. A substantial part of this was in the form of convertible debentures and the debt ratio thereafter declined as conversion took place. The ratio has not been as high as 40 percent since 1953, nor as high as 35 percent since 1963. Respondents allege that because of the postwar requirements for new capital, "A.T. & T. was forced to finance its expansion by floating debt issues, including convertible bonds, until it reached approximately 54 percent debt." Respondents admit that there was an ample supply of funds available for new corporate debt. They chose, however, to issue debt in a form convertible into the higher cost equity capital.

   114. When their principal rate of return witness was asked why respondents did not utilize debt rather than equity financing when interest rates were low, while it is now committed to increasing its debt ratio with interest rates at much higher levels, the matter was [*63] characterized as an example of perfect "20-20 hindsight" and no responsive answer to the question was given.

   115. Respondents defend their capital structure by saying:

   Operating under its present policy of a 30-40-percent debt ratio, the Bell System has been able to provide its subscribers with high quality services at reasonable rates and to attract on equitable terms the vast sums of capital needed, while maintaining a high credit rating (Bell's proposed findings, R.R.

140).

(5) Risk

   116. The reasonableness of respondents' capital structure is further defended on the basis of the comparative risk between their business and others.

   117. Respondents' chief short-term risk is said to be that sufficient revenues may not be generated to produce an adequate margin of earnings after provision for expenses and taxes. Factors affecting Bell's short-term risks are claimed to be inflation, as a consequence of its high fixed costs, relatively inflexible price structure, and large labor costs. There also exists, they say, the danger of discontinuance of service by customers struck by economic adversity, based on the experience of the depression of the 1930's.

   118. As respondents see it, their short-term risk falls between that of the manufacturers, which admittedly have greater risk, and the electrics, which they contend have lesser risk. This risk factor, as it affects capital structure, leads respondents to conclude: "These differences in risk are reflected in differences in debt ratio and in overall earnings requirement. The greater the instability of earnings, the less the enterprise is in a position to finance with debt."

   119. According to respondents, "The telephone industry is faced with serious long-term risks of the loss of earning power through technological or market changes." Respondents, it is claimed, are a capital intensive industry with most of their plant committed to a single purpose and not adaptable to other uses. They alleged that evidence of competition affecting the telephone industry and the Bell System has already appeared. It is claimed, without definite specification, that "many large and small users of communications provide some of their own equipment and services * * *." Perhaps to respond to the claim that none of these risks appear to have made their impact felt upon respondents' continued growth, expansion, and increased income, respondents conclude:

   The fact that these long-term risks have not destroyed or seriously impeded the telephone industry up to this time does not mean they do not exist. No one can envision how the need for communications will be met in the future. Many of today's extinct or dying industries were thought at one time to be monopolies with excellent prospects for the permanent future. (Bell's proposed findings, R.R. 144.)

   120. This statement is in sharp contrast with the testimony of respondents' witnesses Gilmer, Baker, and Nathan, all of whom projected a healthy, growing, and important role for communications in general and respondents in particular.

   121. Without supporting factual data, respondents urge that their riskiness as a business justifies their lower debt ratio as compared with [*64] electric utilities or independent telephone companies. While a number of respondents' witnesses expressed the view that the Bell System is more risky

than the electric utilities, none was able to substantiate his views other than by unsupported generalizations with regard to discontinuance of telephone services and bankruptcies of other businesses experienced during the major depression in the 1930's. The argument is that because respondents are riskier than the electries they need a lower debt ratio. As proof of the truth of this, respondents point to their lower debt ratio -- ignoring the fact that they themselves have established it.

   122. For the period 1960-65, respondents realized an overall rate of return averaging 7.6 percent, while the electric utilities realized only 6.4 percent. Notwithstanding this lower rate of return, the electrics earned 11.7 percent on equity while A.T. & T. earned only 9.3 percent. We note that despite our request in an order setting oral argument, respondents have failed to explain satisfactorily this significant difference.

   123. Respondents submitted an evaluation of the Rell System's operating risks by a management consultant, based on a profit equation in which

revenues are related to pretax profit margin, variable costs and fixed costs, and a margin of safety factor developed. This related the Bell System to the electric utility and manufacturing industry as a whole. The basic factor in the study is variable costs, and all others are considered as fixed. The witness did not study the Bell System to determine this key factor but rather accepted respondents' identification of certain broad accounts for this purpose. Further, an FCC staff witness applied the formula to one specific manufacturing industry with a result opposite of that found for the group by the witness. Finally, another Bell witness, Dr. Morton, testified it is not possible to measure operating risks by the use of this method. Accordingly, we give no weight to the results of this study.

(6) Inflation

   124. Respondents presented a number of witnesses on the subject of the effects of inflation on the Brll System and, in particular, on its rate base.

   125. Richard Walker, of Arthur Anderson & Co., advocated the restatement of the original cost rate base in "current dollars" to offset erosion of capital caused by inflation. Current dollars are adjusted for decreases in purchasing power in relation to a selected base year.

   126. Dr. S. L. Bach presented a general discussion of inflation and recommended it be accounted for in this proceeding. He considers regular depreciation allowances inadequate and inequitable. He alleged, however, that it is possible for the Bell System to "defy" the forces of inflation.

   127. A. R. Tebbutt discussed various price indices, including the telephone price index (TPI) developed for A.T. & T., and how they might be used to

translate historical original cost of plant investment to 1965 dollars.

   128. Finally, John L. Boggs, A.T. & T. cost engineer, constructed an average net investment for the Bell System in revalued 1965 dollars. [*65] On the

basis of Tebbutt's three indices, Boggs found that net investment would be increased by 11 to 14 1/2 percent and annual depreciation expenses by 12 to 18 percent.

   129. A group of State commissions offered a witness, M.W. Van Scoyoc, who took issue with the suggestion that the rate base or rate of return should be adjusted for inflation. He pointed out that a cost of capital approach to rate of return automatically takes into account whatever inflation that has taken place.

   130. On brief, respondents contend that inflation should be taken into account in fixing a fair rate of return on an original cost basis. No method is

proposed, however, for adjusting the return to make such an allowance, but inflation is said to be an additional reason for adopting the comparable earnings standard proposed by respondents.

   D. Other Positions and Data

   131. Respondents' position on rate of return and their capital structure and financial policies was questioned in several important aspects by evidence adduced during the hearing. This was done, principally, by a witness offered by the regulatory commissions of the States of West Virginia and Oregon; by another witness offered by another group of State utility commissions (Iowa, Washington, Oregon, and California); as well as by FCC staff members and consultants.

   132. The West Virginia and Oregon witness, Dr. R.M. Robertson, concluded that a maximum of 7 percent is a fair rate of return for respondents. Robertson used essentially a cost of capital approach. He estimated the actual cost of capital to respondents at various time periods, characterized as historical and current. For historic cost of capital, he concluded the weighted cost to be 6.93 percent and, for current cost, 7.26 percent. These figures were averaged to produce 7.1 percent.

   133. As his next step, Robertson made allowance for income tax savings estimated on the interest expense, arriving at 6.31 percent for historic and 6.4 percent for current cost.

   134. In each case, Dr. Robertson averaged the historic and current results reached, i.e., 7.1 percent, without adjustment for taxes, and 6.36 percent, with adjustment for tax savings. Thus, he arrived at his recommended fair return figure of not more than 7 percent.

   135. M.W. Van Scoyoc, an expert on engineering and accounting aspects of regulation, testified for a group of State commissions on the subject of rate base and depreciation allowances. In addition, he took issue with the comparable earnings method used by Scanlon as not comporting with the language of the Hope case. He contended that the best measure of return to the equity owner is the rate of return investors have been willing to accept in the past, are currently accepting, and are likely to accept in the future. He offered comparative data, since 1920, as to earnings on common stock equity of A.T. & T. and regulated and unregulated enterprises, earnings per share, and market price per share, but did not determine a fair rate to return for respondents.

    [*66] 136. Dr. Lionel W. Thatcher, an FCC consultant, utilized a cost of capital approach which he said will enable the Bell System to retain and attract capital. He rejected the comparable earnings standard relied upon by respondents as not recognizing the importance of market price, which according to him, seldom bears any consistent relationship to book value. He pointed out that the industrials, with which Dr. Morton makes comparison, have market values ranging from 2,200 down to 37 percent of book value. Further, he said that standard poses an almost impossible problem of risk measurement.

   137. Thatcher concluded, on the basis of his analysis, that a fair rate of return for respondents would be a figure between 6.75 and 7.25 percent. He presented data on A.T. & T. and electric utility earnings-price and dividend-price ratios. He also listed four important factors that should be considered in determining investors' expectations of growth: (1) The rate of increase in the number of shares; (2) the rate of increase in dividends per share; (3) ratio of dividends to net proceeds on common stock issues; and (4) the ratio of surplus per share to dividends per share.

   138. Thatcher concluded that there is no significant difference in risk between the Bell System and the electric utilities, and that the Bell System could increase its debt substantially and still have ample interest coverage. Utilizing four different methods of estimating, Thatcher arrived at costs of equity for A.T. & T. ranging between 7.93 and 8.4 percent. He adopted, as a matter of judgment, 8.65 percent which, when weighted with the capitalization ratio of 35 percent debt and 65 percent equity, resulted in a return of 7.01 percent. Using a 45-percent debt ratio, and adjusting interest and equity costs upward, he reached an overall cost of capital of 6.77 percent. He, accordingly, recommended a rate of return between 6.75 and 7.25 percent.

   139. Dr. Myron J. Gordon, also an FCC consultant, utilized a unique approach using an econometric model, or price equation. According to Gordon, this model established the relationship between market price, dividends, growth of dividends, the leverage or use of debt financing, and the stock financing rate. Coefficients were estimated on the basis of a sample of 49 electric utility companies, which Gordon found had no lesser risk than respondents. Gordon found that the annual increase in respondents' total assets had been about 8 percent in recent years and assumed the Commission would authorize continuation of that rate of growth. At that rate of investment, he found 7-7 1/4 percent to be the required rate of return.

   140. Gordon based his study on his concept of the objectives of public utility regulation. Within this framework, he applied economic principles to available data and constructed the model. This model represented a combination of the factors that Gordon considered important to the determination of a rate of return. His recommended rate of return was computed by use of this model.

   141. Gordon testified that:

   The objective of the agency is to serve the consumer, and that objective is realized by setting the price at the lowest level consistent with securing the investment by the utility in replacing, modernizing, and expanding its [*67] capacity that the public requires. The utility management in turn determines its investment with the objective of serving its stockholders. Since the stockholders want the price of a utility stock to be maximized, in deciding whether or not to undertake an investment, a utility management asks whether the investment will increase (at least maintain) the price of its stock. Therefore, the criterion an agency should employ in setting the price of the product is that the rate of return the utility earns on investment will be high enough so that the desired investment and financing by the utility will not depress the price of its stock. (FCC staff exhibit 17, pp. 5, 6.)

Gordon also said that:

   A useful background for arriving at the rate of return A.T. & T. should be allowed is provided by comparing the rate of return, investment rate, financing policies, and share yields of A.T. & T. and a representative sample of electric utility companies. (FCC staff exhibit, 17, p. 10.)

   142. Gordon pointed to the adverse effects of the relatively low ratio of debt in respondents' total capital structure upon the earnings realized by their stockholders within any given rate of return. He noted that the 49 electric utilities he studied had a debt-equity ratio of 1.5, whereas the Bell System average was 0.50 (these refer to the ratio of debt to equity rather than of debt to total capital, and in the latter terms equal 60 and 33 1/3 percent, respectively). As a result the electrics enjoyed earnings on equity of 11.7 percent, while A.T. & T. had earnings of only 9.3 percent, although the average rate of return of electrics on total assets was only 6.4 percent as against 7.6 percent for respondents. His recommendation contemplates that respondents will change their financing policy so that they will obtain new capital by issuing $2 of debt for each $1 of equity until the overall ratio is about 1 (50 percent debt). He found no credible evidence to support the thesis that A.T. & T. earnings are more unstable than those of the electric companies, and stated that if any difference in business risk does exist, its contribution to the rate of return required by investors cannot be material.

   143. Respondents offered testimony in rebuttal to the Gordon presentation. Respondents' witness Scanlon questioned Gordon's recommendation that respondents move to a 50-percent debt ratio by borrowing $2 for each $1 of equity raised until that ratio is reached. He also disagreed with Gordon's assertion there is no credible evidence to show A.T. & T. earnings are more unstable than the electrics and that any differences in business risk do not contribute materially to the investors' required return, but offered no evidence in support of his opinion. At the same time, Scanlon announced that respondents intended to advance toward the upper end of the 30-40-percent range and consideration would then be given to whether a higher range would be appropriate. It would thus appear that Scanlon has in effect adopted Gordon's recommendation that the debt ratio of the Bell System should be increased substantially.

   144. Friend, who also testified for respondents on rate of return, criticized Gordon's study. He called the approach imaginative but seriously deficient and leading to erroneous results. He contended Gordon underestimated the growth rate in earnings expected by investors in electric utilities. Friend preferred to use techniques of averaging a series of broad growth estimates given by investment [*68] houses interested in the promotion and sale of securities. He contended that both approaches to total cost of capital employed by Gordon should be adjusted upward to 8 percent. He listed six

limitations in Gordon's econometric model.

   145. Additional criticism of Gordon, of detail rather than substance, was offered by Dr. John W. Tukey, of the Bell Telephone Laboratory. Tukey contended that Gordon's use of algebraically expressed models with statistically assessed coefficients exceeds previous use of such models. He distinguished between models used as guides to decision-makers, and a model predicting investor reaction to A.T. & T. policy decisions. He listed eight conditions which he believes should be met to remove uncertainty and indefiniteness from the Gordon model. Tukey also testified that a key computation in Gordon's equation, which led to his conclusion 7 percent is the required rate of return, was inaccurate due to being carried only to two stages of iteration. Tukey supplied a 10-stage iteration, which produced results different from Gordon's and led him to be uncertain as to its accuracy.

   146. Another consultant to the Commission, Dr. C. West Churchman, was called to comment on Tukey's criticism of Gordon. Churchman, an expert on operations research and management science and authority on the philosophy of decision-making, addressed himself to the eight criteria of criticism Tukey had leveled at Gordon's model, and found them invalid. He discussed the interplay between the manager and the management scientist in the development of a model, and the need for any critic, who feels a flaw exists in a model, to assess the flaw in terms of its effect on the manager's decision and not merely as a technical deficiency. He pointed out that if Tukey's suggestion for a modification of the model results in only a slight change, it is correct to ignore the suggestion. Churchman finds that most models are susceptible of improvement. He urges that the manager and the scientist share the task of improving the mathematical models.

   147. Gordon stated that the 10-stage iteration suggested by Tukey was a refinement which led to the conclusion that the rate of return should be lowered to 6.75 percent from the 7 percent recommended, but Gordon did not consider this refinement necessary. Gordon also revised the criterion of his model to meet part of the Tukey criticism, and as a result concluded that the required return would fall between 7 and 7 1/4 percent rather than at the 7-percent level

originally proposed.

   148. Subsequently, Tukey stated that Gordon's revised model was an improvement. Tukey was not able to make a definitive judgment of the adequacy of the analyses made by Scanlon, Friend, or Morton on behalf of A.T. & T., because he had not evaluated their testimony and dwas not asked to do so. Tukey also stated he could contribute to the construction of a helpful model, with economic and financial assistance which he assumes is available within the Bell System.

   149. The FCC staff placed in evidence various factual data, in the light of respondents' reliance on the comparable earnings standard and its earnings per share comparisons. These data show various growth factors for the period 1948-64 for A.T. & T. and a few selected industrials and utilities. According to these data, A.T. & T. realized greater growth in net income, 647.3 percent, than any of the [*69] industrials or electrics used for this comparison, and second greatest growth of revenues, 292.6 percent. However, it also had the greatest growth in number of shares of common stock, 278.1 percent, as against 3 to 22 percent for the six industrials and 52 to 181 percent for the electrics. This very great increase in the number of shares, both absolutely and in relation to other companies, had a serious adverse effect on the earnings per share of A.T. & T., so that growth in earnings per share, 97.8 percent, was well below the leading industrials and also lower than one of the utilities, 123.2 percent. Other data indicated that non-Bell telephone companies have debt ratios ranging between 47 and 53 percent, depending on the group studies, with preferred stock accounting for an additional 7-8 percent of senior capital. Thus, the common stock equity for the non-Bell companies ranges between 40 and

45 percent.

   150. In 1965, the capital structures of the 25 largest electric companies, which are typical of the entire industry, included from 40 to 65 percent of senior capital obligations. FPC statistics show, for all reporting electric companies for 1964, 51.8 percent of long-term debt, 9.6 percent of preferred stock, and 38.6 percent of common stock equity. Bell System net income was sufficient to cover interest charges 6.4 times, whereas the coverage for FPC electrics was 2.96 times and for non-Bell telephone companies ranged from 2.29 to 4.3 times. Bell System bonds during the period 1945-48, when the

Consolidated System debt ratio of respondents rose to 54 percent, continued to enjoy Moody's ratings of AA and AAA.

   151. A.T. & T. instructions to the trustees of its pension funds authorize investment in bonds rated as low as A and with earnings coverage of fixed

charges of 1 1/2.

   152. Other staff data demonstrated that, despite rate reductions and jurisdictional separation changes which added interstate revenue requirements, interstate earnings of respondents show quick recovery and continued upward growth. Finally, data were presented on the rates of return approved by various State and Federal regulatory commissions. In Bell System cases, the range of return approved since 1960 was from 5.79 to 6.67 percent; for non-Bell telephone companies, 5.94 to 6.80 percent; for electric utilities in State cases, 5.51 to 7.25 percent; and for electrics before FPC, 6 percent.

   153. From 1946 to 1964, the electric utilities increased gross plant by $47.5 billion, while the Bell System increased its gross plant by $26.2 billion. The additional capital raised by the electrics in that period consisted of $8.9 billion from the sale of common stock, $2.6 billion from the sale of preferred stock, and $18.5 billion of long-term debt. In contrast, the Bell System raised $10.8 billion in common stock, none by preferred stock, and $8.7 billion of debt. The ratio of senior capital (debt and preferred) for the electrics was 61.3 percent in 1946 and 61.4 percent in 1964. Thus, the electrics raised nearly three times as much long-term debt as the Bell System between 1946 and 1964, while starting the period with a debt ratio of over 50 percent.

   154. Respondents offered rebuttal testimony by R.A. Lovett, an investment banker (and former Secretary of Defense). The latter generally urged that matters of debt-ratio and accelerated depreciation [*70] be left to the discretion of management. Lovett found that the unusual dependence of the country on A.T. & T. gives the latter the attributes of a "public trust"; that it ranks with the armed services in its involvement in national security and is an absolute essential to national defense. He believed, however, that the electrics are less risky than A.T. & T. despite their higher debt ratios, their flow-through of liberalized depreciation for tax purposes, and lower rates of return. He did not know whether a 50-percent debt ratio would present a problem to A.T. & T., but he was "apprehensive."

   E. Liberalized Depreciation for Tax Purposes

   155. Section 167 of the Internal Revenue Code has provided, beginning with the taxable year 1954, alternative methods of computing depreciation allowance on a "liberalized" or "accelerated" basis, which, if utilized, would produce substantial reductions in the Bell System payments of Federal income taxes. The Bell System has elected to not to utilize these tax-saving provisions.

   156. Regulated utilities such as respondents are required by our system of accounts to compute and record depreciation charges as an item of expense on the basis of straight-line depreciation. n30 They are also required to record for expense purposes the actual Federal income taxes paid. Thus, a company using liberalized depreciation first computes the straight-line depreciation expense, for book purposes, but utilizes for tax purposes the liberalized or accelerated basis, which produces higher expense charges. This results in an increase in expense deductions and thereby reduces taxable income, with corresponding reductions in the income tax payment. The reduction in income tax expense in that instance is said to "flow-through" to the net income. This would automatically occur under our present regulations should respondents utilize liberalized depreciation for tax purposes.

n30 47 CFR 31.02-80 and 31.02-81.

   157. Some utilities using liberalized depreciation "normalize" the tax payment in the following manner. The utility records straight-line depreciation

for book purposes; it computes depreciation on the liberalized basis for figuring its tax liability. It then computes the theoretical higher tax liability it would have incurred if it had used straight-line depreciation. This difference between the actual and theoretical tax payment is then charged as an additional operating expense, and the amount thereof is placed in a special "normalization" reserve. Both the latter steps must, of course, be approved by appropriate regulatory authority, as they represent a departure from prescribed accounting practices.

   158. Respondents conceded that the use of accelerated depreciation with "normalization" is the theoretically desirable course to follow. Such a course would make available very considerable sums of interest-free funds to meet capital requirements and would not entail the risks which, they contend, result from use of liberalized depreciation with flow-through.

   159. Respondents alleged that eventually the difference in the taxes between those paid under liberalized depreciation and straight-line [*71] depreciation will have to be paid. If the savings are flowed-through now to users, future users, they say, will be charged with these payments, or if rates cannot be raised sufficiently at that time to offset the additional tax costs the stockholders would be required to absorb the increased tax payments. They argued that this is a risk they should not undertake and they, accordingly, decided to forgo the potential tax savings. They said, however, that if the

Bell System should utilize accelerated depreciation, normalization would not be permitted in many jurisdictions in which they operate. About half of the intrastate plant of Bell System is located in States in which commissions have compelled flow-through. The Bell System has, however, never made request for modification of our rules, discussed the problem with our staff, nor sought to obtain a uniform treatment of the problem through the NARUC, with which it

does discuss various regulatory problems.

   160. If the Bell System had utilized liberalized depreciation from the outset in 1954, the reduction in 1965 taxes would have been $223 million and the cumulative amount of reductions in taxes through those years would have been $1,577 million. This accumulative amount would have been the amount available as interest-free capital had accelerated depreciation with normalization been used. The respective amounts assignable to interstate operations would have been $63 million and $404 million, respectively.

   161. If respondents were to begin the use of accelerated depreciation with the year 1965, and assuming the continuation of the present annual $4 billion growth in plant until 1975, they would have a tax reduction of $29 million for 1965, with $8 million allocable to interstate operations; these amounts would grow to $281 million and $80 million per year, respectively, by 1975. The cumulative effect, or normalized reserve, if one were established, would be $2,206 million on total operations by 1975, of which $627 million would be allocable to interstate operations.

   162. Respondents' position was supported by several witnesses and by the independent telephone industry.

   163. All these witnesses objected to flow-through on the grounds it fails to recognize costs currently incurred, and that the Bell System could not achieve normalization should it elect to use accelerated depreciation. This support was based on opinions that respondents' present policy is a sound accounting practice devoid of risk; that there is the danger that regulatory authorities might require flowthrough of tax savings to income; that there is a possibility

of suspension or repeal of these tax provisions; that use of liberalized depreciation would distort reported earnings; and that management decision in this regard should not be disturbed.

   164. None of the supporting witnesses objected to accelerated depreciation with normalization, and some, like Nathan and Keyserling, urged its use with normalization. Dr. Paul W. McCracken, one of the Bell System general economic witnesses, testified that taking advantage of liberalized depreciation would be consistent with the views of economic policy which he advocated.

   165. William J. Powell, an accountant with the Federal Power Commission, with expertise in relation to this question, testified at the [*72] request of this Commission. According to Powell, whenever a company has a stable or growing plant, the use of liberalized depreciation results in a tax reduction, as there is no foreseeable future tax liability. Consequently, there is no need for a tax liability or normalization reserve.

   166. Powell further pointed out that savings lost to the Bell System by failure to use liberalized depreciation, computed by Bell witness Stott at $1,576 million, actually meant a loss in possible reductions in rates of $3,231 million, or nearly twice the amount of the tax savings because of the nearly two-for-one effect on gross revenues or rates.

   167. A witness for the California Public Utilities Commission, M. W. Van Scoyoc, was also of the opinion the Bell System should be utilizing liberalized depreciation and flow-through the resulting savings. He advocated that this Commission should impute the action for regulatory purposes whether or not the system does in fact do so.

   168. Van Scoyoc, as did Powell, presented evidence demonstrating the effect of utilizing liberalized depreciation. In general, he agreed with Powell and also pointed to the effect on consumers, whereby each $1 of tax savings represents about $2 in consumer rates.

F. Discussion

(1) Bell Capital Requirements

   169. Respondents stressed, in support of their comparable earnings presentation, that they are in competition with all other alternative investment opportunities in the manufacturing and utility fields. They said that investors no longer seek companies with stable revenues and dividends and are growth conscious, and that a company which cannot show steady growth will not attract capital. They also contended that investors seek future growth and

market appreciation, and that A.T. & T. stock should perform at least as well as these alternatives to compete effectively.

   170. Assurance of earnings plus growth is no doubt a desirable objective of the investing public. In fact, the investor would be happiest with complete safety and spectacular growth. However, the investing public is not, as the testimony of respondents' witnesses seems to imply, monolithic in its approach to the market. If all investors were disenchanted with stable, secure income, it would be impossible to sell bonds and no one would continue to own or buy securities which have a record of stable earnings and dividends. The record shows that this is not true. Vast amounts are raised annually through the sale of bonds, and many stocks which have relatively stable dividends and earnings records are traded daily.

   171. Investors' interests differ depending on their circumstances. Persons who are retired or depend on their investments for their income are interested primarily in safety and stability. They look more to the assurance of their dividend check and its regularity than to increases in price accompanied by risks of decreased prices or omitted dividends. Similarly, life insurance companies and pension funds still have the vast majority, over 80 percent, of their funds invested in bonds, mortgages, or other fixed return securities. Safety and stability are still the considerations that influence investors.

    [*73] 172. The record shows that respondents have, over the years, satisfied the desire of investors for safety. Since 1958, they have also increased dividends by approximately 50 percent, and per-share earnings have been steadily increasing. Almost uniquely among major corporations, respondents have never omitted or decreased their dividend. Further, if respondents had pursued a different financial policy with respect to capital structure, as discussed below, it is apparent their earnings on equity would have been much greater. Thus, respondents' heavy reliance on the higher-cost equity capital, much marketed at below book cost prices, has diluted the per-share earnings. One issue in 1956 on rights was sold at par value. Further, the large-scale options to employees to purchase stock in the company at prices very considerably below market (since 1958, an average of 31 percent discount) has had a depressing effect on earnings per share. For example, between 1946 and1965 over $2.8 billion of capital was raised by selling 86 million shares to employees at a substantial discount. As a result, there was a substantial dilution of the equity and a depressing effect on the earnings per share.

   173. We shall turn now to respondents' actual experience in the postwar years, which they characterize as a period of inadequate earnings. Nevertheless, respondents raised large amounts of capital in this period --1946-59. In all, respondents raised, through sales of bonds, of convertible debentures, and of stock to the public and to its employees, and from premiums paid to convert debentures to stock, the enormous sum of $13 billion between 1946 and 1959, or an average of nearly $1 billion per year. This tripled the system's capitalization.

   174. The record shows that, from 1946 through 1949, respondents raised over $3.5 billion in debt capital. Of this amount, some $1.1 billion was realized

through the sale of convertible debentures. During this period, over $300 million worth of these debentures, exclusive of the premium for conversion, were converted. Thus, during this period when A.T. & T. alleges it suffered from what it called inadequate earnings, holders of some $300 million of convertible debentures were willing to pay an additional $186 million to exchange their assured interest payments for their share of the "inadequate" earnings represented by equity.

   175. It is also pertinent to note that in this period of alleged inadequate earnings respondents were able to increase their total capital by about 75

percent. This is a much greater percentage increase than has taken place in any subsequent 4-year period. It took almost the entire 7-year period from 1951 to 1957 for a similar percentage increase to again occur. In the 7-year period 1959-65, when equity earnings averaged 9.6 percent and reasonably approximated the percentage respondents now allege they should be permitted to earn on equity, the rate of increase in total capital, premiums, and surplus was only about half as great as in the 1946-49 period.

   176. In the period 1951-57, earnings on equity ranged from 8.09 to 8.84 percent and averaged 8.4 percent, some 1.6 percentage points below what respondents allege they should now be permitted to earn. In these years, respondents raised about $5 billion in debt capital, of which over $2 billion was in the form of convertible debentures. In the same years, [*74] holders of over $2.5 billion of debentures were willing not only to forgo the safety of the prior claim of the debentures, but also to pay over $1 billion in premiums to acquire respondents' stock and have an equity position in a company whose earnings were about 80 percent of what respondents claim they should be allowed as a fair return on equity. Furthermore, these persons who converted their debentures were, or should have been, aware that, by their act of conversion, they were diluting the equity by increasing the number of shares and

reducing the potential earnings per share.

   177. Another factor which must be considered in evaluating respondents' contentions is that the respondents are generating increasing proportions of their requirements for new capital for construction without resort to the public market for debt or equity capital. In the period 1946-49, some 77 percent of respondents' requirements for construction was raised from the public and only 23 percent was generated internally from retained earnings and depreciation charges. In the most recent 4-year period, 1962-65, only 36 percent was raised from the public, whereas 64 percent was generated internally.

   178. In view of all of the foregoing, we find that the record demonstrates that respondents were able to raise vast sums of money representing much greater proportions of their total capital at earnings on equity 15 to 30 percent below what they allege is now needed. Furthermore, such sums were raised at a time when respondents followed a consistent policy of reducing the percentage of debt in total capitalization, thus diluting equity and substantially reducing earnings per share. At present, respondents' announced policy is the opposite. They intend to increase the ratio of debt to total capital from some 31 percent toward the upper end of the 30-40 percent range. This will lead to increased equity earnings per share by additional leverage. It will also decrease the revenues required to support a given level of earnings per share. This is so because each $1,000 of debt capital requires only $40 a year (on a 4-percent embedded cost of debt basis) in revenues after all expenses to pay for interest. On the other hand, each $1,000 of equity capital would require $192 of revenues after all expenses other than Federal income taxes to supply the 10-percent return on equity sought by respondents. Thus, earnings above $40 per $1,000 invested from debt capital increases the earnings per share of the equity holder. The fact that respondents were able to raise successfully billions of dollars at average returns on equity ranging from 7 to 8.4 percent, especially when equity in most of this period was being diluted, does not support respondents' claim that they now need a 10-percent return on equity. On the contrary, such experience does strongly suggest that an equity return of less than 10 percent would provide adequate capital protection and maintain respondents' ability to attract new capital.

   179. In summary, we find that regardless of the opinion testimony offered, the record demonstrates that respondents have in the past, and can in the future, raise the capital they require at returns on equity below 10 percent. Respondents are in a position, and now propose, to offer investors an opportunity for a substantial return on equity with a minimum of additional risk and, at the same time, provide adequate annual growth for the foreseeable future. This growth opportunity [*75] results from two factors. First is the growth in system revenues realized, ranging from 6 to 10 percent per year. To the extent that growth is financed out of internally generated funds, the increased earnings go entirely to equity holders. Second, there is the increased leverage which will result from heavier reliance on debt. Within any allowable overall rate of return, earnings on equity will increase steadily as the proportion of debt to total capital rises, pursuant to respondents' announced policy and our conclusions herein.

   180. We turn now to respondents' contention regarding the attitude of professional investors and the influence they have on the general investing public. Contrary to the opinion expressed by the professional investors who were witnesses on this subject, there is strong substantive evidence of the reaction of the investing public generally as to how they regard respondents' stock. Moller, an official of Merrill Lynch, testified on behalf of respondents. Insofar as is relevant to the specific question before us, he stated:

   (a) That Merrill Lynch served more small investors than any other brokerage firm;

   (b) That he knew more, therefore, about the desires and interests of such investors; and

   (c) That on the day the Commission announced its investigation, Merrill Lynch sent a telegram to its nationwide complex of offices and representatives removing A.T. & T. from the recommended "buy" list and that it had not yet restored the security to such list.

   181. However, the record shows that, despite this warning not to buy A.T. & T. stock and despite the uncertainty caused by the announcement of this investigation, the number of shares of stock of A.T. & T. that Merrill Lynch held on behalf of its investors actually increased from 7,309,279 at the end of

1964 to 7,423,995 at the end of 1965, and to 7,881,061 in August 1966. Under these circumstances, it appears that, whatever the influence of the professional investor may be generally, the stock of respondents is so highly regarded that investors not merely hold it, but seek, insofar as they are clients of Merrill Lynch, to increase their holdings substantially, even against professional advice and in the face of a pending proceeding which the professional investor felt could adversely affect the future earnings of the company.

(2) Risk

   182. Respondents' comparable earnings approach is premised on the relative risks encountered by its business as against others. It is a truism that the

greater the risk the greater the return must be, whether on debt or equity capital. This is particularly applicable in case adequate alternatives are

available. If one investment is less risky than another, it will attract capital at a lower price. If A.T. & T. were, in fact, riskier than any particular large manufacturing company, persons would not invest in A.T. & T. stock unless an opportunity to make a higher return were offered to them. Risks are generally considered in two categories: Long-term risks and short-term risks.

   183. Long-term risks are defined as the risks of permanent loss of earning power as a result of technological or market developments. Short-term or current risks are defined as those associated with the day-to-day management of the business. The principal short-term risk was [*76] described as the risk that sufficient revenues may not be generated to produce an adequate margin of earnings. We shall consider each in turn with respect to both manufacturing entities and electric utilities, the two groups with which respondents made most of their comparisons.

   184. Respondents' position is that the long-term risks are not very different for the Bell System as compared with manufacturing companies generally or electric utilities. Respondents' witnesses conceded that manufacturing companies have greater long-term risks from the standpoint of competition. They claimed, however, that utilities have offsetting disadvantages because of their rigorously limited franchises which inhibit diversification as well as entry into, and withdrawal from, markets; because their plant is longer lived and more specialized; and because they require greater amounts of plant investment in relation to revenues.

   185. Insofar as technological change is concerned, reference was made to developments "no one can envision." In addition, it was asserted that, in the future, there would be an accelerated pace of dynamic change in communications technology, which portends increasing risk to the Bell System. The nature of the risk, its scope or type was not specifically identified. Reference was made to canals, toll roads, manufactured gas, and street railways, as publicly regulated carriers which have suffered permanent loss of markets.

   186. There is no doubt that respondents, like other regulated utilities and carriers, are subject to franchise. However, a virtual monopoly in long distance interstate communication, as well as in intrastate toll and local exchange, in areas where they operate, which results in revenue in excess of $12 billion a year, can scarcely be characterized as a "rigorously limited" franchise. Respondents have not, it is true, diversified outside the communications field, but they have fully taken advantage of all opportunities to diversify within the filed. They provide local exchange service, toll service, leased channel service, service to television entities, service via leased satellite circuits and now are starting to provide the interconnection facilities in the data market which, according to some experts, will grow tremendously.

   187. The dangers of technological change should not be minimized. They have affected, and can affect, seemingly prosperous industries or utilities. However, the record shows that respondents are well protected, if not completely insulated, from any foreseeable danger. The major dangers are twofold. First is that the service provided will no longer be required. No one has suggested, or even hinted, that, in the foreseeable future, the need for any of respondents' services will disappear or even lessen. Instead, every witness agreed that need for respondents' communications facilities will not merely continue but will grow, at least as fast as the economy as a whole. Bell's major witness testified that there was no basis for assuming that the need for an interconnected nationwide telephone network will disappear.

   188. The second danger is that, although the need for communication service will continue, other more effective and efficient means, operated by different entities, will replace it. Here, too, respondents are protected if not completely insulated. The virtual monopoly of the railroads for long distance transportation of persons and materials, [*77] which has been invaded by buses, trucks, the passenger car, and the airplane, is cited as an example of such danger. On analysis, it appears that respondents have not, and do not, face such problems. In the provision of this service they use all the

alternative media available to provide communications service, i.e., open wire, carrier, cable, coaxial cable, high frequency radio, and microwave. Moreover, there is an inherent limitation on the extent to which radio, as a principal communications medium, is available by virtue of the relative scarcity of frequencies, and the licensing policies of the Commission reflecting such scarcity. Even insofar as satellite communications are concerned, their position appears secure. Respondents hold, as authorized by law, 25 percent of the stock of the Communication Satellite Corp. and will, therefore, share in the profits if Comsat is successful. They have a large share of the ownership of the earth stations used to provide satellite services, under a policy which indicates that their ownership will be related to their use of satellites, so that respondents will earn on their investment proportionately to their use. Finally, under our authorized user decision (4 FCC 2d 12, 421; 6 FCC 2d 593), Comsat may not ordinarily compete with them for the high revenue leased circuit business.

   189. Aside from any of the above considerations, respondents have integrated, under single ownership and control, not only the nationwide communications network, but also a supporting research and development unit, as well as a large manufacturing subsidiary. They, thus, are in a position to be in the forefront of technological development, as exemplified by their research on the waveguide and laser. They also manufacture and provide the equipment to exploit the fruits of their research and development. Finally, they market their services by means of their nationwide communications network. The record amply supports Scanlon's conclusion that no industry appears distinctly on the horizon as a replacement for the Bell System.

   190. Respondents have further protection in the field. All of their expenses for research and development are treated as a current cost of doing business and are included in the computation of their revenue requirements. Hence, the users, through their rates, are paying such costs as they are incurred, regardless of the results thereof, without the restraint of the competitive price structure of others in the field, as is generally true in the case of nonregulated manufacturers.

   191. The final element of long-term risk mentioned by respondents is that the manufacturers' plant is shorter lived and requires less in terms of investment and in terms of revenue. Again, respondents overlook several pertinent factors. Depreciation of plant is a charge against the ratepayer and is collected pursuant to schedules fixed or approved by regulatory authority. There is very little danger that plant actually devoted to the service will not be fully paid for. Unlike many manufacturers, respondents rarely, if ever, junk obsolescent plant because of competitive impetus. Costs are averaged and plant is continued in service until replaced in a schedule proposed by them. For example, respondents now propose to take several decades to convert fully the nationwide switching system to electronic switching, thus permitting a gradual and controlled phaseout of electromagnetic switching systems [*78] now in use. If respondents were operating in a competitive field, they might very well be required to accelerate plant retirements even though the plant may not have been fully amortized. This schedule has been taken into account in the setting of respondents' depreciation rates so that the customers will fully pay for the retired plant. We agree with respondents' witness, Nathan, that the Bell System is in a more favorable position to control the rate at which innovations are introduced.

   192. Respondents' witnesses concede that manufacturing companies have greater short-term risks than has the Bell System and that this greater risk is

reflected in the greater instability of earnings rates of manufacturing companies. They contend, however, that because manufacturing companies usually have smaller proportions of bonds in their capital structures, this substantially equalizes the risks associated with common stock equity for manufacturing companies and the Bell System. These generalizations are not supported either by the evidence of record with respect to earnings and risk or by the long-term ratings given respondents' securities in relation to manufacturing companies.

   193. Respondents made comparisons between their earnings and large groups of manufacturers. They then averaged earnings of the manufacturing entities over a period of years and grouped the averages. The fallacy of such a procedure is obvious. Very few, if any, individual investors own stock in hundreds of manufacturing entities. Even those who do not compare the risks involved in purchasing respondents' stock with the purchase of stock in one of a hundred other companies. The question is always, "Do I buy A or B"? Not "Do I buy A or 100 others which on the average will give me the return I want"? Secondly, average earnings are deceptive. In the period 1959-65, respondents' earnings varied from 9.5 to 10 percent on equity. However, the 528 manufacturers from which respondents culled their alleged comparable group showed variations from a loss of 51.9 percent to a profit of 49.3 percent on equity. Earnings from year to year vary for most manufacturers to an extent that makes them not proper subjects for comparison with respondents. The difference in capital structure does not compensate for the differences in risk. Aside from the wide variations in earnings, there is a constant stream of business failures among manufacturing companies in all years, prosperous or depressed. Respondents, of course, have not experienced any failure in any of the companies they own.

   194. The differences in risk are reflected in the ratings of recognized investment advisory services generally relied upon in financial markets and cited frequently in this proceeding. These ratings indicate that the securities of the Bell System are superior to those of practically all manufacturing companies. In this connection, the bonds of the Bell System all carry Moody's highest rating of AAA, except those of one company representing less than 2 percent of the total, which carry the next lowest rating of AA. In contrast, the bonds of only three manufacturing companies in the group of 528 manufacturing companies relied upon by respondents' witness carry Moody's highest rating of AAA (Bell exhibit 11, p. 29). Of 500 major industrial companies included in the "Fortune" magazine 1964 annual financial [*79] survey, 160 companies had rated bonds outstanding at the end of 1964. The bonds of only seven of these companies were rated AAA. Much the same relative risk rating is indicated by rankings of common stock by the investment advisory services. A.T. & T. common stock is classed in the highest investment rating by Standard and Poor's and the third from the highest rating by Value Line Investment Survey. Of the 528 manufacturing companies relied upon by the principal Bell witness in his comparable earnings presentation, only 38 companies, or about 7 percent, had a ranking as high as A.T. & T. by Standard and Poor's and only 45 companies, or about 8 percent, had a ranking as high as A.T. & T. by Value Line.

   195. On the basis of the foregoing, we find that the earnings of manufacturing companies do not provide a useful or reliable measure for fixing the return to be allowed respondents herein.

   196. Witnesses for respondents expressed divergent views on comparing the long-term risks of the Bell System with electric utilities. Some took the position that electric service is more essential than telephone service. Others stressed that Bell is such an indispensable element in our country's well-being that it has many of the attributes of a public trust. It appears to us, on the basis of the evidence, that the long-term risks are minimal for both the Bell System and the electric utility industry. So far as individual electric utility companies are concerned, it appears that Bell faces fewer long-term risks. Individual electric companies have direct competition from the gas industry as an alternative means of providing the same service. Thus, space heating and cooling, water heating, cooking, and refrigeration can be done by either gas or electricity, and this lively competition is reflected in numerous advertising campaigns. There is no such choice between telephone companies. While Bell faces competition from Western Union and the mail, such competition does not encompass as much of the field of service, particularly in the respondents' major message toll market. We note that Bell witness Nathan did not consider the competition of Western Union significant to the Bell System. USITA witness Foster stated in regard to competition, "I do not believe that there has been adverse financial results in adverse effect on the ability of the telephone industry to serve the public. We find that competition to the Bell System is confined to a relatively small segment of its total market." On the record herein we can find no basis for forecasting any significant change in the effect of competition upon the Bell System. While privately owned systems compete to some extent, they offer no danger to the backbone nationwide switched network. Individual electric utilities find competition from publicly and privately owned power systems. Such power systems can and do supplant services provided by an individual electric company. On the whole, electric companies individually do face a higher degree of risk than does Bell. However, by any test, Bell does not face any long-term risk as great as any of the individual electrics.

   197. Respondents' witnesses provided only general statements to support their views that the short-term risks for the Bell System are greater than those for electric utilities. They discussed these risks [*80] under the general classifications of (a) risks related to investment and (b) risks related to earnings.

   198. With respect to risks related to investment they took the position that the telephone plant must be planned to meet the needs of customers much more than electric plant; that telephone plant is much more complex than electric plant; and that a telephone company must supply all of the facilities used by its customers, whereas the customers supply the electric equipment on their premises.

   199. The fact that plant must be planned to meet needs of user or is more complex becomes important as an element of risk only to the extent that there is danger that it will not be used for the purposes planned and will become idle. We have already discussed fully the prospect of such occurrences and found them nonexistent or minimal. These factors, therefore, are not substantial in assessing comparative short-term risks.

   200. We are unable to find merit in the second contention, namely, that Bell faces a greater risk because it supplies equipment on customer premises, while the electrics do not. Telephone tariffs initiated and filed by respondents generally prohibit use of customer-owned equipment as a condition of their service offerings. Bell has substantially increased its investment and revenues through this practice. The equipment is depreciated from rates collected from the user. Thus, respondents and its investors are gainers, rather than the assumers of additional risk, as a result of this policy.

   201. Certain of respondents' witnesses also contended there is risk associated with its capital intensive business and long-lived plant. They acknowledged, however, that Bell plant, with an average life of about 20 years, is shorter lived as compared with electric plant, that has an average of about

37 years.

   202. Respondents contended that telephone companies are more susceptible than electric companies to loss of earnings in the event of a business decline. Respondents failed to demonstrate this, however, and relied on general contentions that telephone companies have more competition than electric companies; telephone expenses are less variable than those of electric companies, hence, cannot be as well controlled in a recession; electric rate structures insure a lesser decline of revenues during a recession; electric companies have fuel adjustment clause by which increased fuel costs can be passed on to customers; electric companies can shut down their highest cost plant as output demand declines; telephone companies have a higher proportion of labor cost which makes them more vulnerable in an inflationary period; and telephone service is more susceptible to cancellation in a recession. The only indication of record to support these contentions is that in the pre-World War II great depression years of 1932-35, the operating revenues of the Bell System fell somewhat more than did the electrics. We do not accept so remote a period as indicative of current conditions or possibilities. We think that the dependence upon telephone service is now so deeply embedded in the fabric of our society and economy that the experience of the 1930's is no longer valid. This assumption is buttressed by the fact that respondents no longer contend, as they have in the past, that interstate service merits a higher return [*81] than intrastate service on the ground that it is more subject to fluctuation and riskier.

   203. Respondents criticized comparisons of stability of earnings by staff witnesses because they are confined to the postwar period. Respondents claim

that the variation in earnings since 1950 is not a valid basis for any conclusions as to risk and that any such comparisons ought to be made for an earlier period. We do not agree. The 15 years since 1950 include periods of peace and war, recession and expansion, growth and stability. They cannot be rejected as being nonrepresentative.

   204. A basic defect of Bell's analysis of cyclical risk is its concentration on the phase of business decline (negative risk) to the neglect of the phase of

business expansion (positive risk). In a growth economy, by definition, the phase of business expansion must predominate, for otherwise there will not be growth for the entire period under consideration. Respondents expect the economy to grow even more rapidly in the decade ahead than during the post-World War II period, and for the Bell System and the communications industry generally, heavy demands are anticipated. Respondents, thus, inconsistently argue that the general economy and the Bell System will grow in the future at a very fast rate, while, at the same time, contending they are subject to great risk from business declines.

   205. In contrast to respondents' general contentions, there is specific evidence which controverts their claim as to risks of the Bell System. As we mentioned earlier, the period since 1950 has been one of general economic growth. Although the period was marked by the recessions of 1953-54, 1957-58, and 1960-61, expansion predominated. Over the period 1950-65, Bell System sales increased more rapidly than either the gross national product, the sales of electric utilities, or the sales of manufacturing firms. The average annual increase during this period was 9.5 percent for the Bell System, 7.7 percent for the class A and class B electric utilities, 6.2 percent for GNP, and 5.9 percent for manufacturing companies. For the 1959-64 period (the latest period used by Scanlon in his comparable earnings study), the average annual increase was 7.3 percent for Bell, 6.7 percent for class A and class B electrics, 5.9 percent for GNP, and 5.4 percent for manufacturers.

   206. In 1950, the experienced earnings on average total book capital for the Bell System and for the class A and class B electrics were the same -- 6.1 percent. Over the 1950-54 period, Bell's return on total capital increased more that experienced by the electric utilities, so that, by 1964, Bell's rate of return was 7.7 percent, as compared to an average rate of return of 7.1 percent for electrics.

   207. Commission witness Weintraub found that since 1955, as measured by value added, the telephone industry grew at an annual rate of 7.6 percent, as compared to 5.5 percent for the gross national product, and that respondents have shown little or no vulnerability to the business recessions that have occurred. In fact, throughout the period, and despite the recessions, the Bell System grew in number of telephones as well as in average revenue per telephone. A study by Lovejoy and Bowers entitled "Cyclical Sensitivity to Public Utilities, 1947-59" (" [*82] Land Econ.," November 1962), concluded that there is little or no evidence that the telephone industry's revenues were sensitive to changes in business activity during the 1947-59 period, and a recent publication of USITA n31 asserts that "Telephones are a basic utility with build-in recession resistance." USITA witness Foster found no observable sensitivity of exchange revenues to changing business conditions and that the recessions affected toll revenues only by slowing down the rate of growth. During the postwar period, not only did the total number of telephones show annual increases, but, also, the monthly revenues per telephone increased during recession years from $8.83 in 1953 to $9.14 in 1954; from $10 in 1957 to $10.25 in 1958; and from $10.80 in 1960 to $11.03 in 1961.

n31 USITA, "Annual Statistical Volume, 1965," vol. I, p. 7.

   208. Aside from the foregoing, the record shows that respondents' witnesses, in their presentations, omitted a number of significant factors which serve to reduce the risks of the Bell System as compared with those of electric utility companies. For example:

   (a) The Bell System, with its integrated research, manufacturing, and utility operations, reduces its risks associated with planning for growth and technological changes as compared with those of electrics. No electric system has a comparable protection.

   (b) The Bell System, being nationwide in scope, has the advantage of diversification as to the areas it serves, as well as the advantage of diversification as to all of the customers it serves. An electric company usually operates in a relatively small area, such as a city or a portion of a State. Therefore, it has all of the risks of fluctuations in the local economy and the risks of its much smaller diversification of customers. Since an electric company serves a local market, a customer moving out of this market area represents a total loss to that company. By contrast, the Bell System serves all 50 States and the loss experienced by one subsidiary company is likely to be a gain for another with little or no change in the position of the Bell System as a whole.

   (c) Respondents' witnesses have stressed that risks are greatly increased by taking liberalized depreciation for income taxes and treating those taxes on a flow-through basis. Most electric utilities follow this practice, while Bell does not.

   (d) Since respondents contend that the higher the debt ratio the higher the risk, it is relevant to note that the electrics have a debt ratio which is above 50 percent, whereas respondents' is presently below 35 percent and, even with the announced policy of raising it, it will not reach 40 percent for several years.

   209. That electric companies are considered to have greater risks than those of the Bell System is borne out by the ratings of securities by investors' advisory services. Of the 128 electric companies relied upon by a Bell witness, bonds of only seven of these companies carry Moody's highest rating of AAA, which is the rating accorded 98 percent of the Bell System bonds. Much the same picture is presented with respect to the financial ranking of common stocks. The common stock of A.T. & T. carries Standard & Poor's highest ranking of A +. Only 11 of the 128 electric companies had a ranking this high. A.T. & T. common stock has a ranking of A - by Value Line Investment Survey, whereas only 32 of the 128 electric companies had a ranking this high.

   210. By any reasonable test, neither the short-term nor the long-term business risks of Bell are any greater than those of the electrics, contrary [*83] to the contention of respondents. In fact, the evidence would indicate respondents to be less risky than individual electric companies.

(3) Capital Structure

   211. The policy of maintaining the debt proportion of total capitalization in the range of 30 to 40 percent was established in the 1920's. The ratio was above 40 percent, however, from 1947 to 1953; it has been below 35 percent, the midpoint of the range utilized by respondents' witness Scanlon in his rate of return calculations, since 1963.

   212. For the purpose of discussion, we will assume that respondents are quite comparable to the electric utilities, even though they may, as we note elsewhere, be less risky. Therefore, the capital structure, history and growth, and experience in raising capital of the electric utilities are most pertinent.

   213. We have already noted that, while the combined revenues of all electric utilities are somewhat greater than the Bell System's revenues, the postwar growth of the revenues of the latter has been greater. So, too, the Bell System gross plant has grown at a faster rate during the postwar period; nevertheless, the gross electric utility plant in 1946 and in 1964 was much larger than the Bell System gross plant on the corresponding dates.

   214. The electric utilities financed their massive postwar expansion (1946-64) of $47.5 billion of gross plant by raising capital in the following amounts: Common stock $8.9 billion, preferred stock $2.6 billion, long-term debt $18.5 billion, retained earnings $4.3 billion, for a total of $34.3 billion. The Bell System, on the other hand, financed $26.2 billion increase in gross plant through raising capital as follows: Common stock $10.8 billion, no preferred stock, $6.9 billion of long-term debt, and $4.7 billion of retained earnings, for a total of $22.4 billion. The electrics had a 61-percent ratio of senior capital (debt and preferred stock) at the beginning of the period and the same ratio at the end. The Bell System, on the other hand, started the period with a ratio of long-term debt of 38 percent and ended the period with a ratio of 33 percent.

   215. In view of the fact that respondents have announced, during this proceeding, a new policy with respect to their debt ratio, we will not dwell on the advantages which the electrics have enjoyed from their debt policy in contrast to the more conservative policy followed by the Bell System. It is necessary, however, to mention, at least, the effect on earnings per share due to the great emphasis which respondents have placed on various comparisons in this record.

   216. At the 10-percent return on equity sought by respondents herein, each dollar of equity financing requires nearly five times as much gross revenue as a dollar of debt financing. Thus, the ratepayer is penalized if more of the financing is by equity than is required. Further, too high a proportion of equity can affect the stockholders adversely. It is, therefore, obvious that the shareholders will benefit considerably from respondents' announced change in debt policy, inasmuch as the earnings on A.T. & T. equity will move closer to the electric utilities as the policy is implemented. Such equity earnings are pointed out by some of respondents' witnesses as a desirable objective.

    [*84] 217. Most of the debt issued by respondents is through the individual operating companies, and the bonds representing such debt are marketed in competition with similar issues of electric utilities and other enterprises. Because of the elimination of duplications of equity when the Bell System is considered on a consolidated basis, the consolidated debt ratio is greater than the average debt ratios of the individual companies making up the consolidation. Thus, to achieve a 50-percent debt ratio for the Bell System consolidated, neither A.T. & T., nor any individual Bell operating company, would be required to maintain a debt ratio in excess of about 33 1/3 percent. Also, in marketing bonds, respondents are in a favorable position because of the diversification of bond issues they make available to the investors. Respondents have flexibility as to which company and which area of the country will be represented by any given bond offering at a given time.

   218. Respondents, by their postwar experience, did, indeed, demonstrate they could achieve a 50-percent debt ratio. However, due to the convertibility features of the security issues, the ratio was soon reduced to that dictated by the then-established policy.

   219. Respondents say, however, that the low debt ratio with which they entered the postwar period enabled them to raise the extremely large amount of capital required for expansion, and that such a "cushion" should be retained so that they can meet any similar situation in the future. Aside from the fact that respondents have already changed the policy to which they point, we should note that, short of a war which would do widespread physical damage within the United States, it is most unlikely that a situation similar to the post-World War II years will arise. Even if we should assume, however, that such a demand could arise again, it does not appear that respondents would have any problem in meeting the need for the required capital. This is demonstrated by the electric utilities, who raised almost three times as much debt capital as respondents in the post-World War II period, even though starting the period with a debt ratio of approximately 60 percent.

   220. We find, therefore, that a continuation by respondents of their past policies with respect to capital structure will not be conducive to the raising of future required capital in a reasonably economical fashion. Instead, the continuation of such a policy would substantially increase the cost of service to the users and, equally important, have a depressing effect on the earnings and dividends per share, with detriment to the market price of A.T. & T. stock and the investors' interests.

   221. The following excerpt from a decision of the District of Columbia Public Utilities Commission adequately expresses the opinion to which we adhere:

   "* * * the prerogative of management to determine the capital structure of a utility corporation does not and cannot extend to a prerogative to fasten unreasonably high rates upon the consumers at its will. The balance of interests between investors and consumers requires that the latter pay the cost of capital for a reasonable capital structure in the light of the risks involved as against the relative costs, taking into account the tax and other pertinent considerations. The management, acting as a representative of the stockholders, may decide, as a matter of internal policy, to adopt a course of issuing no, or an unreasonably small amount of debt, but if it does so, it [*85] is the stockholders who must pay the cost of the additional insurance provided for them by that decision, and not the consumers." [Emphasis supplied.] Re Chesapeake & Potomac Telephone Co. (D.C. Public Utilities Commission, 1954), 6 PUR 3d 222, pages 239-240.

   222. Accordingly, in fixing the rate of return to be allowed, we shall take into account this "additional" and extraordinary amount of risk insurance respondents have given its stockholders by its low debt ratio policy. We note respondents' announced intention to alter this policy. Under the rate of return we shall allow, and in light of conditions which we find are likely to obtain in the foreseeable future, respondents are in a position to improve equity earnings by increasing their debt ratio, as they now propose to do. We further note respondents' representation that when a 40-percent debt ratio has been reached, they will review the situation to determine whether they should further increase that ratio. In this connection, we find that the record indicates respondents could effectively support a debt ratio above 40 percent. We believe such a course will be in the interests of users, investors, and the respondents.

   223. A further manner in which A.T. & T. financial policy differs from the electrics is with respect to the proportion of earnings paid out as dividends or payout ratio. A number of respondents' witnesses discounted the importance of dividends to investors; their witness Morton, however, stressed their importance. Morton states that: "The investor in the long run gets dividends, and the value of his stock in the long run depends on the dividends. It is, therefore, reasonable to look at the opportunity cost of capital as an opportunity to receive dividends." Morton pointed out that from 1961 to 1965, A.T. & T. dividends have provided investors with only a 5.8-percent return on book equity, while the FPC electric utilities have returned to their stockholders as dividends 8.2 percent on book equity. However, it is pertinent to note that the payout ratios for that period were 61 percent for A.T. & T. and 69 percent for the electrics.

   (4) Accelerated Depreciation

   224. A final factor in connection with rate of return is the policy followed by respondents of not using accelerated depreciation for tax purposes as permitted by section 167 of the Internal Revenue Code.

   225. Respondents have stated that they have not taken advantage of the provisions for accelerated depreciation for tax purposes because in many States and possibly in the Federal area they might be required to flow-through the tax savings. They apparently presume such a flowthrough would result in the entire amount going to reductions in rates, although whether it did so would depend on regulatory action. If not carried to rates, the savings would, of course, benefit stockholders. In any event, they say, to take accelerated depreciation would increase the risks of doing business. If this argument is valid, then respondents now have lower risks in this respect than most manufacturing companies and electric utilities which do use accelerated depreciation. It is not known to what extent manufacturers might flow-through the tax savings in competitive pricing, but most of the electric utilities do flow-through the savings. Thus, to the extent that risk is a factor in [*86] rate of return, and respondents urge strongly that it is, failure to take accelerated depreciation reduces their risk in comparison with that of entities which do take advantage of accelerated depreciation. We shall, therefore, in our decision herein consider this factor along with other relevant considerations, as they affect the allowable rate of return.

   226. In the second phase of the proceeding, we shall evaluate the substantive arguments as to the propriety of the course followed with respect to accelerated depreciation and what action, if any, we should take with respect thereto.

   G. Conclusions as To Rate of Return

   227. We have reviewed thoroughly the testimony of all of the witnesses and considered the evidence of record fully insofar as it relates to rate of return.

As noted earlier, we have rejected respondents' contention that, as a matter of law, we must accept their concept of comparable earnings. Upon analysis of the arguments and data advanced in support of their concept, we have found that the manufacturing companies, individually or as a group, are not comparable to respondents. Therefore, we cannot accept comparisons such as offered by Scanlon, respondents' principal witness on this subject, as valid support for their claim for a return of 10 percent on equity.

   228. Morton's use of what are essentially earnings-price ratios determined by the relationship of earnings, book value, and market price is similarly lacking in logic and reflects the deficiencies of the same broad averaging employed by Scanlon. The adjustment of A.T. & T. earnings to support an equivalent of 200 percent of book value suffers not only from the large subjective adjustment applied to the result, but from his effort to treat A.T. & T. as if it were unregulated, rather than by using any direct measurement of risk which Morton acknowledged could be made.

   229. Friend's method of combining dividends with estimated growth has basic merit and, in a broad sense, is what Commission consultants Thatcher and Gordon have done. The extremely unrefined method of estimating the growth factor for A.T. & T. which Friend utilized, as well as an admission that it is inferior to the comparable earnings method of Scanlon, which we have rejected, destroys the usefulness of his method and, therefore, does not provide us with a meaningful guide.

   230. Thus, we are unable to accept the conclusions of respondents' witnesses Scanlon, Morton, and Friend that A.T. & T. requires a return on equity of approximately 10 percent and an overall return of 7 1/2 to 8 1/2 percent. The generalized statements of other witnesses concurring in a return of that magnitude are also lacking in convincing support.

   231. The measurement or evaluation of risk proposed by the three designated witnesses, in our judgment, do not aid us in determining for respondents a return "commensurate with returns on investments in other enterprises having corresponding risks," one of the objectives which we noted at the outset.

   232. The two consultants retained by the Commission, Dr. Thatcher [*87] and Dr. Gordon, both propose a rate of return in the range of 6 3/4 to 7 ¼ percent. Dr. Thatcher followed the cost of capital approach and utilized an imputed capital structure of 45 percent debt and 55 percent equity.

   233. As we have already set forth, Dr. Gordon proposed a new and challenging approach to the question of fixing a rate or return in proceedings involving regulated entities. His econometric model, it appears to us, has promise of being a useful tool in this difficult and often vexing area of regulation.

   234. We have found his approach and methods useful in analysis and evaluation of the effect of capital structure, i.e., the ratios of debt and equity upon overall cost of capital. It lends support to our conclusion that a regulatory determination of revenue requirements can rely, in part, on announced and anticipated changes in capital structure.

   235. We have not had the opportunity to analyze, evaluate, and test fully his model to determine all of its implications insofar as fixing an overall rate of return is concerned. However, we believe that it merits further attention as a means of making available more objective data and substantive support for the exercise of the subjective judgments in fixing a rate of return. We would, therefore, encourage further study and refinement of the model to make it more useful in resolving the special problems which arise in the field of regulated utilities.

   236. Turning now to the conclusions we are required to reach in this proceeding, we find that of all of the different enterprises regarding which data appear in the record, the electrics offer the most fruitful basis for comparison. Like respondents, they provide a vital service, they are regulated, they have raised vast amounts of capital, and have shown substantial growth. We have earlier compared the risks involved and found respondents to be no more risky than the electric industry, and may be less risky. Insofar as earnings are concerned, we note that the electrics now earn less on overall capital than do respondents, but more on equity. This difference is due primarily to the financial policies followed by the respective managements. The electrics have relied much more heavily on debt and senior capital for the raising of the vast sums of new capital they require than have respondents. The lower cost of debt, and the leverage given equity by the larger reliance on debt, have given the equity holders of electrics higher earnings per share. Respondents, on the other hand, have relied much more heavily on equity financing. This decreases leverage and requires considerably more revenue to support a given level of earnings. Since the record justifies the conclusion that respondents are no more risky than the electrics, it follows that respondents can support a capital structure with a debt ratio much closer to the electrics than at present.

   237. The second factor which enabled the electrics to show greater equity earnings is their use of accelerated depreciation with flow-through. To the extent that the tax savings have not been fully reflected in reduced rates, earnings have increased. Respondents have, of course, not taken advantage of accelerated depreciation.

   238. We note that respondents are now moving toward a higher debt ratio and, accordingly, will be providing increased leverage to the [*88] stockholder. However, because of the vast sums involved, this will take several years. We believe, however, that respondents' announced policy and our conclusions herein with respect to capital structure will, as Gordon predicts, have a beneficial effect on the market for respondents' stock as well as their cost of capital.

   239. We have also noted the higher cost of debt at present, the effect it will have on embedded debt costs in the future period for which we are fixing the return, current business and economic conditions, and all other relevant factors. We have also considered respondents' actual earnings and financial experience during the entire postwar period. We have taken account of the returns allowed telephone companies and electrics in formal proceedings and the returns actually being realized by them. On the basis of all of the foregoing, and applying our considered judgment, we conclude that a fair rate of return to respondents on their interstate operations is in the range of 7 to 7 1/2 percent. Since we have considered respondents' overall earnings requirements n32 in reaching our conclusions with respect to the allowable rate of return on the 25 percent of their total operations represented by the interstate segment of their business, it is pertinent to note what the immediate effects of our decision on respondents' overall equity earnings will be. Thus, even if our findings and conclusions hereinafter set forth with regard to jurisdictional separations and interstate revenue requirements were to be implemented at once by rate reductions, it is clear from the record that respondents' total earnings on equity would then still exceed 9 percent on the basis of the 1966 test period. With the increase in their debt ratio in accordance with their announced policy, combined with the continued long-term growth in traffic and revenues, as well as more efficient use of facilities, respondents' equity earnings can be expected to move upward.

n32 Our Findings and conclusions herein relate solely to the services subject to our jurisdiction and are not intended to apply to the intrastate services subject to the jurisdiction of the various State regulatory agencies which, of course, will make their own determinations on the basis of conditions applicable within their own jurisdictions.

   IV. JURISDICTIONAL SEPARATIONS

   A. General

   240. Respondents are engaged in furnishing both intrastate and interstate communications services, including exchange, message toll telephone, private

line, and teletypewriter exchange (TWX). The intrastate services are subject to the jurisdiction of the several State regulatory bodies, and interstate services, except as exempted by section 221(b) of the Communications Act, are under the jurisdiction of this Commission. The major portion of telephone property of the associated companies is used in common for both intrastate and interstate services. Similarly, the major portion of the expense is incurred in the joint rendition of these services. All of the plant of A.T. & T. (the parent company) is used exclusively for interstate and foreign communications services.

   241. In general, telephone plant is divided into two broad classifications: Exchange plant, which is plant used primarily to furnish [*89] local exchange service, and interexchange plant, which is plant used to furnish toll (long distance) service. The exchange plant generally consists of the telephone instruments, drop and block wires, and other station equipment located on the customers' premises; the subscriber lines from the customers' premises to the telephone company's central office; the portion of the central office equipment used to switch and connect one subscriber line to another and other facilities required to establish a connection between telephones in the same exchange or between a telephone in one exchange and the toll lines to another exchange.

   242. The interexchange plant generally consists of switching equipment and circuit or toll lines plant. The switching equipment, which may be manual or dial, is the equipment used to connect toll lines to toll lines, or toll lines to local central offices. The toll lines circuit plant is composed of central office equipment, which creates and controls the circuits, and outside plant, such as conduit, poles, wire, and cable.

   243. Some of the telephone plant is designed and used specifically for a particular service. However, a major portion of the plant is used in common for exchange, State toll, and interstate toll service. This is particularly true of the plant used to provide message toll telephone service. Examples are the telephone instruments and subscriber lines which are used jointly to handle exchange calls and long distance messages, either State or interstate. Likewise, many toll circuits between cities are used in common to transmit both intrastate and interstate long distance messages.

   244. Because the major portion of respondents' property is used in common for both intrastate and interstate services, an appropriate method of separating or allocating those costs in essential so that each regulatory jurisdiction -- State and Federal -- may determine, for ratemaking purposes, the investments, revenues, expenses, taxes, and reserves applicable to the services subject to that jurisdiction. References herein to the States are intended to include the 48 contiguous States and the District of Columbia. They do not include Alaska and Hawaii which, by virtue of their geographic location and the nature of the facilities required to serve them, cannot properly be governed by the separations procedures which are appropriate for the other States. Since 1941, the formulation and, from time to time, the revision of the procedures employed in the Bell System for such separations have been largely the product of cooperative efforts involving this Commission, the State commissions through the NARUC, and the telephone industry. In 1947 the procedures were incorporated in the original "Separations Manual." n33 Since then, there have been four major revisions to the "Separations Manual," anmely, the "Charleston Plan" (1952), the "Modified Phoenix Plan" (1956), n34 the "1962 Simplification Changes," and the "Denver Plan" (1965). Each of these [*90] revisions increased the allocation of plant investment and expenses to interstate operations. It is estimated that these revisions have resulted in a shift from intrastate to interstate of about $500 million in revenue requirements based on 1965 volume of business. The 1963 edition of the manual with addenda covering the subsequent revisions of the procedures constitutes the separations methods employed by the Bell System from which are derived the interstate operating results presented by Bell in this proceeding.

n33 The "Separations Manual" has been used by telephone regulatory agencies for ratemaking purposes, although this Commission has never formally approved or adopted the procedures thereof. The procedures of the manual are also employed within the Bell System in connection with the distribution among the system companies of revenues collected from their rendition of interstate services.

n34 This plan was a modified version of an earlier plan which was recommended by the NARUC at its convention in Phoenix, Ariz., but was rejected as improper by the FCC, hence, the name "Modified Phoenix Plan."

   245. The circumstances relating to the introduction of the "Denver Plan" are particularly pertinent to this procedure. During the year 1965, continued studies to determine equitable methods for allocating exchange plant costs among the operations were undertaken by the telephone industry and the NARUC Committee on Communications Problems. A revision in the exchange plant separations was devised by A.T. & T. and proposed at a meeting of the committee in Denver, Colo., on July 22 through 23. This revision was designed to reflect the value of the local plant to toll message services. This so-called "Denver Plan" will be described in more detail later. There was divided opinion in the industry as to the merits of the plan. The NARUC approved the revisions on an interim basis through resolution adopted on October 2, 1965. The FCC indicated it would interpose no objection to use of the revised procedures on the interim basis recommended by the NARUC, subject to such changes as may be required as a result of this proceeding.

   246. In our order of October 27, 1965, instituting this investigation, we stated:

   Although the content of the "Separations Manual" is the product of cooperative studies and consultations involving the NARUC, this Commission, and the telephone industry, it has never been formally evaluated, approved, or adopted by this Commission in the context of either a ratemaking or rulemaking proceeding. This is true with respect to the original "Separations Manual" as issued in 1947 and the several revisions that have been subsequently incorporated therein. * * * This, in connection with our determination of revenue requirements of the Bell System applicable to its interstate services, the rates for which are at issue herein, we will consider the propriety of the principles and procedures of the "Separations Manual" including the most recent revision.

   B. Applicable Principles and Criteria

   247. In the course of these proceedings, a number of different proposals have been advanced by respondents and other parties with respect to principles and procedures of separations that should be adopted by the Commission for the purpose of determining the revenue requirements of respondents. Before discussing these proposals in light of the evidence presented with respect thereto, it will be will to review certain related policy and legal considerations.

   248. It is urged by respondents and other parties that a fundamental principle to be observed in making jurisdictional separations is the need for uniformity in the procedures applied by both Federal and State authorities for ratemaking purposes. We subscribe fully to this objective, as we have in the past. Such uniformity obviates the danger that certain amounts of plant investment and expenses may be [*91] assigned to more than one jurisdiction to the detriment of ratepayers. Equally important, it obviates the risk that certain amounts of plant and expenses will be recognized in neither jurisdiction, to the economic detriment of the company and its owners.

   249. At the same time, we recognize the power inherent in each ratemaking authority, barring effective preemptive action by the Federal jurisdiction, to employ any separation procedure which meets the test of legality and fairness. In this regard, sections 221 (c) and (d) of the Communications Act reflect an explicit expression of congressional intent that the Commission should have the power to determine the costs of service furnished by telephone companies subject to its rate regulatory jurisdiction. n35 We cite this not for the purpose of establishing that our action herein will operate to preempt the authority of the State jurisdictions in matters of separation procedures to be employed by them, n36 but only to demonstrate what is contemplated by the scheme of regulation fashioned by Congress for the guidance of this Commission with respect to interstate telephone rates.

n35 Sec. 221 provides in part as follows:

   * * *

   "(c) For the purpose of administering this Act as to carriers engaged in wire telephone communication, the Commission may classify the property of any such carrier used for wire telephone communication, and determine what property of said carrier shall be considered as used in interstate or foreign telephone toll service. Such classification shall be made after hearing, upon notice to the carrier, the State commission (or the Governor, if the State has no State commission) or any State in which the property of said carrier is located, and such other persons as the Commission may prescribe.

   "(d) In making a valuation of the property of any wire telephone carrier the Commission, after making the classification authorized in this section, may in its discretion value only that part of the property of such carrier determined to be used in interstate or foreign telephone toll service."

n36 Whether or not our action herein will have preemptive effect on State regulatory jurisdictions is a legal question that we are not obliged to rule on at this time.

   250. The record shows that there is no unanimity of viewpoint among respondents and the other parties as to the appropriate procedures which should apply for jurisdictional separation purposes. As will be discussed hereinafter, there are substantial disagreements among the respondents, the States, Western Union, GSA, and the independent telephone companies as to the separation methods that we should invoke concerning respondents' costs associated with their interexchange circuit plant. Similarly, there is substantial disagreement between respondents and the NARUC, on the one hand, and the independent sector of the industry, on the other hand, as to what procedures are reasonable for the separation of respondents' exchange plant. This substantial lack of agreement exists notwithstanding the opportunity afforded the respondents and other parties by the Commission, in the course of this proceeding, to arrive in concert at a common approach to the separations problem for our consideration herein.

   251. It is relevant to note here that the most recent revision (the "Denver Plan") in separations procedures was embraced by the respondents and the States (acting through the NARUC), without the formal or informal approval of this Commission. That the Commission had serious questions regarding the reasonableness of this revision was made known to the NARUC and the industry prior to its implementation. In our letter to the NARUC, dated October 27,1965, we advised the States as to the nature of our reservations and [*92] indicated that, while we would interpose no objection to the use of the plan on an interim basis, we were simultaneously instituting the instant proceeding in which we would evaluate the propriety of the plan and that our interim acquiescence was subject to the determinations made herein.

   252. The independent companies contend that any separations procedures determined in the context of this proceeding should be applicable to all companies comprising the telephone industry. We recognize that appropriate telephone separation procedures are of vital interest to the independents. Such procedures must be employed by the independents as the underpinning for the establishment and regulation of their rates by State and Federal authorities. Separation procedures are also critical to the independents in connection with their division of revenues with Bell System respondents for traffic handled jointly by Bell and the independents, to the extent that such divisions must be related to the costs incurred by the latter in handling such joint traffic. We also recognize that the "Separations Manual," by its terms, is considered to be applicable to all segments of the industry, with certain specified exceptions.

   253. It should be stressed that the Commission is not prepared, on the basis of the record herein and in the context of the issues of this proceeding, to determine whether the separation procedures concluded to be appropriate for application to the costs of the Bell System respondents are equally applicable

to the costs of the independents. The thrust of the instant proceeding is to determine, among other things, the costs and revenue requirements of the Bell System respondents for their interstate services. Neither the costs nor rates of the independents are under examination in this proceeding. No independent company is a respondent in this proceeding. Thus, although we can understand and sympathize with the desire of the independents to establish separation

procedures which can apply uniformly to all segments of the industry, the record and the issues herein foreclose such determination. This is not to say that separation procedures which are appropriate for ratemaking purposes in the case of the Bell System are not equally appropriate for ratemaking and division of revenue purposes of the independents. They may well be. But whether or not they are is a question that we can only resolve in a proceeding directed to such a determination. There are provisions of the Communications Act which are designed to accommodate such a proceeding and which can be invoked at any time by an independent telephone company or by the Bell System (e.g., sec. 201(a) of the act).

   254. There is no dispute with respect to the applicable legal principles and controlling standards to apply in a determination of appropriate jurisdictional separation procedures. It is generally recognized that the applicable principles and standards necessary to accommodate State and Federal authority have been firmly established since the decision of the U.S. Supreme Court in 1913 in Minnesota Rate Cases, 230 U.S. 352, where it was held that a jurisdictional separation was essential and that it must be made on the basis of relative use.

   255. In 1930, the Supreme Court, in Smith v. Illinois Bell Tel. Co., 282 U.S. 133, for the first time considered the problem of separations [*93] as it specifically related to the establishment of rates for telephone service. The Court emphasized that a separation of telephone property, revenues, and expenses, between intrastate and interstate operations, "is essential to the appropriate recognition of the competent governmental authority in each field of regulation." Referring to the practical difficulty of separating certain exchange plant between local and long distance service, the Court observed:

* * * While the difficulty in making an exact apportionment of the property is apparent and extreme nicety is not required, only reasonable measures being essential * * *, it is quite another matter to ignore altogether the actual uses to which the property is put. It is obvious that, unless an apportionment is made, the intrastate service to which the exchange property is allocated will bear an undue burden -- to what extent is a matter of controversy. We think that this subject requires further consideration, to the end that by some practical method the different uses of the property may be recognized and the return property attributable to the intrastate service may be ascertained accordingly.

   256. The actual or relative use standard has purportedly been the touchstone of all cooperative endeavors of the FCC, the State commissions, and the industry in devising separation procedures. The procedures which antedated the 1947 "Separations Manual" and which, with certain modifications, were incorporated into the manual, contemplated that separations would be made in the following manner. In general, and subject to the use of procedural "shortcuts" that would not distort separations results, the costs of plant used wholly for a single service were to be assigned directly to that service. The costs of plant used jointly for State and interstate services were to be apportioned on the basis of the proportionate use made of the plant by each service as determined by relative occupancy and relative time measurements. For example, interexchange toll lines plant within a given State, used jointly for intrastate and interstate message toll telephone calls, was apportioned between the two services on the basis of their proportionate occupancy of the common plant, as measured by message-minute-miles (MMM's). If the units of interstate use (MMM's) equaled the intrastate units (MMM's), the cost of such jointly used plant would be equally apportioned between the intrastate and interstate services. Subscriber plant, consisting of the telephone instruments and subscriber lines connecting the subscriber to a central office, as previously noted, is used in common for exchange, intrastate, and interstate calls. Thus, under the 1947 manual procedures, the costs of such plant were apportioned on the basis of the relative use made thereof by each of the several services as measured by minutes of occupancy.

   257. It has been urged over the years that the procedures set forth in the manual represent an overly puristic or unduly slavish application of the principle of relative or actual use to jurisdictional separations. It has been contended that such procedures fail to take account of, or give weight to, other factors which are essential to a fair and equitable jurisdictional allocation. As evidence or symptoms of such inequities, the NARUC has, in the past and on this record, cited the historical condition that the rates for intrastate message toll telephone services in almost all States are substantially higher than interstate rates for message toll telephone calls of comparable distances. The [*94] fact that such rate disparities to exist, however, is, in our view, a consequence of the multiplicity of regulatory jurisdictions over message toll service. Not only are different ratemaking philosophies and practices applied in each of those jurisdictions, but the structure of costs and patterns of usage differ from State to State and between the Federal and each State jurisdiction. So long as there is no single jurisdiction over all toll service in the United States, there is bound to be a diversity in ratemaking and disparity in rates among jurisdictions. Accordingly, we do not regard the mitigation or elimination of disparity between state and interstate toll rates as a valid consideration in determining the propriety of separations procedures. We stress that the purpose of such procedures is to determine the amount of investment, expenses, taxes, and reserves subject to Federal and State jurisdiction. Their purpose is not to effect an artificial or contrived equalization of the cost per unit of service among all jurisdictions. Once the total costs have been ascertained by reasonable methods for each jurisdiction, each then applies its own ratemaking principles in establishing appropriate rates.

   258. However, we recognize that allocation of costs is not simply "a matter for the slide rule. It involves judgment on a myriad of facts. It has no claim to an exact science." Colorado Interstate Gas Company v. Federal Power Commission, 324 U.S. 581, 589. Thus, although a jurisdictional separation must take account of and measure the different uses to which common property is put by several services, it must also take account of any conditions or circumstances which clearly affect the reasonableness or equitableness of the results produced by such measurements. The issues presented by this record arise from the differing contentions and views of the parties as to the nature of the conditions or circumstances that warrant consideration in a proper jurisdictional separation of certain elements of the Bell System's plant and expenses. Furthermore, there is a wide divergence of views as to the appropriate method by which such conditions or circumstances should be reflected in the allocation procedures.

   C. Summary of Proposed Changes

   259. Separations changes have been proposed by the Bell System, the independent telephone group (consisting of U.S. Independent Telephone Association, G.T. & E. Service Corp., United Utilities, Inc., and National Telephone Cooperative Association), the Western Union Telegraph Co., the National Association of Railroad & Utilities Commissioners, and the General Services Administration for the executive agencies of the United States. The changes proposed by each party are summarized here and will be discussed later.

   1. The Bell System proposes the elimination of the "Modified Phoenix Plan" of separating interexchange circuit plant, and substituting a plan that approximates methods used in the 1947 "Separations Manual." With respect to exchange plant, the Bell System proposes a new "Use-Worth Plan" to replace a portion of the "Denver Plan" which relates to the separations of subscriber plant. It recommends retention of the "Denver Plan" insofar as it applies to the separation of dial switching equipment and exchange trunk plant.

    [*95] 2. The independent telephone group proposes an interexchange plan to arrive at a nationwide average investment per conversation-minute-mile (CMM). For exchange separations, the group proposes the "Comparable-Use-Unit (CUU) Plan" applicable to all exchange plant.

   3. Western Union proposes the elimination of the "Modified Phoenix Plan" and the substitution of a plan that involves an averaging process for interexchange circuit plant. For exchange separations, Western Union proposes an allocation for subscriber plant that reflects an estimate of availability.

   4. The NARUC supports the Bell System "Use-Worth Plan" for subscriber plant, but opposes the elimination of the "Modified Phoenix Plan" for separations of interexchange circuit plant.

   5. GSA recommends the retention of the features of the "Modified Phoenix Plan," the establishment of the exchange plant categories of the "Charleston Plan," and the use of peak monthly minutes of use for the allocation of all plant.

   D. Interexchange Plant

   260. Historically, two methods have been used for the separations of interexchange circuit plant to determine the allocation of the cost of this plant between State and interstate jurisdictions. The first plan was contained in the 1947 edition of the "Separations Manual" and remained in effect until 1956, when the "Modified Phoenix Plan" was approved as a method for the allocation of the interexchange circuit plant.

   261. Under the 1947 plan, interexchange circuit plant costs were assigned to the following subcategories:

   (a) Circuits used wholly for interstate service. service.

   (b) Circuits used wholly for intrastate service.

   (c) Circuits used jointly for both interstate and intrastate service.

   262. The book cost of the circuits used wholly for a single service were assigned directly to that service. The book cost of circuits used jointly for both services were allocated between services on the basis of the conversation-minute-miles of traffic for each service using the facilities jointly.

   263. The 1947 manual did not identify or separate property, revenue, and expenses of special services, since at that time these services represented a relatively insignificant part of the operations. The manual stated that special services need not be accorded special treatment provided the property, revenues, and expenses were handled consistently.

   264. In July 1956 the "Modified Phoenix Plan" became effective as the method of separations of the respondents' interexchange circuit plant and is currently in use. The "Modified Phoenix Plan" procedures provided a new method of allocation of the book costs of interexchange circuit plant of the associated companies of the Bell System used primarily for message toll service. Although the Long Lines plant terminating in each State plays an important role in the allocation process, it is still assigned 100 percent to the interstate operations.

   265. The "Modified Phoenix Plan" has as its design the treatment of all Bell System toll lines plant in a given State which is used to serve subscribers in

that State as if such plant were jointly used to render both intrastate and interstate services. Thus, the book costs of Long Lines plant which terminates in the State are combined with the [*96] book costs of associated company terminating plant, even though such Long Lines plant is used exclusively for interstate calls originating or terminating in the State. The combined book costs are then apportioned on the basis of relative use measured by the relative

number of State and interstate conversation-minute-miles occupying the combined plant. The book costs so assigned to intrastate service are then deducted from the total book costs of the associated company interexchange circuit plant and the remainder is assigned to interstate operations. All of the Long Lines book costs are, of course, still directly assigned to interstate, since those costs are added to the associated company book costs under the "Modified Phoenix Plan" procedures only as a catalyst to effect a lower cost per unit of intrastate use of associated company plant.

   266. This procedure results in the same cost per unit of use, messageminute-mile (MMM), of all message toll lines plant in a State serving customers in that State. This treatment averages the Long Lines lower unit cost per MMM with the higher unit cost of associated company plant, thereby assigning a larger amount of costs of this plant to interstate.

   267. The rationale given for the "Modified Phoenix Plan" at the time of its introduction was as follows: (a) All of the terminating plant, whether owned by the associated company or Long Lines, was used by customers in that State; (b) such plant was engineered and operated as an entity to provide both intrastate and interstate toll service to subscribers in the State; (c) the Long Lines terminating plant, although under separate ownership, is frequently a part of joint plant, such as particular pairs of wires in a jointly owned cable or wires supported on pole lines jointly owned with the associated company; and (d) thus it was appropriate to combine such plant in making the apportionment between intrastate and interstate services on the basis of the conversation-minute-mile of use made of such plant.

   268. Proposed plans for changes in present methods used to allocate the cost of interexchange plant between State and interstate have been presented by the respondents and three of the intervenors with a fourth intervenor, the NARUC, recommending no change in the methods presently in use.

(1) Respondents' Position

   269. Respondents propose that the "Modified Phoenix Plan" be eliminated and that the cost of interexchange circuit plant be allocated by methods generally consistent with those used prior to 1956, so as to assign all costs of circuits used wholly in either interstate or intrastate operations to the operation in which they are so employed, and so as to allocate between such operations only the costs of jointly used circuits. Respondents also propose that certain broad averaging techniques presently used to determine the costs of associated company and independent company interexchange circuits be eliminated by adopting a method of separation which provides for computing line haul and terminal costs separately to reflect the effect of lengths of circuits on the total cost per mile. The effect of the Bell proposal for changes relating [*97] to interexchange plant would be to shift about $176 million of revenue requirements from interstate to intrastate operations.

   270. The Bell System contends that the present procedures for separating interexchange circuit plant cost should be changed because:

   (a) There has been a 50-percent increase in the length of haul of interstate messages since 1955. This has aggravated the over-assignment of costs to interstate under the "Modified Phoenix Plan" since it is based on an average cost per mile of State and interstate use.

   (b) There has been a significant change in the assignment of cost to interstate since mid-1956 when the "Modified Phoenix Plan" was made effective for use in the Bell System Division of Revenues. In 1956, 56 percent of interexchange circuit plant book cost of associated companies was assignable to interstate, compared to 44 percent based on preexisting procedure. In 1966, 65 percent of these costs was assigned to interstate under "Modified

Phoenix."

   (c) The distortions created by the application of "Modified Phoenix Plan" are increasing. The effect of the "Modified Phoenix" procedure on Bell System message circuit book cost was estimated to be an increase of approximately $152 million assigned to interstate based on 1954 data. Currently the amount assigned to interstate under this procedure is about $481 million. This is approximately 3.2 times as much as was assigned in 1954, although the total book cost of Bell System message circuit plant has increased only about 2.1 times in the same period.

   (d) The proportion of traffic which is alternately routed today is declining and is less than was anticipated when the "Modified Phoenix Plan" was conceived in the early 1950's.

   (e) The present procedure in the manual for assigning costs to plant categories on the basis of the overall average cost per circuit mile, developed in the early 1940's, should be abandoned. These procedures were developed when practically all circuits were derived by the use of voice grade open wire and

cable facilities. Carrier techniques employing wire line, coaxial cable, and microwave radio facilities have been expanded so that presently over 90 percent of associated company circuits and 95 percent of Bell System circuits are provided through this technique. Under this technique a large part of the circuit costs are concentrated at the ends of circuits. It is contended that the present method fails to reflect the difference in the proportion of terminal cost and line haul cost for circuits of various lengths.

   (f) The averaging techniques of "Modified Phoenix" result in assigning book cost to intrastate operations based on a $62 combined Long Lines ($36) and associated company ($82) terminating book cost per circuit mile instead of the average book cost per circuit mile of associated company plant.

(2) NARUC Position

   271. The NARUC and the California Public Utilities Commission oppose the Bell proposal for the elimination of the "Modified Phoenix Plan." In support

they state:

   (a) The greatest economies in interexchange circuit plant costs continue to be in the Long Lines plant. While the costs of both Long Lines and associated company plant per circuit mile have decreased since 1954 the ratio between them has increased. In 1954, associated company plant cost was approximately 2.6 times as much as Long Lines plant. In 1963, associated company cost was 3.6 times that of Long Lines.

   (b) There has been no showing of any burden on interstate service occasioned by the "Modified Phoenix Plan." On the contrary, interstate service has flourished and grown. From 1955 to the present, in excess of $175 million in interstate toll reductions have been consummated, estimated at the time of the rate changes. In addition, two major changes in separations methods have resulted in a transfer of revenue requirements to interstate totaling $145 million net at the time of the change. In spite of these reductions in rates and increases in expenses, the Bell System interstate ratio of revenues to expenses has improved each year since 1956.

    [*98] (c) The toll network has been designed as an entity, and every portion benefits every other portion and provides economical operation of the total toll system.

   (d) The average cost per circuit mile for the associated companies is much higher than the cost per circuit mile for Long Lines. The lower cost of Long Lines is primarily related to the longer length of haul in the interstate plant and is due to the mix of the various types of facilities at the

various lengths of haul.

   (e) The associated company feeder plant, which contributes to the low cost, high volume trunk routes, is an essential part of the interstate toll telephone network and, although high in cost, results in substantial economies in the interstate operations.

(3) Independents' Position

   272. The independent telephone group recommends the conversation-minute-mile (CMM) plan for the separation of interexchange plant. The CMM plan expands the averaging principle of the "Modified Phoenix Plan" and has the same basic features as the message-minutemile plan which is discussed hereinafter in connection with the plan proposed by Western Union. It would average the cost of interexchange dial switching plant and interexchange circuit facilities, and apportion all of these costs on the basis of relative nationwide conversation-minute-miles. In effect, this plan would produce the same average book cost per MMM for intrastate and interstate throughout the Bell System. The

independents argue that:

   (a) This plan provides a uniform unit of use measurement in terms of a nationwide interexchange circuit and intertoll switching plant cost per conversation-minute-mile.

   (b) This plan recognizes that calls may be routed between all parts of the country automatically and the relatively small tributary circuit groups are the traffic sources of the large backbone groups and are so designed to assure maximum utilization of the large groups.

   (c) The averaging of the CMM plan benefits the short haul tributary circuits.

   273. The independent telephone group's position is that if the CMM plan is not adopted they would favor continuation of the "Modified Phoenix Plan" with inclusion of the independent telephone companies.

(4) Western Union Position

   274. Western Union (WU) proposes the elimination of the "Modified Phoenix Plan." In its place the WU plan would, in general, combine the interexchange circuit plant costs of the associated companies and independent companies in each study area with Long Lines costs (based on a nationwide average) for the study area for each service separately. The plan would then accumulate the total nationwide costs for the message toll service and separate such costs between State and interstate in each study area by multiplying the intrastate MMM's by the nationwide average cost per MMM to determine costs to be allocated to intrastate message toll. The remaining costs in the study area would then be assigned to interstate message toll service.

(5) Conclusions

   275. At the outset we may dispose of the proposals of Western Union and the independent group as regards the separation of interexchange plant. These plans, the so-called MMM and CMM plans, would equalize [*99] costs and usage on a nationwide basis and thus would ignore the actual economic conditions existing in each jurisdiction. In order for such plans to afford a proper method of separation under existing law, it would have to be found that similar economic conditions obtained in each jurisdiction throughout the system, Capital Transit Co. v. P.U.C., 213 F. 2d 176 (D.C. Cir. 1953), cert. den. 348 U.S. 816. Otherwise, the ratepayers in one jurisdiction would be in the position of subsidizing the ratepayers in another. On the record before us, we are unable to make the requisite finding to support such a system-wide approach. In fact, the evidence of record indicates to the contrary, i.e., that costs and usage vary substantially among jurisdictions.

   276. The bases for the use of the "Modified Phoenix Plan" were stated to be:

   "This procedure is consistent with the fundamental plan under which Bell System toll lines plant is engineered and operated as an entity in serving the customers in a State, as described above. It recognizes that Long Lines plant, and more particularly terminating toll lines plant, is in fact mostly a part of a plant owned jointly or used jointly with the associated companies. Because of this commingling, it is appropriate that the combined terminating plant be treated as an entity and apportioned on the basis of the use (MMM) made of the combined plant.

   "Moreover, the bulk of Long Lines terminating plant is largely particular wires on a joint open wire line, particular conductors in a jointly owned cable, together with the appropriate share of supporting structures (poles, conduit, right-of-way) and, in other cases, wholly owned cables on jointly owned supporting structures. Thus, while a given terminating circuit may be used solely for interstate, it is generally a part of a joint plant.

   "The proposed procedure would result in the same cost per unit of use (per MMM) of all message toll lines plant in a State serving subscribers in that State. This is consistent with the existing procedures of the 'Separations Manual' for apportioning exchange plant costs among exchange service, intrastate and interstate toll service, as well as for apportioning jointly used toll lines plant.

   "An effect of the proposed procedures is to spread the benefits of Bell System research and development, which have been directed toward maximum

economies for telephone service as a whole, but which have produced the greatest toll lines economies on heavy routes and longer circuits which are predominantly interstate." (NARUC, Proceeedings, 1954, pp. 274, 275.) We think these are sound reasons, and unless some or all of these no longer obtain, it would seem that we would not be justified in overturning a separations principle which has endured for the past 11 years.

   277. Respondents, in advocating elimination of the "Modified Phoenix Plan," contend that (1) it is inherently unsound, and (2) analysis of the arguments in support of "Modified Phoenix" emphasizes the importance of its elimination. With regard to its contention that the plan is inherently unsound, A.T. & T. alleges that historically it was adopted as a compromise and contains some features of the objectionable "Full Phoenix Plan," and that changes in conditions since 1956 have aggravated what A.T. & T. contends is an overassignment of costs to the interstate jurisdiction. Suffice to say, with respect to both of these allegations, that neither of them goes to the merits of "Modified Phoenix" as a rational method of separations. There is nothing inherently wrong with a compromise of views on something as to which the Supreme Court has recognized the difficulty of making an exact [*100] apportionment. n37 Moreover, the fact that changed conditions have quantitatively increased the proportionate differential between interstate and intrastate costs contemplated by the plan has no bearing on a determination as to whether the plan itself has merit. As we have indicated, and as A.T. & T. apparently recognizes, a more logical approach is to analyze the original reasons for "Modified Phoenix" and determine whether such reasons still obtain.

n37 Smith v. Illinois Bell Tele. Co., supra; Colorado Interstate Gas Co. v. F.P.C. supra.

   278. The principal reason for the cost averaging within a State, which is the main feature of "Modified Phoenix," is the fact that the Bell System toll lines plant is engineered and operated as an entity in serving the customers in a State. A.T. & T. concedes that "[this] is of course true." It contends, however, that it does not follow that costs incurred solely in furnishing either State to interstate services should be averaged among themselves, or averaged with costs related to jointly used facilities.

   279. The "Modified Phoenix Plan" admittedly contemplates the averaging of certain costs. Such averaging was introduced in recognition of the fact that, within each State, the interexchange plant, as respondents admit, is engineered and operated as an entity, and thus, to a large degree, the various parts are interdependent. In an effort to show that such interdependence is no longer valid as a reason for averaging costs, respondents point to a factor which, if it were of sufficient substance, would tend to negate the validity of the cost averaging. This factor is the trend toward using direct circuits between metropolitan centers instead of switching circuits together to handle such traffic. However, respondents' witness was unable to quantify this trend, and, in fact, stated that most of the traffic today was still handled over the switched network.

   280. Aside from the above-discussed considerations which we regard as the fundamental reason for the "Modified Phoenix Plan," the 1954 justification refers to the spreading of the benefits of Bell System research and development, which have been directed toward maximum economies for telephone service as a whole, but which have produced the greatest toll line economies on heavy routes and longer circuits which are predominantly interstate. Respondents' evidence indicates that recently techniques have been developed to reduce the unit costs of short haul circuits. However, respondents conceded that nothing has been developed which will offset completely the advantage in unit costs which long haul, high volume circuits have.

   281. Respondents have, with some degree of success, made showings designed to meet other arguments advanced by certain parties to the proceeding in relation to "Modified Phoenix." Such arguments include matters relating to the benefits of "feeder" circuits, mitigation of toll rate disparity, and inapplicability to independent telephone companies (which latter is not at issue in this proceeding). In the light of our view as to the basic justification for "Modified Phoenix," we need not treat with these matters in detail.

   282. Respondents have proposed the elimination of what they characterize as "broad averaging" in the determination of interexchange [*101] circuit costs of the local companies. Under present methods, an overall average statewide book cost per interexchange circuit mile is determined by dividing all interexchange circuit costs of such plant of the local company in a State or study area by the total circuit miles in the State. Respondents contend that this method is improper because it spreads the terminal costs of the circuits over the length of the circuits, whereas terminations are not a function of distance. Under respondents' proposal, the terminal costs and line haul costs of all interexchange circuits in a study area would be determined separately, and an average cost per termination and average line haul cost per mile would be computed. These unit costs would then be applied to the count of terminals and miles of interexchange circuits to determine the costs for six categories of circuits. The six categories, based on circuit usage, would be message circuits used for interstate only, intrastate only, and jointly used and special service circuits used for interstate private line only, intrastate private line only, and those used solely for TWX services. The costs thus derived would be directly assigned for those categories used for one service only, and the costs of the jointly used circuits would be allocated on the basis of conversation-minute-miles.

   283. We are not convinced that the refinement sought by the proposed method is necessary for the purposes of jurisdictional separations. Also, respondents propose that, in those categories which they designate as being used for one service only, they would include circuits handling other traffic so long as it did not exceed 5 percent of the usage of the circuit. The justification given is that the intrastate usage on "interstate only" circuits would approximately offset the interstate usage on the "intrastate only" circuits. We fail to see the justification of the more precise method proposed on the one hand, coupled with mere assertion of approximately offsetting corrections on the other hand. Moreover, the proposal to eliminate broad averaging is contrary to the averaging principle we have heretofore found to be acceptable in connection with the "Modified Phoenix Plan." We will continue to use, therefore, the present manual procedures for the determination of interexchange circuit costs of the Bell System associated companies.

D. Exchange Plant

(1) General

   284. The Bell System gross investment in exchange plant amounted to $24 billion in 1966, including an allocated portion of the investment in land and buildings. This amount was assigned to three major categories as follows: Subscriber plant (station equipment and subscriber lines), $15.4 billion, switching plant, $7.0 billion, and exchange trunk plant, $1.5 billion. Of the total gross investment, approximately $2.1 billion was allocated to interstate operations under the provisions of the "Denver Plan" of separations which was introduced in 1965.

   285. A subscriber line is a communication channel which extends between a telephone station, PBX, or TWX station and the central office which serves it. station equipment consists of that equipment and related wiring placed on the customers' premises in which the [*102] subscriber lines are terminated. Unlike other classes of plant, such as dial switching equipment and exchange and toll trunks, subscriber plant is not traffic sensitive, that is, it is not engineered to handle expected volumes of traffic. The purpose and function of subscriber plant are to make both local and long-distance telephone service available to the subscriber through connection with the switching and trunk plant. It does so without regard to the volume or classes of calls the subscriber may originate or receive. Calls made to or from any subscriber station and over any subscriber line may be of short or long duration or may be to or from nearby or distant stations without increasing the quantities of such plant. In essence, therefore, subscriber plant provides subscribers with continuing availability for access to and from the exchange and toll networks of respondents. Particularly important is the fact that the costs of providing subscriber plant are not significantly affected by the amount or types of actual use made of such plant. By contrast, the switching and other classes of plant

are engineered to be responsive to the volumes and types of traffic generated by subscribers, and hence the costs of such plant are directly affected by the expected use to be made thereof by subscribers.

   286. As a consequence of this characteristic of subscriber plant, although such plant is available for use of the subscriber 24 hours a day, it is in actual use, on a nationwide average basis, only 29 minutes out of the 24 hours. For the remainder of the time, the plant stands idle but available for the subscriber's use. In other words, actual use for all services, intrastate and interstate, accounts for only 2 percent of total time such plant is available for use. Of this 2 percent, interstate toll service makes actual use of the plant for an average of 4 percent, intrastate toll for an average of also 4 percent, and exchange service for an average of 92 percent. It is urged that the allocation of the costs of such plant among the services solely on the basis of this pattern of relative use produces inequitable jurisdictional separations results.

   287. Another distinctive feature of subscriber plant derives from the significant differences between the rate structures applicable to exchange and toll services. Exchange service is generally provided under flat rate tariffs which tend to encourage unlimited local calling and unlimited conversation time. Under such circumstances, exchange rate tariffs offer no deterrent to the use of subscriber plant. Toll service, on the other hand, is generally offered as a measured rate service, with each call to be charged for individually, and the charges vary with both distance and time involved. Thus, telephone subscribers are, for economic reasons, inhibited or disciplined in their use of toll services. This is borne out by industry experience which has shown a substantial increase in the exchange service calling rate whenever toll service has been converted to flat rate extended area exchange service.

   288. Because the amount of use of subscriber plant for exchange or toll service has little if any bearing on the amounts and costs of such plant, there

is general agreement on this record that some reasonable method of allocating the costs of these facilities among jurisdictions which is not based wholly on

time in use should be employed.

[*103] (2) The "Denver Plan"

   289. The "Denver Plan" purportedly represents a procedure for separating the costs of subscriber plant between intrastate and interstate operations in a manner which would, in addition to actual use thereof, reflect other considerations. Thus, the "Denver Plan" introduced a "user" factor designed to

reflect the relative number of users of the respective operations (interstate toll, intrastate toll, and exchange) and thereby recognize the relative worth of the availability of this class of plant for local and long distance services. More specifically, it called for allocating subscriber plant between jurisdictions by employing a use-user factor which is a straight averaging of a subscriber line time-in-use (SLU) factor (actual use) and a subscriber line user factor. The latter factor is computed by multiplying (1) the ratio of interstate toll messages to total toll messages by (2) the ratio of the number of toll users to the total number of toll plus exchange users.

   290. The "Denver Plan" also provides for a significant change in the apportionment of the investment in the local dial switching plant which is used to connect subscribers to other subscribers in the same exchange area or to the interexchange facilities for toll calls to other exchange areas. Prior to the

introduction of the "Denver Plan," the separation procedures then in effect ("Charleston Plan") provided for the allocation of the book costs of this plant on the basis of the relative dial equipment minutes of use (referred to as the DEM factor), which is essentially the same as the SLU factor. In the calculation of the DEM factor, the "Denver Plan" provides that the toll minutes of use are to be weighted to reflect the higher switching cost per minute of toll use compared to the cost per minute of exchange use. The weighting applied to the DEM factor approximates the relative actual use that would be determined if each component of plant included in the local dial switching category was studied separately. It was estimated at the time of this innovation that such weighting would increase the interstate DEM factor for the Bell System by about 50 percent. The costs of exchange trunk plant (i.e., circuit plant interconnecting offices within a given exchange area) are divided under the "Denver Plan" into five categories which are assigned directly to a specific intrastate or interstate service when used exclusively for either service or, when jointly used, are apportioned between the operations on the basis of appropriate factors measuring relative use.

   291. The Bell System, the independent telephone group, Western Union, NARUC, and GSA all agree that the provisions of the "Denver Plan" applicable to subscriber plant (i.e., station equipment and subscriber lines) are unsound and should be changed. It is urged in this regard that the "user" part of the "use-user" formula for the jurisdictional separation of the costs of such plant has not worked satisfactorily and should be eliminated -- by the substitution of revised procedures which will be discussed hereinafter.

   292. The Commission, as previously noted, reserved endorsement of the "Denver Plan" because of substantial questions presented as to its reasonableness, particularly with respect to treatment of subscriber plant. On the basis of this record, we have no reason to recede [*104] from our doubts, notwithstanding that the user factor was suggested as a measurement of the relative worth of this class of plant to exchange and toll service. Moreover, there is no rational basis upon which we can deduce the logic for according equal weight to the actual use of subscriber plant and the relative worth of this class of plant to local and long distance service by the straight averaging of both factors. Hence, we believe that no further attention to this facet of the "Denver Plan" is required herein.

(3) Proposals of the Parties

   293. We turn now to the specific proposals advanced on this record by the various parties as to the procedures that should be adopted for the jurisdictional separation of exchange plant. In summary, the Bell System and the NARUC advocate a "use-worth" formula for the allocation of subscriber plant, which formula again seeks to reflect the relative worth of the plant to interstate toll, intrastate toll, and exchange operations. They propose that such relative worth be measured by relating it to the length of haul of calls made by the use of such plant, rather than to the number of users, as in the case of the "Denver Plan." With reference to dial switching equipment and exchange trunk plant, the Bell System and NARUC urge retention of the existing procedures of the "Denver Plan."

   294. The independents advance a similar concept, with some differences in method of application. Their plan, described as the "Comparable Use-Unit Plan," or CUU plan, would also provide for separating dial switching equipment and exchange trunk plant, as well as subscriber plant, on the basis of both time in use and length of haul of originating messages.

   295. Western Union proposes separating subscriber plant costs by the application of a use-availability factor which reflects equally the relative time in use and the opportunity for use of subscriber plant. The proposal is that a 25-percent interstate availability factor would be averaged with the actual use factor in each study area. The 25-percent interstate availability factor is derived simply by saying the exchange plant is available 50 percent for exchange calls and 50 percent for toll calls, and that one-half of the toll availability (or 25 percent of total) is for interstate toll. It would retain the existing procedures with respect to the separation of dial switching equipment and exchange trunk plant. GSA proposes that subscriber plant costs, exchange trunk, and local dial switching equipment costs be combined and separated by a common factor reflecting peak monthly subscriber line minutes of use, determined separately for exchange, State toll, and interstate toll services.

(4) Discussion of Proposals

   296. Each of the foregoing proposals reflects a common objective, namely, to accomplish an apportionment of exchange plant costs which is not based entirely on actual time in use or occupancy of such plant by exchange and long distance services. This objective is not without merit, particularly as it applies to subscriber plant, in view of the distinctive characteristics of such plant, as described above, in comparison with other classes of plant.

    [*105] 297. The most significant feature of the "Denver Plan," regardless of the infirmities in its method of application, is that it sought to give recognition in a jurisdictional allocation of subscriber plant to the value of its availability for toll use, as well as to its actual use in the different operations. Although respondents were the principal author of the "Denver Plan, " they recommend, in this proceeding, that relative worth of subscriber plant to the interstate toll, intrastate toll, and exchange operations can be better recognized by relating it to the length of haul of calls made by the use of such plant rather than to the number of users. The NARUC and independents support this concept although, as previously indicated, the independents strongly advocate a somewhat different application of the concept.

   298. Unlike the "Denver Plan," which sought to reflect the value of availability of subscriber plant for toll services, the instant probasic

assumption that a unit of use of subscriber plant is worth more, or that subscriber plant has greater utility or value, as a means of communication between customers as the distance between them increases. The proponents for this approach urge that the relative worth of the use of subscriber plant can be reasonably measured by means of factors related to the average lengths of haul of the originating messages. They also contend that a reasonable measure of the effects of distance on the relative worth of the use of subscriber plant to the interstate and intrastate operations can be developed mathematically from the relationships among the rates for the several mileage steps in the interstate and intrastate toll rate schedules, which typically provide for increased charges per minute of use as distances increase, but at a diminishing rate of increase in the charge.

   299. The philosophy underlying this conceptual approach to the jurisdictional allocation of subscriber plant costs is summarized by respondents in paragraph 51 of their proposed findings and conclusions as follows:

   A telephone call on the average is more valuable as a means of communication between customers as the distance between them increases, because alternative means involve increasing outlays of time and money as the distance between the communicants increases. This economic fact has been reflected in the rate schedules for telephone service. In the typical toll rate schedules, the charge for a minute of conversation is greater at longer distances than at shorter distances. In the exchange rate schedules, the local rates are higher in the larger exchanges, where the numbers of telephones available for connection in the exchange are greater and the distances called locally tend to be longer on the average than in smaller exchanges. Thus, the relationships between mileage steps in toll rate schedules and average mileage for exchange calls provide a basis for measuring the effects of distance on the relative worth of the use of subscriber plant to the respective services.

   300. The treatment of distance, as the only measure of worth or value of exchange plant to the toll operations, is subject to question. To be sure, distance does give an element of value to long-distance calls. Users do pay more for longer-haul calls than for calls of shorter distances. From this standpoint, the value of the call to the user may be said to correspond to what he is willing to pay for it. From the respondents' standpoint, the rates collected for calls of longer distances [*106] are greater than the rates collected for calls of lesser distances. But this is simply because respondents ' costs of furnishing service tend to increase with distance. In our view, there are so many other factors which go to determine the value to the user of a given call that we believe that distance, as the sole criterion for measuring the worth of subscriber plant to exchange, intrastate, and interstate services, is inadequate and unconvincing. In local telephone service, although distance is no factor in pricing, business subscribers pay more for the service than do residential subscribers. This is because the value of the service to the business subscriber is considered greater than it is to the residential subscriber. We cite this as an example of one factor other than distance that represents value. Certainly, as among individual users, the value of telephone calls cannot be measured by distance alone.

   301. The function of subscriber plant, as cited in the separations brief of respondents at page 42, "is to make available to the subscriber both exchange and toll service, without regard to the volume or classes of calls the subscriber may generate." In this context, the real significance of distance in the jurisdictional allocation of subscriber plant is the restrictive effect of toll rates on the use of such plant in toll operations. This restriction results from the fact, previously noted, that toll users must pay for each call on the basis of distance and time involved. Thus, the question presented is how

best to make allowance for this restrictive effect of distance on the interstate use of subscriber plant.

   302. The effect, although not the design, of the CUU and use-worth plans gives at least partial recognition to the restrictive nature of the distance in the interstate toll rate schedule. They do not, however, give recognition to other factors such as the deterrent effect of the per call basis of charging.

   303. Another disparity between conception and application is also present in the case of the CUU plan of the independents. At page 26 of their brief and proposed findings and conclusions, the independents, after noting the effects on actual use of exchange plant resulting from differences between the rate structures for exchange and toll services, state:

   The comparable use-unit concept provides the proper recognition of the utility or value of the exchange plant for unrestricted exchange service and for

restricted toll service. Here again we fail to see how relating the worth of exchange plant to the distance of calls actually using that plant compensates properly for the inhibiting effect of the measured toll rate schedules on toll usage.

   304. Moreover, we see no justification for that aspect of the CUU plan of the independents which would apply the concept of distance as a measure of relative worth to dial switching and exchange trunk plant as well as subscriber plant. Switching and trunk plant are traffic sensitive, in that they are engineered to handle anticipated volumes of calls. Hence, even though these classes of plant are designed to operate in an integrated manner with subscriber plant, as urged by the independents, it would appear to us that actual use of that plant by the several services is, in and of itself, a fair and equitable basis [*107] for allocating its costs for jurisdictional purposes. In any event, the evidence in this record does not persuade us to conclude otherwise.

   305. In view of the foregoing, we cannot accept the concept of distance as the sole index of measuring the availability of subscriber line plant to the interstate operation, although we agree that it has an effect on the measurement of such availability. This is particularly true where the distance factor is used as the sole index in formulae which purport to measure the worth factor with the mathematical precision which is attributed to the Bell and independent plans. Aside from the difficulties which we have with distance as the basic concept of such plans, they also contain certain defects which detract from their validity as efforts to apply this basic concept.

   306. Both plans use the interstate message toll rate schedule as a measure of the relative value of the exchange plant to the interstate toll service. The interstate rate schedule does, to some extent, reflect value of service. However, a major factor entering into the ratedistance relationship is the specific revenue requirement such schedule is desired to recover in total at a given time. Although, within the framework of this requirement, value of service, as well as cost of service, relationships have always been reflected generally, the ratedistance relationship, as established by the various rate zones, is affected to a substantial degree by the traffic patterns between rate centers in designing rates to produce the desired amount of revenue. Thus, the use of a curve based on the interstate rate schedule in the manner in which it is used in the Bell and independent formulae, to produce the precision in separations which these formulae purport to achieve, is of doubtful validity.

   307. The formulae contain, as an essential element, an amount representing the average distance of an exchange call. In the case of Bell, this is assumed to be 3 airline miles and, in the case of the independents, 8 route miles. Respondents made no study of actual calls to obtain this figure. Rather, it was developed from averages of estimates of certain operating personnel in various operating areas which were, in turn, averaged for the Bell System. It is not clear from the record just how the 8 route miles used in the independents' formula was determined, but it, too, appears to have been an estimate and is not based on an actual study. Moreover, the use of route miles in the same formula with airline miles representing interstate distances appears to be logically fallacious. The importance of these mileage figures to the validity of the formulae is illustrated by the fact that a difference of 1 mile in the Bell formula represents a difference of approximately $50 million in revenue requirements.

   308. Bell has used in its formula the 0.3 power curve, while the independents have similarly used the cube root curve (0.333). Both were allegedly chosen because they correlate closely with the curve of the interstate rate schedule. Aside from the question of the validity of using the interstate rate schedule in this manner, it is mathematically apparent that a number of other power curves are similarly closely correlated with the curve of the interstate schedule. The use of such other curves, however, would produce considerably different dollar results in revenue requirements assigned to the interstate operations. [*108] For instance, use of the 0.2 power curve in the Bell formula would reduce such assignment by approximately $200 million.

   309. The plan for separating exchange plant suggested by Western Union, whereby an availability factor of 50 percent for exchange and 25 percent each for State and interstate toll is assigned, has the appeal of simplicity. Unfortunately, its greatest vice lies in that very simplicity; it is completely arbitrary. Such an arbitrary assignment of 25 percent has no demonstrated relationship to the value to the interstate toll users of the availability of the subscriber plant for such service. Indeed, it would be almost as easy to rationalize a 1/3-1/3-1/3 relationship of the three principal uses of the plant. Moreover, the simple averaging of the availability factor with the use factor has, as we have stated regarding the "Denver Plan," no rationale basis.

   310. GSA's plan was stated in broad terms without details as to how it would be implemented and without sufficient detailed facts to test its reasonableness. Aside from the theoretical objections to this plan voiced by certain of the parties, we find that the record is insufficient to establish its reasonableness.

(5) Conclusions

   311. On the basis of this record, we are satisfied that actual use is a proper and equitable standard for the allocation of the costs of exchange plant between State and interstate jurisdictions, except in the case of the subscriber plant consisting of station equipment and subscriber lin