THE EVOLUTTON 0F ACCOUNTING THOUGHT AND PRACTICES RELATED TO BOND REFUNDING Thesis for the Degree of D. B. A. . MICHlGAN STATE UNIVERSITY GEORGE H. NEAL 1971 yuqc q This is to certify that the thesis entitled THE EVOLUTION OF ACCOUNTING THOUGHT AND PRACTICES RELATED TO BOND REFUNDING presented by George Howard Neal has been accepted towards fulfillment of the requirements for D.B.A. Accounting - Business degree in __ Mm, Major professor Date November 2, 1971 0-169 “:1, BIN05NG BY ‘ K HUAU & SUNS HI BUUK BINULRY WC, LIBRARY BINDEP‘ ' .......... 'nnulonu ABSTRACT THE EVOLUTION OF ACCOUNTING THOUGHT AND PRACTICES RELATED TO BOND REFUNDING BY George Howard Neal Accounting for the several types of cost incurred in the refunding of bonds was the subject of considerable controversy during the 19305. Continuing interest shown in the prOblem by accounting and regulatory authorities during recent years prompted this study. More specifically, its objectives are as follows: 1. To review develOpments in accounting thought related to bond refunding. 2. To compare practices and analyze size relationships between refunding-related costs, earnings, and dividends of public utility and nonregulated industrial firms. 3. To determine inferable support for alter- native accounting procedures from recent models of accounting theory. 4. To investigate the question of whether different procedures may be justified for public utilities than for nonregu- lated firms. The literature indicates that theoretical argument has revolved around whether refunding—related costs should be subject to immediate write-off or one of several gradual George Howard Neal write-off procedures. Advocates of immediate write-off tend to View these costs as a loss realized upon refunding. Advocates of gradual write-off usually attempt to relate the costs to perceived periods of interest savings. Data concerning practices by samples of utility and industrial firms during the periods 1936-1945 and 1956-1965 were obtained largely from questionnaires and previous re— ports. Utilities tended to use gradual write—off to a greater extent than immediate write-off during the earlier period. But there was increased use of immediate write-off during 1956-1965. A majority of industrials used immediate write-off during both periods. Utilities deferring write-off tended to have larger amounts of such costs relative to retained earnings and net income than industrials or utilities using immediate write— off during 1936—1945. Percentages were smaller and less conclusive for the 1956-1965 period. A chi-square test indicated significantly higher median dividends for utili— ties using immediate write-off during both periods. Similar results were not obtained for industrials. The results of a correlation test of refunding-related costs and dividends were not strong but did not refute the inference that the relative size of these costs was at least partially George Howard Neal responsible for use of gradual write-off procedures by pub— lic utilities during 1936-1945. Accounting for bonds outstanding at current values and their related interest at market rates is proposed by Eldon S. Hendriksen in Accounting Theory. Complete applica- tion of this procedures, called the full-accrual approach in this study, requires continuous adjustment of premium or discount since the bond liability is always stated at the amount required for current redemption. Prior complete accrual of this amount leaves no unamortized cost related to a refunded issue in the accounts. The Hendriksen model is described, illustrated, and analyzed in the study. Guidelines stated by Raymond J. Chambers in Accounting, Evaluation and Economic Behavior and by Robert T. Sprouse and Maurice Moonitz in Accounting Research Study No. 3 are summarized and interpreted as lending support for the full- accrual approach. The standard of relevance of accounting data to expected use, as proposed in the American Accounting Asso- ciation's A Statement of Basic Accounting Theory, is inferred to permit capitalization and amortization of refunding-related costs (in the historical-cost model) when such procedure conforms accounting data to that used George Howard Neal by utility commissions in the rate-making process. But the A. A. A. model prOposes the reporting of current-cost as well as historical-cost data, and this current-cost model, as applied to bond liabilities, yields results that agree with the Hendriksen full-accrual approach. The writer recommends (l) conformity of externally reported utility data to that used in rate making, (2) re- jection of current practice of reporting what appear to be irrelevant data, (3) further consideration of current costs for use in rate making and financial reporting, and (4) research of user views and predictive assistance of data based on different procedures of accounting for debt. THE EVOLUTION OF ACCOUNTING THOUGHT AND PRACTICES RELATED TO BOND REFUNDING BY v‘ George HQRNeal A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF BUSINESS ADMINISTRATION Department of Accounting and Financial Administration 1971 @C0pyright by GEORGE HOWARD NEAL 1972 ACKNOWLEDGMENTS I wish to express my sincere appreciation to several individuals who have contributed greatly toward my success in the graduate program at Michigan State University. I am especially indebted to Professor Roland F. Salmonson, chairman of my dissertation committee, for his many helpful criticisms and suggestions. His standards of academic excellence are matched only by his infinite patience and understanding. For their consultation as members of the committee, I am grateful to Professors Ronald M. Marshall and Harry M. Trebing. I also wish to thank Dr. Trebing for helping me to obtain financial assistance. For typing the final draft of the dissertation, I am grateful to Mrs. Josephine McKenzie. Further, I am in- debted to the entire faculty and administration of the Graduate School of Business Administration for the Oppor— tunity to pursue my studies at Michigan State University. And, of course, I appreciate the continual encouragement of my wife, LaDelle. ii TABLE OF CONTENTS LIST OF TABLES. . . . . . . . . . . . . . . . . . . . LIST OF EXHIBITS. . . . . . . . . . . . . . . . . . CHAPTER I. INTRODUCTION . . . . . . . . . . . . . . . . Objectives of the Dissertation . . . . . . Hypotheses and Methodology . . . . . . . Hypothesis No. l Hypothesis No. 2 . . . . . . . . . . . Hypothesis No. 3 Background for the Study . . . . . . . . . The Practice of Bond Refunding . . . . . Related Accounting Problems. . . . . . . Significance of the Study. . . . . . . . . Limitations in Sc0pe . . . . . . . . . . . Note on Terminology. . . . . . . . . . . . Organization of the Dissertation . . . . . II. THE HISTORY OF ACCOUNTING THOUGHT RELATED TO BOND REFUNDING . . . . . . . . . The Basic Nature of Bond Financing Costs . PrOposed Alternative Accounting Procedures Related to Refunding. . . . . . . . . . . Immediate Write-off in the Year of Refunding. . . . . . . . . . . . . . Write-off to Normal Maturity of Retired Bonds . . . . . . . . . . . . . Write-off to Maturity of Refunding Bonds . . . . . . . . . . . . . . . . . iii page vi viii UT 11 11 13 18 20 22 24 27 27 38 39 48 53 Chapter Page Other Preposed Procedures. . . . . . . . 6O Present State of Authoritative Sanction. . . . . . . . . . . . . . . . 63 Summary. . . . . . . . . . . . . . . . . . 69 III. TRENDS AND EFFECTS OF ACCOUNTING PRACTICES O O O O O O O O O O O O O O O O O 7 3 General Practices Related to Bonds Outstanding . . . . . . . . . . . . . . . 74 Practices Related to Refunding During the Period 1936-1945. . . . . . . . . . . 76 The Clendenin Report . . . . . . . . . . 76 The Lemke Report . . . . . . . . . . . . 80 Practices Related to Refunding During the Period 1956-1965. . . . . . . . . . . 83 Regulated Public Utilities . . . . . . . 84 Unregulated Industrial Corporations. . . 86 Tax Effects. . . . . . . . . . . . . . . 87 Procedure-Selection Reasons. . . . . . . 88 Comparative Statistics of Net Income and Retained Earnings Relationship of Bond Financing Costs and Dividend Payouts . . . . . . . . . . . . . . . . . 91 Summary. . . . . . . . . . . . . . . . . . 108 IV. APPLICATION OF GUIDELINES FROM RECENTLY DEVELOPED MODELS OF ACCOUNTING THEORY . . . 111 The Hendriksen Model . . . . . . . . . . . 112 The Full—Accrual Approach. . . . . . . . 114 Historical-Cost Procedures: Hendriksen Evaluation. . . . . . . . . . . . . . . 125 The Sprouse-Moonitz Postulates—Principles Model . . . . . . . . . . . . . . . . . . 128 iv Chapter Position on Assets . . . . . . . . . Position on Liabilities. . . . . . . The Chambers Current-Cash—Equivalent MOde l O O O O O O O O O O O O O O O 0 Summary. . . . . . . . . . . . . . . . V. APPLICATION OF AMERICAN ACCOUNTING ASSOCIATION STANDARDS AND GUIDELINES. . Implications for General-Purpose Reporting . . . . . . . . . . . . . . Accounting Under Regulation. . . . . . Costs of Debt Capital in the Rate Base 0 O O O O O O O O O O O 0 Costs of Debt Capital in the Rate of Return. . . . . . . . . . . Effects of Alternative Procedures. . Evaluation of Alternative Procedures Summary. . . . . . . . . . . . . . . . VI. SUMMARY AND CONCLUSIONS. . . . . . . . . Historical Review. . . . . . . . . . . Evaluation of Procedures . . . . . . . Recommendations. . . . . . . . . . . . BIBLIOGRAPHY. . . . . . . . . . . . . . . . . . . Page 130 132 135 145 149 151 158 161 165 170 176 187 191 196 200 205 210 .r& a; u a 6 I. B. T‘\ Table 10. 11. LIST OF TABLES Page Practices of Accounting for Financing Costs Related to Bonds Outstanding 1956-1965 . . . . 76 Accounting Treatment of Unamortized Financing Costs Applicable to Bonds Refunded as Reported by Clendenin 1941 . . . . . . . . . . 78 Accounting Treatment of Unamortized Financing Costs Applicable to Bonds Refunded as Reported by Lemke for the Period 1936-1945 . . 81 Accounting Treatment of Unamortized Financing Costs Applicable to Bonds Refunded During the Period 1956-1965 . . . . . . . . . . . . . 85 Accounting Treatment of Tax Effects Related to Refundings by Public Utilities During the Periods 1936-1945 and 1956-1965. . . . . . . . 88 Stated Reasons for Choice of Procedures 1956-1965. . . . . . . . . . . . . . . . . . . 89 Frequency of Data by Source and Period . . . . 93 Wald-Wolfowitz Runs Test of Unamortized Bond Financing Costs and Dividends as Per Cents of Retained Earnings 1956-1965 . . . . . . . . 97 Percentage Relationships of Refunding-Related Items to Retained Earnings and Net Income Before Charge-Off. . . . . . . . . . . . . . . 102 Results of Median Test of Dividends of Firms Using Immediate and Deferred Write- off Procedures for Financing Costs Related to Bonds Refunded. . . . . . . . . . . . . . . 104 Dividends Declared in the Year of Refunding as Per Cents of REBCO and NIBCO and as Corre- lated with Bond Financing Costs Charged Off as Per Cents of REBCO. . . . . . . . . . . . . 107 vi Table 12. 13. 14. Page The Full-Accrual Approach Illustrated. . . . . 122 Computation of Utility Return and Net Income Using Three Refunding-Related Write-off Procedures with Tax Allocation . . . . . . . . 174 An Illustration of Estimated Cash-Flow Benefits of a Timely Refunding . . . . . . . . 181 vii Exhibit I. LIST OF EXHIBITS Page Accounting Procedures for Items Related to Bonds Refunded as Approved by Regulatory Authorities and Professional Accounting Organizations . . . . . . . . . . 65 viii ‘Ia .,.. 4 v... CC! I... q ”a . :"A I‘- lb" (I' I II, (I) CHAPTER I INTRODUCTION Historically, one of the major problems in account- ing has been the assignment of items of cost and revenue to the proper periods to achieve realistic measurement of busi- ness income. Closely related is the problem of fair pre- sentation of assets and liabilities in the balance sheet. Discount, issue cost, and premium related to long—term debt have been among the items of controversy as to prOper accounting treatment. A particularly troublesome question has been that of apprOpriate disposition of these items when bonds are refunded prior to maturity. Refunding usually involves retirement of a bond issue by using cash proceeds of other bonds issued for the purpose of acquiring funds to be used to retire the original debt. A direct exchange of issues is also considered a form of refunding. The central concern of the accounting problem has been whether financing costs applicable to the retired issue and remaining in the accounts at the date of refunding 2 should be written off immediately upon retirement or de- ferred and charged off gradually over some future period. Preposed alternatives have also included some prior dis- position of the items such that no balance remains upon the occurrence of refunding. Accounting and regulatory authorities have issued various pronouncements concerning the matter. Pro and con discussion of procedures has appeared in the literature of accounting theory many times. But, to date, there has not been general agreement in the accounting profession as to superiority of a particular procedure sufficient to exclude acceptance of other alternatives. Objectives of the Dissertation In view of the historical controversy and evidence that the problem still exists, the areas of investigation in this study involve two phases of research. The first phase consists of an historical review of thought and prac- tice trends related to accounting for bond refunding, accompanies by an analysis of size-relationships between financing costs applicable to bonds refunded and net in- 1 come, retained earnings, and dividends. The second phase 1For definition of financing costs, see pp. 19—21, infra. 3 consists of an investigation to determine what procedures may be supported by application of guidelines prOposed in recently developed generalized models of accounting theory. The specific objectives of the study are: 1. To show develOpments in accounting thought related to refunding, with emphasis upon pro and con argument prevailing in the literature near the time of approval of various procedures by accounting bodies and regulatory authorities. 2. To determine practice trends, if any, indi- cated by procedures used during periods of attention to the problem by accounting and regulatory groups and what pragmatic basis for adOption of certain procedures may be inferred from the size-relationships analysis. 3. To determine whether logically consistent. application of guidelines comprising the main framework of each of three recently develOped accounting theory models leads to the full-accrual approach or permits some historical cost procedure. 4. To determine whether different procedures appear warranted for public utilities than for nonregulated industrial firms by appli— cation of usefulness oriented evaluation standards and communication guidelines 1Eldon S. Hendriksen, AccountingfiTheory (Rev. ed.; HOmewood: Richard D. Irwin, Inc., 1970), pp. 450-455, con- tains perhaps the strongest statement of the case for what is referred to in this dissertation as the full-accrual approach. The procedure consists of accounting for the bond liability and related interest charge at current market price and yield rate. Complete application of the procedure achieves automatic adjustment for premium and discount as part of the accrual. 4 prOposed in the American Accounting Asso- ciation's A Statement of Basic Accounting Theory. Hypotheses and Methodology Research related to the first objective consists of summarization and discussion of theoretical and practical argument offered in the literature by pr0ponents and critics of various procedures. This review includes state- ments of regulatory commissions, federal income tax authori- ties, and professional accounting organizations as well as preposals of individual writers. To show the variety of procedures currently in effect, a tabulation of those approved by the various authorities will be presented. This includes the results of a questionnaire survey of the state commissions. Certain questions as to possible findings related to the second and third objectives prompted the researcher ‘ to formulate working hypotheses for these aspects of the study. These are given as tentative assertions of possible findings of the research. Hypothesis No. 1 and Hypothesis No. 2 are related to the second objective. Hypothesis No. 3 is related to the third objective. The statement of each is followed by a description of the methodology to be used in its investigation and a statement of the basis for adoption of the hypothesis. The first hypothesis related to the second objective is given below. Hypothesis No. l.--A pragmatic basis for gradual write-off of financing costs appli- cable to bonds refunded may be inferred for public utilities, in contrast to industrials, on the grounds that utility balances of such costs tend to be of sufficient size to sig- nificantly impair earnings available for dividends. Using a list of refunding firms reported by Lemke as a working nucleus and obtaining additional information from financial summaries published in Moody's Manuals, amounts of discount, issue cost, and call premium applicable to the issues retired were determined for 40 utility cases and 33 industrial cases of refundings during the period 1936-1945.1 By a combination of investment manual sta- tistics, Securities and Exchange Commission reports, and questionnaires similar data were obtained for 36 utility and 27 industrial refundings occurring during 1956-1965. As a test of Hypothesis No. l, the data for these two periods are analyzed and reported in the study as follows: 1. Unamortized financing costs applicable to refunded bonds, the amount charged 1Bernhard C. Lemke, "The Treatment of Bond Discount and Premium in Connection with Refundings," (unpublished Ph.D. thesis, University of Minnesota, 1946), pp. 175-176, 184. 6 off in the year of refunding, and the dividend for that year are shown as per cents of retained earnings and net in- come before charge-off. Cross compari- sons are made for: a) Public utilities vs. nonregulated indus- trials. b) Immediate write-off vs. deferred write—off. c) Decade ended December 31, 1945 vs. decade ended December 31, 1965. 2. The results of certain statistical tests are reported. These are: a) A runs test of 1956-1965 utility and indus— trial samples assumed to be representative of pOpulations consisting of known refund- ings for the period. Randomness was assumed for the 1936-1945 samples since these included many of the cases reported, although analyzed differently, in prior studies. b) A median test for differences in dividends of firms deferring write—off of refunding- related financing costs and those of firms writing these costs off entirely in the year of refunding. c) A correlation test of association between dividends and refunding—related charges written off in the year of refunding. The basis for Hypothesis No. l was provided pri- marily by the results of the Lemke study. That report showed aggregate amounts of unamortized bond financing costs appearing in balance sheets at or near refunding as . . 1 per cents of retained earnings and total assets. Amounts lIbid. 7 applicable to refunded issues specifically were not distin- guished in most cases. Other sources referring to similar size-relationships have not shown quantitative documentation. The second hypothesis related to the second objec- tive is stated below. Hypothesis No. 2.—-There has been a trend during the 1956-1965 period toward greater use of write-off over the life of the new issue by public utilities than indicated by earlier studies for the 1936-1945 period. Methodology for the second hypothesis consists of tabulations of practices for 1956—1965 compared to patterns reported by Lemke and Clendenin for the 1936-1945 period.1 The Clendenin study reported practices determined by a sur- vey in 1941 but did not include any analysis of quantita- tive data. Approval of write-off over the life of the new issue by the Accounting Principles Board of the American Institute of Certified Public Accountants in 1965 was the basis for Hypothesis No. 2. Introductory remarks in Opinion No. 6 indicated that currently accepted practice influenced revisions of positions previously taken in the 1Ibid.. pp. 180, 193. J. C. Clendenin, "How 118 Major Corporations Account for Bond Discount," The Journal of Accountangy, 8 Accounting Research Bulletin series.l Also, since antici- pated rising interest rates as well as currently lower rates were mentioned as circumstances justifying the pro— cedure, there arises the question of a tendency of firms to relate accounting write—offto the life of the new issue when that period is employed in rate-of-return calculations for capital budgeting purposes. A third hypothesis is related to the third objec- tive and is stated below. Hypothesis No. 3.——Logically consistent application of guidelines contained in the deductive models used in the study, including those which do not specifically discuss re- funding, provides greater support for the full—accrual approach than for other proce— dures of accounting for long-term debt and related interest. The methodology related to Hypothesis No. 3 con- sists of presentation of the full-accrual approach as developed and supported in the Hendriksen book and exam— ination of two other accounting theory models as follows: 1. Sprouse—Moonitz Postulates—Principles Model: Sprouse, Robert., and Moonitz, Maurice. "A Tentative Set of Broad Accounting Principles for Business Enterprises," Accounting Research Studero. 3. American Institute of Certified Public Accountants, 1962. 1American Institute of Certified Public Accountants, Accounting Principles Board, "Status of Accounting Research Bulletins," Opinion No. 6 (New York, 1965), pp. 38, 43. 9 2. Chambers Current-Cash-Equivalent Model: Chambers, Raymond J. Accounting, Evaluation and Economic Behavior. Englewood Cliffs: Prentice—Hall, Inc., 1966. The use of the models in this study will consist of describing the general framework of the approach taken in each model and drawing from it the guidelines shown or in- ferred to be related closely to accounting for long-term debt. The author's supporting argument will be summarized, but extensive derivations will not be attempted. The pur- pose of this analysis is merely to show what solution to the bond-refunding accounting problem would follow from logi- cally consistent application of the guidelines contained in the models. The basis for the third hypothesis lies in the general controversy over the many alternative procedures of accounting for various types of transactions, of which bond refunding is just one example. In 1959 the American Insti— tute of Certified Public Accountants launched a research program aimed at reduction of inconsistencies and narrowing the areas of differences in accounting practice by formu- lating accounting principles based on research other than . . . l . . just surveys of practices that eXlSt. Since that time 1American Institute of Certified Public Accountants, Special Committee on Research Program, "Report to Council of the Special Committee on Research Program," The JOurnal of Accountancy, CVI (December, 1958), 62—63. 10 several proposals purported to be contributions toward the deve10pment of theoretically sound accounting guidelines based largely on deductive logic have appeared in the literature. The essence of Hypothesis No. 3 is that three such models can be shown to support the full-accrual approach, essentially as developed by Hendriksen. Most models of financial accounting theory examined by this researcher appear directed primarily to general- purpose income measurement and reporting. But the evalua— tion standards and communication guidelines prOposed in the American Accounting Association's Statement represent a pragmatic approach to accounting theory deve10pment related to the criterion of usefulness in achieving desired results. In this study, the model is used to evaluate procedures related to utility accounting under rate regulation as Opposed to those of firms Operating in the nonregulatory environment. No particular hypothesis is asserted with respect to the fourth objective of the dissertation, although there is a discussion of the extent of support for the full— accrual approach as well as others in the light of the "standards and guidelines." 11 Background for the Study The Practice of Bond Refunding Historically, the usual purpose of refunding bonds has been to reduce the cost of long-term debt financing during periods of low interest rates.1 But contraction of a cumbersome number of different issues, getting rid of in- denture restrictions, and postponement of ultimate repayment have all been cited as reasons for refunding.2 While little refunding occurred until around the turn of the century, about 60 per cent of all bonds retired during the entire period 1900—1943 were refunded.3 Refunding issues accounted for around 45 per cent of approximately $72 billion total face amount of straight bonds offered to investors during these years.4 Most of the refunding was done by railroads and public utilities, but refunding bonds comprised about 35 per cent of total offerings by nonregu- lated firms during the period. 1Raymond P. Kent, Corporate Financial Management (Homewood: Richard D. Irwin, Inc., 1964), pp. 745-746. 21bid. 3W. B. Hickman, The Volume of Corporate Bond Financing Since 1900 (Princeton: Princeton University Press, 1953). 41bid.. pp. 111, 113. 5Pearson Hunt, Charles W. Williams, and Gordon Donaldson, Basic Business Finance (Homewood: Richard D. Irwin, Inc., 1961), pp. 562-563. 12 A record volume of refunding occurred during the 1944-1947 period when interest rates were lower than for any comparable period since the 1920s.1 Since the 19403, the practice has continued but at a relatively lower level. Upturns have been evident during the 1954-1955 and 1962- 1963 periods.2'3 Much of the evidence indicates that bond refunding is a major financial practice primarily of regulated in- dustry, although it is engaged in to some extent by non- regulated industrials. Heavy fixed capital requirements and the conditions of rate regulation tend to limit avail- ability of equity funds to the regulated firm. Hence, public utilities continually rely on bond issues as a major source of long—term funds. Securities and Exchange Commission policy under the Public Utility Holding Company Act of 1935 requires that bonds and preferred stock of companies subject to the act must be fully refundable at the option of the issuing lPearson Hunt, Charles W. Williams, and Gordon Donaldson, Basic Business Finance (Homewood: Richard D. Irwin, Inc., 1961), pp. 562-563. 2Ibid. 3Arthur J. Stegeman, "tlst Annual Statistical Report-1965," Electrical World, CLXIII (February 22, 1965), 113. 13 corporation.1 The commission reported that a majority of electric and gas issues offered at competitive bidding from mid-1957 to mid-1966 were refundable bonds.2 It does not seem unreasonable to anticipate that many high-yield issues of recent years will be refunded in the event of future rate declines, although it is not a purpose of this study to predict such phenomena of the financial market-place. At any rate, bond refunding continues to a suffi- cient extent to attract the attention of professional accounting bodies and regulatory authorities. It has also been discussed in the literature of capital budgeting in recent years. Related Accounting Problems The accounting problems related to bond financing stem largely from issuance at prices other than face value, perhaps the rule rather than the exception as bond prices and interest rates react to conditions in the securities 1U. S. Securities and Exchange Commission, 32nd Annual Report (Washington: U. S. Government Printing Office, 1966), p. 82. 2Ibid.. pp. 82-83. 3Oswald D. Bowlin, "The Refunding Decision: Another Special Case in Capital Budgeting,“ The Journal of Finance, XXI (March, 1966), 55-68, and Earl A. Spiller, Jr., "Time- Adjusted Rate of Refunding," Financial Executive, XXXI (July, 1963), 32—35. 14 market. By the time refunding reached peak pr0portions in the 19308, the view of premium and discount as adjustments of the interest rate and, hence, of the effective bond lia- bility was receiving considerable support in the literature. But the procedure usually followed in practice was to show the liability at face value and carry discount as a deferred charge on the asset side of the balance sheet.1 The practice of refunding created additional prob- lems with respect to accounting for remaining balances in- cluding some combination of unaccumulated discount, issue costs, call premium, incidental call costs, and duplicate interest charges applicable to the bonds retired. Some examples of the different views prevailing in the 19305 are as follows: 1. The National Association of Railroad and Utility Commissioners approval of immediate write-off as the basic procedure. 2. The state of Wisconsin's stipulation of write-off over the life of the refunding issue to the extent that the resulting "effective" rate was not greater than 1Thomas H. Sanders, Henry R. Hatfield, and Under- hill Moore, A Statement of Accountinngrinciples (New York: American Institute of Certified Public Accountants, 1938), p. 78. 2National Association of Railroad and Utility Commissioners, Uniform System of Accounts for Electric gtilities (New York: The State Law Reporting Company, 1937), p. 14. .- u H 15 that for the retired issue prior to refunding. 3. A writer's recommendation of a hybrid approach involving immediate write-off at refunding of discount and issue cost of the old issue but deferral of call premium over the life of the new issue. Attention to the problem by the American Institute <>f Certified Public Accountants has included the following: 1. Approval of immediate write-off and write- off over the period to normal maturity of the retired issue in 1939.3 2. Recommendation of a charge in the income statement at least equal to the tax saving resulting from deduction of the items related to refunded bonds regardless of the disposi- tion of the remainder to be written off.4 lHerbert C. Freeman, "The Treatment of Unamortized lDiscount and Premium on Retirement in Refunding Operationsx' jghe Journal of Accountangy. LX (October, 1935), 255. 2V. Childs Klug, "The Accounting Treatment of Unamortized Discount and Premium in Utility Refunding (Nperations," The Journal of Land and Public Utility Economics, XII (November, 1938), 411. 3American Institute of Certified Public Accountants, (hummittee on Accounting Procedure, "Unamortized Discount and Redemption Premium on Bonds Refunded," Accounting lageearch Bulletin No. 2 (New York, 1939) (As reprinted in idadexed volume of numbers 1 through 37, 1949, pp. 9-10). 4American Institute of Certified Public Accountants, Chammittee on Accounting Procedure, "Unamortized Discount and Redemption Premium on Bonds Refunded (Supplement). " Jaggcountinngesearch Bulletin No. 18 (New York, 1942) (As J:‘eprinted in indexed volume of numbers 1 through 37, 1949, l6 3. Reaffirmation of the 1939 position in 1953 but expression of a definite prefer- ence for write—off over the original life of the refunded bonds. 4. The previously noted approval in 1965 by the Accounting Principles Board of write- Off over the life of the new issue.2 Since inception of the Institute's new research program, one study sponsored by the organization, Accounting Research Study No. 3,held that unaccumulated discount and call premium on refunded bonds should be written Off as a loss.3 But the research studies do not represent the offi- cial position of the Institute. Official pronouncements are issued by the Accounting Principles Board at the present time. These were formerly given by the Committee on Accounting Procedure in the Accounting Research Bulletin series. Two significant deve10pments on the part of national regulatory authorities have occurred since the 19405. One American Institute of Certified Public Accountants, Committee on Accounting Procedure," Restatement and Revi— sion Of Accounting Research Bulletins," Accounting Research Bulletin No. 43 (New York: 1953), p. 142. 2A. I. C. P. A., Accounting Principles Board, Op. cit., p. 43. 3Robert T. Sprouse and Maurice Moonitz, "A Tentative Set Of Broad Accounting Principles for Business Enterprisesfl Accounting Research Study No. 3 (New York: American Insti- tute of Certified Public Accountants, 1962), p. 40. 17 of these was adOption in 1958 of a variation of payback time as an allowable alternative to immediate write-off.l This involves write-off over the period required for interest savings to accumulate to the amount to be written Off and may be used without prior permission of the commis- sion. Formerly, immediate write—off was the basic require- ment, but departures from the general rule have apparently been liberally allowed. The other deve10pment was a pro- posal in 1971 by the Federal Power Commission to adOpt the three procedures approved by the American Institute of Certified Public Accountants.2 Surveys of practices around the time of the Insti- tute's publication of Bulletin No. 2 disclosed that public utilities tended to adOpt some form of gradual write-off to 3 4 a much greater extent than nonregulated industrials. ' Reasoning offered in support of the various pro- cedures has included references to the completed-transaction 1National Association of Railroad and Utility Com- missioners, Uniform System of Accounts for Class A and Class B Utilities (Washington: N. A. R. U. C., 1959), pp. 59-60. 2U. S. Federal Power Commission, "Notice of Pro- posed Rule-making," Docket No. R-424 (August 6, 1971), pp. 5, 13-14. 3Lemke, Op. cit., pp. 180-193. 4Clendenin, Op. cit., pp. 40-43. 18 concept, the "doctrine" of conservatism, and different views concerning matching of costs and benefits. In addition to theoretical argument, pragmatic considerations, especially in relation to public utility circumstances, seem to have been given significant attention in judging the accept- ability of procedures.1 Significance of the Study Any area of accounting in which a considerable diversity Of Opinions and practices exists appears to be fair game for further research. Since bond refunding is such an area, perhaps the analysis and conclusions of this study may contribute toward narrowing the areas of differ- ences in accounting procedures by suggesting criteria for acceptance or rejection of alternatives. While it may be argued that bond financing costs might not be of material amount in the statements of a given firm, the line of distinction between materiality and immateriality is sometimes quite nebulous. Furthermore, inapprOpriate accounting for several "immaterial" items may aggregate to considerable distortion in a financial report. Specific factors that appear indicative of the 1George 0. May, "Accounting Principles and Regula- tory Expediency," The Journal of Accountangy, LXXI (February, 1941), 116. l9 relevance of the study as a current research project include the following: 1. The Securities and Exchange Commission policy on refundability of bonds under the Public Utility Holding Company Act of 1935 and the large proportion of total Offerings consisting of refundable bonds in recent years. 2. Approval by regulatory authorities and pro- fessional accounting organizations of dif— ferent procedures within the 1956-1965 period and the more recent prOposed rule— making of the Federal Power Commission. 3. Recent attention in financial literature to bond refunding as a capital budgeting problem. 4. Recent evidence of interest in the problem by the Canadian Institute of Chartered Accountants as well as the American Insti- tute Of Certified Public Accountants.2 5. Lack of quantitative documentation in the literature of size relationships between financing costs applicable to refunded bonds and retained earnings, net income, and dividends. 1Moody's Public Utility Manual (August, 1969» pp. a144 and a174, contains lists of bonds that are noncallable and nonrefundable, at least for a number of years. Nonre- fundable bonds are noncallable for the purpose of being refunded. Regulatory authorities usually do not allow non- call provisions, but exceptions are made in the interests of parties affected. S. E. C. policy is aimed at preventing diseconomies in raising capital that might result from issuance of bonds that cannot be refunded, according to Public Utilities Fortnightly, LX (September 12, 1957), pp. 417-418. 2H. S. Moffet, Accounting for Costs of Financing (Toronto: The Canadian Institute of Chartered Accountants, 1964), p. 19. 20 6. The Opportunity afforded by the study to examine a long-standing controversial problem by reference to theoretical models prOposed during the recent era of emphasis upon the deductive approach to accounting principles development. Limitations in SCOpe In the foregoing statements of problems, objectives, and research methods, some limitations were stated in the interest of clarity. Certain other constraints are con- sidered necessary to keep the project within manageable pro- portions in relation to time limits, costs and feasibility of the research, and the aims of the study as defined. Although interest in accounting for bond refunding has recently been indicated by neighboring foreign account- ants, the present study is confined to the United States. Selection of the specific period 1956—1965 was made for two reasons as follows: 1. TO present data for a recent period during which national regulatory and accounting authorities have taken official action on refunding accounting. 2. TO present data for a recent period of comparable length to the earlier period represented by the Lemke study. Except as to reported practices, the present study is not strictly comparable to that of Lemke because of the different variables and techniques used. The Lemke study 21 included percentage relationships only of bond financing costs to retained earnings and total assets.' In this study, no analysis is made of assets. Instead, computations were made for net income and dividends as well as retained earnings. This was done under the assumption that effects on current and prior earnings and, hence, desired dividend position are more likely to influence the choice of write- Off procedures than relationships between financing costs and total assets. Data representing practices and size—relationships for regulated industry are confined primarily to the class of public utilities generally consisting of gas, electric power, and water companies. A few telephone companies were included. Quantitative data concerning railroad re- fundings were omitted because of limited availability and to provide comparisons for firms similar to those primarily reported upon by Lemke and Clendenin. Of 118 firms surveyed by Clendenin, 14 were rail- roads, but Lemke provided only limited footnote reference ro railroad refunding},2 Past and present Interstate Com- merce Commission requirements are mentioned in the study, lClendenin, Op. cit., 37. 2Lemke, Op. cit., p. 147. 22 but statistics and effects are not given except by reference to the Clendenin report. This dissertation is intended to be a study in the history and theory of accounting procedures rather than of financial management related to bond refunding. Relevance Of some mention of capital budgeting lies in the possibility that accounting write-off may be related to the period of expected benefits employed in making the refunding decision. It is not a purpose of this study to rigorously analyze the theoretical models used as to the validity of claims or implications of logic methodology except as to consistency of specific conclusions in the general frame- work Of the model. Also, the prOpriety of logic as a tool of accounting theory will not be discussed. This has already been extensively treated in the Dominiak study and in other writings. Note on Terminology There seems to this writer to be little doubt that the lack Of uniformity in accounting terminology is part of a problem of effective communication between the accountant 1Gerald F. Dominiak, "An Investigation of the Applicability of Methods of Logic to the Analysis of Accounting Problems," (unpublished Ph. D. dissertation, Michigan State University, 1966). 23 and the lay user of financial statements. And bond account- ing is one of the areas in which terminology is not stand- ardized. Some writers use tht term "accumulation" to refer to the periodic build-up of the effective liability by the amount of the discount write-off and "amortization" to mean gradual reduction of the liability balance by the premium adjustment. Such usage is predicated on the assumption that discount and premium are deducted from or added to the principal on the liability side of the balance sheet. Examination of a number of actual financial state- ments indicates that "amortization" is often used to refer to the write-Off process, whether in reference to a debit or a credit balance. The balance is often a debit consist- ing of one or more elements of discount, issue cost, call premium, duplicate interest, and incidental redemption costs. Such a balance is often labeled, especially in pub- lic utility reports, as ”unamortized debt discount and ex- pense." Writers on the subject often ignore some of the specific components of the balance and designate the pack- age as "unamortized discount and call premium." Sometimes a credit representing premium received upon issuance of bonds is netted out against the other items. Where the netting out process results in a credit 24 balance or the premium is shown separately, the balance is often labeled "unamortized premium" or "net unamortized premium." In this dissertation, "accumulation" is accepted as appropriate when used to refer specifically to the discount adjustment, but "amortization" appears more descriptive of gradual write-off of the other items. For convenience, the more general term "write-off" will often be used. The amount to be written off will be designated as "unaccumulated discount, issue cost, and call premium" ex- cept where a collective term such as "financing costs" or "refunding-related items" seems adequate and more convenient or where individual components are discussed in isolation. Organization of the Dissertation Chapter I contains an overall preview of the study The historical background and present state of the problem are reviewed. Objectives, hypotheses, and research methods are stated and discussed. Chapter II provides a review of the history of accounting thought related to bond refunding, with special attention to developments near the time of approval of alternative procedures by accounting and regulatory organi- zations. A tabulation of presently sanctioned procedures 25 is included. Chapter III shows comparative tabulations of prac- tices for the periods 1936-1945 and 1956-1965. Financing costs are analyzed percentagewise and by application of certain statistical tests in comparison to dividends in relation to retained earnings and net income. Chapter IV includes a summary of the character- istics and rationale of the full—accrual approach based on the Hendriksen model. Application of the procedure is illustrated. Guidelines from two other models are summa- rized and discussed as to procedures supported for long- term debt accounting, with emphasis upon implications for the refunding problem. Chapter V provides a discussion of accounting for long-term debt in the light of the evaluative standards and communication guidelines prOposed in the American Accounting Association's A Statement of Basic Accounting Theory. Pro- cedures are discussed under two sets of assumed circum- stances: those related to general—purpose financial reporting in the nonregulated environment and those of the public utility firm under rate regulation. Chapter VI concludes the study with a review of the highlights of previous chapters and principal conclu— sions of the dissertation. Certain fundamental constraints 26 are discussed as criteria for limiting the differences be- tween public utility and nonregulated industrial accounting for bonds refunded. CHAPTER II THE HISTORY OF ACCOUNTING THOUGHT RELATED TO BOND REFUNDING This chapter contains a review of theoretical and practical argument concerning the nature of bond financing costs and accounting procedures for refundings. The writ- ings are reviewed from the early 19005 to the late 19605, although much of the controversy was concentrated in the 19305 and 19405. The discussion includes pronouncements of regulatory commissions, tax authorities, and professional accounting organizations as well as the prOposals of indi- vidual writers. Emphasis is placed upon views prevailing near dates of significant attention to the refunding prob- lem by accounting and regulatory authorities. A summary of procedures currently approved by various authoritative bodies is presented to show the divergence that still exists. The Basic Nature of Bond Financing Costs The literature of economics, accounting, and finance is replete with concepts and theories of interest. Most of these revolve around the question of the function performed 27 28 by the interest rate. In some way, all convey the notion that the interest rate is a price of capital. Hicks says of the interest rate that "it is a price, like other prices, and must be determined with them as part of a mutually interdependent system."1 But Hicks points out that there is actually a whole system of interest rates in any economy having more than one type of lending.2 Corporate bond rates are only one category. And there is usually a prevailing rate for a given class of bonds according to risk, term to maturity, and other characteristics. A corporation usually issues bonds at a price re- lated to the prevailing yield or market rate of interest. Proceeds often differ from face value by the amount of a premium or discount. Hence, the relevant rate of interest for the measurement of expense (or investor's earnings) is correspondingly different from the nominal rate on face value. Legally, of course, the bonds must be redeemed at face value if held by the lender until maturity. And periodic cash outlays must be made according to the nominal 1J. R. Hicks, Value and Capital (London: Oxford University Press, 1946), p. 154. 2 . Ibid. 29 rate of interest. Redemption prior to maturity usually requires payment of a call premium in addition to face value. The accountant must find some way of recording and reporting on the several elements of bond transactions. While the bonds are outstanding the main concern is with appropriate reflection of the liability and the periodic interest charge. The presumed objective is fair presenta— tion of financial position and results of Operations. This requires accrual accounting and adherence to the matching principle. Few published guidelines for accrual accounting were available through the nineteenth century. Littleton indi— cated that only about ten per cent of about fifty book- keeping texts appearing during the period 1788-1899 . 1 . . attempted to discuss accruals. He Cited National Book- keeping, 1861, by Bryant and Stratten, as an apparent rarity in that it provided for mortgage interest to be shown as a liability in the balance sheet, offsetting a red-ink debit to an interest account.2 But the increased use of long-term borrowing that accompanied the growth of 1A. C. Littleton, Accounting Evolution to 1900 (New York: American Institute Publishing Company, 1933), p. 150. 2Ibid. 30 the large-scale corporate organization required greater attention to accounting procedures related to debt trans- actions. Nineteenth century courts favored charging capital asset accounts with interest paid during construction, apparently considering the borrowing to be tantamount to acquiring the prOperty for which the funds were used.1 Littleton indicated that professional accounting Opinion prevailing at the time seemed to be fairly well reflected by these early court decisions.2 Tax authorities and regu- latory commissions have continued to authorize charging interest to construction within prescribed limits. The practice is quite common in the public utility area, as revealed by examination of a number of annual reports. Much of the problem in accounting for premium, discount, and issue costs seems to involve differences in views as to the nature of these items. In 1906, George 0. reported almost universal treatment of bond discount as a lIbid.. pp. 218-219. 2 . Ibid., p. 221. 3Consumers Power Company, Annual Report 1966, p. 23, is typical of many. 31 capital charge.1 "Capital charge" apparently referred to a charge to a physical asset account since May reported the basis of the charge to be the view of discount as represent- ing a share in future enhancement in value of property ex- pected to accrue to the stockholders' benefit.2 .Around the time of May's writing, however, a policy of conservative capitalization only on the basis of costs incurred or results achieved appeared to be underway. That such a change was indeed evolving was evidenced by the Interstate Commerce Commission requirement in 1907 that the railroads end the practice of charging prOperty accounts with discount upon issuance of the bonds.3 Uniform systems of accounts prescribed by commissions in recent years do not seem to allow the charge to physical plant accounts except possibly as a prorata portion of the interest attributable to the construction period.4 1George 0. May, "The PrOper Treatment of Premium and Discount on Bonds," The Journal of Accountancy, II (July, 1906), 176. 21bid. 3Robert E. Healy, ”Treatment of Debt Discount and Premium Upon Refunding," The Journal of Accountangx. LXXIII (March, 1942), 201. 4U. S. Federal Power Commission, Uniform System of Accounts Prescribed for Public Utilities and Licensees, Class A and Class B, effective March 1, 1965 (Washington: U. S. Government Printing Office, 1965), pp. 9, 37. Other national systems appear subject to similar interpretation. 32 May considered charging a prOportionate part of the discount to eXpense over the life of the bonds to be a con- servative practice. But, also citing conservatism, he sug- gested it as a write—off against intangible assets or extraordinary expenses. And he considered crediting premium gradually to income over the life of the bonds to be an acceptable practice.1 May did Object to crediting the entire amount of premium to income in the year of receipt and to using it as an offset against regular expenses. He apparently visu- alized the investor as losing the premium if the bonds were held to maturity and the issuer as having to return it in the event of retirement prior to maturity. Hence, it should not be written off immediately after issuance of the bonds. But discount was apparently regarded as accruing to the benefit of the bondholder if the firm achieved suffi— cient success to redeem the bonds at a price equal to or greater than face value.2 Although recommending different treatment of premium and discount, May did concede that both could be considered capitalized changes in the 1May, "The PrOper Treatment of Premium and Discount On Bonds," 177. 21bid.. pp. 176-177. 33 interest rate. By the 19305, the view of premium and discount as components of the effective interest rate appeared to be widely supported in the literature. But, following the era of charging discount directly to prOperty accounts, the item continued to be carried on the asset side of the balance sheet in practice. Apparently the first published indict- ment of the practice to attract significant attention among accountants was a statement by the American Account- ing Association in 1936. The Executive Committee of the association held unaccumulated discount to be a prOper deduction from face value of the obligation.2 In defending the committee's position, Paton claimed that discount was a measure of the difference be- tween proceeds of the loan and the stated amount due at maturity and, hence, a part of the true interest accruing over time.3 He thought the treatment of discount as an 1Ibid., p. 177. 2American Accounting Association, Executive Com- mittee, "A Tentative Statement of Accounting Principles," Accounting and Reporting Standards for Corporate Financial Statements and Preceding Statements and Supplements (Re- printed from the Accounting_Review, JUne, 1936), p. 61. 3W. A. Paton, "Presentation of Bond Discount," Paton on Accounting (Reprinted from The Accounting Review, September, 1937), ed. H. F. Taggart (Ann Arbor: Bureau of Business Research, University of Michigan, 1964), p. 340. 34 asset was a leftover from the early view of the balance sheet as a final statement of debit and credit ledger balances and a reflection of the concept of face value as the true liability for accounting purposes.l Some present-day textbooks claim that showing dis- count among the assets represents the prOprietary view of creditorship equities as claims to assets rather than the entity concept of liabilities as sources of assets.2 Sometimes discount is still referred to as a type of prepaid interest.3 But Paton pointed out that discount remains unpaid until maturity.4 Actually, it remains un— paid until retirement whether at or before maturity. The typical theory text holds that the effective interest rate encompasses premium and discount and that these are prOper additions to or deductions from face value of bonds outstanding to show the effective liability. The effective rate (the investor's yield rate) reflects the difference between net proceeds received and total 1.15121... p. 335. 2N. M. Bedford, K. W. Perry, and A. R. Wyatt, Advanced Accounting (2nd ed.; New York: John Wiley and Sons, Inc., 1967), pp. 67-68. 3Ibid. 4Paton, "Presentation of Bond Discount," p. 340. 35 outlays required over the life of the bonds to pay interest at the nominal rate and redeem the debt at maturity.1 Application of the effective rate in the accrual process represents the interest method of discount and premium amortization recently approved as a theoretically sound procedure by the Accounting Principles Board.2 At least one writer regards the true effective rate to be the current market rate, the application of which in the accrual process results in stating the liability at the current mar- ket price of the bonds.3 This is the amount required to retire the bonds at the balance sheet date. According to Paton, the relevant proceeds on which to base the effective rate consist of the issue price paid in by the first bona fide bondholder.4 Apparently an intermediary who aids in marketing the issue is not re— garded as a bona fide bondholder, and issue costs such as 1G. A. Welsch, C. T. Zlatkovich, and J. A. White, Intermediate Accounting (Homewood: Richard D. Irwin, Inc., 1968), Chapter 20, includes the typical intermediate theory presentation. 2American Institute of Certified Public Accountants, Accounting Principles Board, "Omnibus Opinion-1967," Opinion No. 12 (New York, 1967), p. 194. 3Eldon S. Hendriksen, Accounting Theopy (Rev. ed.: Homewood: Richard D. Irwin, Inc., 1970), p. 452. 4 Paton, "Presentation of Bond Discount," p. 338. 36 commissions paid to him by the issuer are not included in interest as viewed by Paton. Unlike interest paid for the use of funds, issue costs are viewed as payments for ser— vices rendered in aiding the firm to raise capital. Likening bond issue costs to organization costs and architect's fees, Paton and Littleton clearly place them in the asset classification.1 In contrast to unpaid discount, issue costs are typically prepaid. Paton and Littleton recommend systematic write-Off of issue costs over the life of the bonds, but they do not include them in the adjustment of face value to the amount of the effective liability.2 These writers regard the effective liability of the issuing corporation to be reciprocal to the effective receivable (the carrying value of the investment) from the lender's side of the transaction.3 In the event of premature retirement of the bonds, call premium must be dealt with by the accountant. Unlike discount, issue cost, and premium received upon issuance, 1W. A. Paton and A. C. Littleton, An Introduction to Corporate Accounting Standards (American Accounting Association, 1940), p. 32. 2Ibid., p. 95. 3Ibid¢l pp. 39-40. 37 call premium is incurred pply if the bonds are retired prior to the regular maturity date. In practice there is usually no provision made for call premium in the periodic interest accrual. On an after-the-fact basis, however, the amount turns out to be part of the total outlays to the bondholder when premature retirement occurs. The entire package of financing costs to be dis— posed of in some manner when bonds are redeemed prior to maturity usually includes some combination of the following: 1. Unaccumulated discount 2. Unamortized premium received at issuance 3. Issue costs 4. Call premium 5. Incidental call costs 6. Duplicate interest (see below) The last item usually arises when there is a refunding and both the old and the new issue are outstanding simultaneously for a short time. Some firms then charge interest on the issue being retired to the "unamortized" balance. 38 PrOposed Alternative Accounting Procedures Related to Refunding As noted in Chapter I, the controversy surrounding the accounting treatment of refunding-related items devel- 0ped largely during the low-interest, high-volume refunding activity of the 19305. In 1942, Walter A. Staub indicated that bond refunding was not a live problem in accounting until about a decade before.l Apparently his statement was an accurate one. A system of accounts prescribed by the Federal Power Commission in 1922 provided for writing off to profit and loss the discount, expense (issue cost), and premium on reacquired securities but did not mention specifically re- acquisition by refunding.2 A book published in 1926 as a reference review of utility accounting practices briefly discussed refunding but did not indicate specific accounting procedures.3 1Walter A. Staub, "Treatment of Discounts and Premiums in the Refunding of Bonds," The Accountants Digest, VII (March, 1942), 266,(Reprinted from The Controller, January, 1942). 2U. S. Federal Power Commission, System of Accounts Prescribed for Licensees Under the Federal Water Power Act, (washington: U. S. Government Printing Office, 1922), pp. 12, 79-80. 3W. G. Bailey, and D. E. Knowles, Accountipg Procedures for Public Utilities (Chicago: A. W. Shaw Company, 1926), p. 424. 39 In the 19305, however, conflicting Opinions and practices developed as firms were faced with the problem of accounting for large balances of unamortized financing costs remaining at refunding. The sections which follow are concerned with the procedures that were the subject of a controversy which remains unsettled. Immediate Write-off In The_gear of Refunding When the Committee on Accounting Procedure approved immediate write-off of the entire balance of unamortized financing costs in 1939, the procedure was said to conform most (of the three considered) to hitherto accepted account- ing doctrine.l Such doctrine included the completed—trans- action view that losses or expenses should be accounted for as such no later than the time of completion of the series of transactions resulting in their incurrence. The achievement of balance-sheet conservatism was also cited as an advantage of the procedure.3 In those days, the "doctrine" of conservatism was generally taken 1American Institute of Certified Public Accountants, Committee on Accounting Procedure, "Unamortized Discount and Redemption Premium on Bonds Refunded," Accounting Research Bulletin No. 2 (New York, 1939) (As reprinted in indexed volume of numbers 1 through 37, 1949, p. 9). 2 Ibid., p. 14. 3Ibid. 40 to mean that accountants should anticipate no profits and provide for all losses. In slightly different language, it meant that balance sheet figures should be kept within the historical dollar-cost limits of expected economic benefits, the probable realization of which was considered objectively indicated. Freeman indicated about four years before the committee's action that accountants favored immediate write-off for fear of overstating retained earn- ings by the amount that would otherwise be deferred. Freeman claimed that the broad finance view in support of immediate write-off was that the current cost of money should not be distorted by superimposition of costs incurred in a previous period and related to closed transactions. The Committee on Accounting Procedure favored making the write-off to income instead of retained earnings. The increasing importance of the income statement over the balance sheet was cited in support of this attitude, and charging retained earnings was said to understate charges 1Herbert C. Freeman, "The Treatment of Unamortized Discount and Premium on Retirement in Refunding Operations," The JOurnal of Accountangy. LX (October, 1935), 256. 21bid. 41 to income for borrowing costs.1 In addition to the arguments of accounting theory, the committee cited approval of immediate write—off by regulatory authorities.2 Immediate write-off to retained earnings (then called surplus) was the general rule pre- scribed in regulatory systems of accounts. But the systems often contained the escape clause that gradual write-off could be followed with special permission of the commission? Among the national commissions, the Interstate Commerce Commission was exceptional in not providing such a clause but allowing premature write-off to retained earnings up to 1943.4 In published regulations and rulings, statements concerning accounting theory observed by the commissions are Often skimpy at best. In favoring immediate write-off, the commissions apparently accepted the accountants' usual 1 . . A. I. C. P. A., Committee on Accounting Procedure, Accounting Research Bulletin No. 2 (Reprint 1—37, pp. 9- 10). 2Ibid., p. 9. 3National Association of Railroad and Utility Com- missioners, Uniform System of Accounts for Electric Utilities (New York: The State Law Reporting Company, 1937), p. 14. 4 Bernhard C. Lemke, "The Treatment of Bond Discount and Premium in Connection with Refundings," (unpublished Ph.D. thesis, University of Minnesota, 1946), p. 170. 42 arguments of conservatism and the completed-contract con- cept. But departures from the general rule on the bases of "the public interest," "administrative expediency," and the like were not unusual. An example was the Pennsylvania Water and Power Company case in 1940. In this case the Federal Power Com— mission declared immediate write-off to represent sound accounting principles and claimed that financing costs related to bonds refunded were subject to the same kind of accounting as losses from fires and floods.1 But the com- mission argued that the interests of investors and rate- payers (customers) were sometimes best served by gradual write—off where the amounts involved were large enough to create a deficit in retained earnings and possibly preclude advantageous financing. The alternatives considered in the Pennsylvania case were within the bounds of guidelines approved by the Committee on Accounting Procedure. But May criticized the commission for unnecessarily subordinating accounting principles to administrative expediency.3 1George 0. May, "Accounting Principles and Regu— latory Expediency," The Journal of Accountancy, LXXI (February, 1941), 116-118. 21bid. 31bid. 43 The Federal Power Commission represents the class of regulatory authorities whose function is to insure ade- quate service and protection from price exploitation for utility consumers. The Securities and Exchange Commission performs a different function--policing the securities mar- kets. The S. E. C. was mainly interested in full disclo- sure of material information and did not specify a partic- ular write-off procedure except in connection with bond retirements from the proceeds of capital stock sales.1 The requirement of immediate write-off in such cases may be con- sidered a part of the general body of support for the pro- cedure, although this type of transaction falls outside the usual concept of refunding discussed in this study. Cranstown Opposed the S. E. C.'s position on the grounds that so long as cash was used to retire bonds, the accounting should not be dictated by whether the funds were obtained through the issuance of stocks or new bonds. Tax requirements constitute yet another class of regulation. In general, the federal income tax approach has been that financing costs related to retired bonds are 1William D. Cranstown, "S. E. C. Release No. 10," The Journal of Accountanpy_(Commentator Department), LXVII (March, 1939), 179. 21bid. 44 components of gain or loss in the year of retirement for cash, regardless of the means of acquiring the cash. This was the position taken in the Great Western Power Company case in 1936 and in previous cases cited as precedents. At the time the Great Western case came before the Supreme Court, the tax regulations, essentially as promul- gated under the Revenue Act of 1918, referred directly to discount and premium but were silent as to issue cost. The disposition of these costs was a principal issue before the court. The ruling was that issue costs were deductible under the same theory as discount but that the items were subject to write—off over the life of the new bonds in a case of direct exchange of issues rather than a cash refunding. Besides theoretical argument in support of other procedures, objections to immediate write-off have generally followed a pragmatic line similar to that taken in the Pennsylvania Water and Power Company ruling. For example, Accounting Research Bulletin No. 2 contains the admonition that depletion of retained earnings by immediate write-off 1Great Western Power Company v. Commissioner of Internal Revenue, 297 U. S. 543 (1936). 21bid. 3Ibid. 45 might sometimes render temporarily illegal the payment of preferred dividends. Freeman argued that the notion of retirement of the old bonds as terminating a disadvantageous contract mis- placed the emphasis.2 He prOposed emphasis on the new con- tract but held that part of the cost of the new benefits consisted of items originally determined in the old one. Having expressed some misgivings about making the complete write-Off to retained earnings in its original pronouncement, the Committee on Accounting Procedure in 1942 declared showing the full charge in the income state— ment to be sound accounting in accordance with tax pur- poses.4 Taken at face value, the point may imply that a procedure is sound because it is acceptable for tax pur- poses. But, rather than insisting on the full-charge 1A. I. C. P. A., Committee on Accounting Procedure, Accounting Research Bulletin No. 2 (Reprint 1-37, p. 16). 2Herbert C. Freeman, "Unamortized Discount and Premium on Bonds Refunded," The Journal of Accountancy (Correspondence), LXVIII (December, 1939), 399. 31bid. 4American Institute of Certified Public Accountants, Committee on Accounting Procedure, "Unamortized Discount and Redemption Premium on Bonds Refunded (Supplement)," Accountipg Research Bulletin NO. 18 (New York, 1942) (As reprinted in indexed volume of numbers 1 through 37, 1949, p. 151). 46 treatment in the income statement, the committee recommended deduction of an amount at least equal to the tax reduction, regardless of the procedure followed for the remainder.1 The argument offered was that charging the loss to retained earnings and reducing the tax charge to income would give the misleading appearance of increasing net income for the year.2 The committee suggested that the offsetting credit should be to the unamortized financing costs or to a re- serve for future taxes.3 This position was later modified to specify that any write—off to retained earnings or over the life of the old issue should be limited to the remainder after allowing for the tax reduction.4 Support for the immediate write-off procedure has continued to appear in the literature since the Institute's original pronouncement. Writing in 1940, Paton and 5 Littleton stated the following: lIbid., p. 152. 2Ibid. 3Ibid. 4American Institute of Certified Public Accountants, Committee on Accounting Procedure, "Restatement and Revi- sion of Accounting Research Bulletins," Accounting Research Bulletin No. 43 (New York, 1953), pp. 130-131, 142. 5 Paton and Littleton, Op. cit., p. 95. 47 ...there is scant justification for carrying for- ward costs attaching to a closed contract: to do so would inflate the cost of securing funds just as the carrying forward of a part of the cost of retired plant would overstate the amount invested in the existing layout of facilities. Somewhat similar reasoning has been stated more recently by Bierman. He considers discount a liability valuation account and suggests that failure to write it off upon retirement is similar to leaving the allowance for depreciation in the accounts after disposal of a fixed asset. Lemke concluded immediate write-off to be appro— priate on the basis of consistency with approved practice for nonrefunded retirements and the view of the old and new bond issues as separate transactions.2 Other very recent advocates of immediate write—off have included Moffet, Sprouse—Moonitz, and Hendriksen. In his 1964 Canadian study, Moffet expressed a preference for immediate write—off on the basis of consist— ency with recognition in practice of profits and losses in accordance with realization.3 lHarold Bierman, Jr., Financial Accounting Theory (New York: The MacMillan Company, 1965), p. 89. 2Lemke, Op. cit., p. 158. 3H. S. Moffet, Accounting for Costs of Financing (Toronto: The Canadian Institute of Chartered Accountants, 1964), p. 19. 48 In the Henriksen and Sprouse-Moonitz studies, the balance of unaccumulated discount and call premium remain— ing at refunding is held to be a realized loss resulting from interest rate and bond price changes.l'2 Hendriksen differs from the other writers, however, in advocating use of the current market rate of interest in the accrual pro- cess and recognition of holding gains and losses while the bonds are outstanding. The Sprouse-Moonitz and Hendriksen cases are discussed extensively in Chapter IV of this study. Write-Off to Normal Maturipy of Retired Bonds Although approving immediate write-off in the 1939 bulletin, the Committee on Accounting Procedure stated perhaps a stronger case for gradual write-off over the remaining life to normal maturity of the refunded issue. This procedure was deemed clearly permissible. But the committee refrained from expressing a preference of suffi— cient definitiveness to require qualification in auditors' reports when some other approach was used.3 1Robert T. Sprouse and Maurice Moonitz, "A Tentative Set of Broad Accounting Principles for Business Enter- prises," AccountingyResearch Study_Np, 3 (New York: Ameri- can Institute of Certified Public Accountants, 1962), p. 41. 2Hendriksen, Op. cit., pp. 453-454. 3A. I. C. P. A., Committee on Accounting Procedure, Accounting Research Bulletin No. 2 (Reprint 1-37, p. 10). 49 Reasoning given in support of this second alterna— tive included the following:1 1. The merit of reflecting expense apprOpriately in a series of income accounts. 2. The claim to close conformity with the (then) current trend in accounting Opinion. The committee cited a tendency to bring an increased number of transactions within the rule of charging income with costs over periods of expected future benefits rather than when payment is made or ascertained.2 Apparently with the matching concept in mind, the committee attributed merit to spreading discount and premium over the remaining normal life of the old issue when savings exceeded the amount to be written off.3 Also, the potential illegality of preferred dividend payments, cited as a dis- advantage of immediate write-off, was avoidable by use of gradual write-off.4 Walter A. Staub, a member of the Committee on Accounting Procedure, indicated that he had concurred with other members in approving write-off over the normal life 1Ibid. 2Ibid., pp. 14-15. 3Ibid., p. 16. 41bid. 50 of the old issue on the basis of the argument that this was the period of direct benefits from a lower interest rate. This view, simply stated, seems to have been the essence of much of the support for this procedure in the 19305 and 19405 when the almost obvious purpose of most refundings was achievement of interest savings. Application of the matching concept to the "objectively determinable" period of savings was considered to provide sufficient theoretical support. It was easy to see that the higher rate on the old issue would have continued to be paid over that period in the absence of retirement. In revising previous bulletins in 1953, the Com— mittee on Accounting Procedure stated a definite preference for write-off over the life of the old bond issue.2 The evidence Offered seemed to consist mainly of a summary of the previous argument. One apparent difference was more precise emphasis on the view of discount as the cost of an Option enabling the firm to enter into a more advantageous arrangement from the point of refunding to the original maturity date of the first issue.3 1Staub, Op. cit., p. 268. 2A. I. C. P. A., Committee on Accounting Procedure, Accounting Research Bulletin No. 43, p. 131. 31bid. 51 The view expressed was similar to that of Cranstown around the time of the first pronouncement. He claimed that discount might be viewed as a demand charge to secure continuation of capital availability.1 The concept of a "demand charge" seems to imply that discount is a concession to induce the prospective lender to take bonds subject to premature call at the Option of the issuer. Very little outright advocacy of write-off over the original life of the refunded issue has appeared in theory literature in recent years. The typical theory text usually contains a limited discussion of the procedures acceptable to the Institute but seldom offers new argument pro or con. No accounting system published by a utility commis— sion and examined by this writer actually prescribed write- off over the period to normal maturity of the refunded issue. But the procedure has been allowed on a permissive basis. For example, the California commission was reported to have allowed all three of the alternatives dealt with by the Institute's committee, each in different cases.2 Objections to the procedure cited by the Committee on Accounting Procedure included lack of balance sheet l Cranstown, Op. cit., 180. 2Freeman, "Unamortized Discount and Premium on Bonds Refunded," 397. 52 conservatism and the view of the items involved as more properly representing costs of terminating a disadvantageous contract than of entering into a better one.1 Thus, objec- tions to the procedure seem to take the form of positive arguments in favor of other alternatives. Freeman expressed the Opinion that if the saving in interest over the remaining life of the old issue were the sole purpose of refunding, then the maturity of the new issue would be made to coincide with that of the old.2 He thought the company should, apparently as a general rule, be able to refinance on a more favorable basis over the shorter period.3 RObert E. Healy, senior member of the Securities and Exchange Commission at the time of his writing in 1942, said that write-off over the remaining life of the old issue was less supportable in logic than either of the other two considered by the Institute's committee.4 His reasoning was that the procedure had no relation to any facts existing 1A. I. C. P. A., Committee on Accounting Procedure, Accountipngesearch Bulletin No. 2 (Reprint 1-37, p. 19). 2Freeman, "Unamortized Discount and Premium on Bonds Refunded," 399. 3 Ibid. 4Healy, Op. cit., 206. 53 at the time of making the entries, including the literal nonexistence of a maturity date for an issue that has already been retired.1 Write-off to Maturity of Refunding Bonds In rejecting write—off over the life of the new issue in 1939, the Committee on Accounting Procedure stated very little in favor of the alternative but a great deal in opposition to it. Basically, the procedure was held to be in conflict with generally accepted accounting principles and lacking in theoretical support.2 Specific disadvan- tages mentioned were as follows:3 1. The use of the procedure might encourage disadvantageous refunding by exaggeration of annual savings. 2. No logical relationship existed between the unaccumulated discount on the old debt and the term of the new. 3. The use of the procedure was said to under- state the annual cost of money over the remaining portion of the original life of the retired issue. lIbid. 2A. I. C. P. A., Committee on Accounting Procedure, Accounting Research Bulletin No. 2 (Reprint 1-37, p. 10). 3Ibid.. pp. 18-19. 54 4. The procedure was unconservative with respect to both the balance sheet and the income statement. 5. No basis existed for assuming benefits beyond the normal maturity date of the retired issue because of the difficulty of foreseeing interest rates that far in advance. Concerning the last point, the committee indicated that the benefits would have to be measured by the differ- ence between the rate which a new loan would bear if made at the normal maturity date of the old issue rather than prior to that time.1 Cases of refinancing of refunding issues, apparently in response to unanticipated rate changes, were cited in substantiation of the stand taken on write-off over the life of the new issue.2 But the procedure had its prOponents among those writing on the subject. Freeman mentioned the view that refunding just involves retention of capital already raised.3 He said that this seemed to reasonably relate unaccumulated discount on the old bonds to the new issue. In addition, he thought that call premium on the Old could, in effect, be considered a discount on the new bonds 1Ibid., p. 18. 2Ibid. 3Freeman, "The Treatment of Unamortized Discount and Premium on Retirement in Refunding Operations," 258. 55 and a relevant cost of refinancing that could be avoided by not calling the original issue prior to maturity. He sug- gested that since the firm Obtains the benefits of the re- financing over the entire term to maturity of the new issue the shorter period should not be burdened with the cost of benefits received later.1 May, who was vice-chairman of the Committee on Accounting Procedure, replied to Freeman that a "modest" amount of premium and discount on the old bonds might be assigned to the new ones when refunding occurred near the maturity date of the first issue. He assumed that in such a situation management's purpose would be related to the future and not primarily concerned with interest savings over the short period to maturity of the old issue.2 But May did not support such a procedure in the general case. In his 1942 article, Healy stated reasoning similar to that of Freeman, holding that the costs of both issues should be considered relevant to the combined periods.3 But he thought that practice should follow procedures 11bid. 257-259. 2George 0. May, "Unamortized Discount and Premium on Bonds Refunded," The Journal of Accountangxo LXIX (January, 1940), 50. 3Healy, Op. cit., 206. 56 supported by court decisions for tax purposes under the conditions prevailing at the time.1 Apparently the conditions to which Healy referred were that many firms engaging in refunding had sufficiently large balances of financing costs to effect substantial tax reductions in the year of refunding. As previously noted, immediate write—off was the basic requirement, and interperiod tax allocation had not yet become p0pular. Gushee criticized Healy for failure to stand by his theoretical convictions and the Institute for failure to consider an accounting approach recognizing face value and the coupon rate as of only nominal character.2 He held that the legalistic aspects of the old and new issues as separate contracts to be of little consequence as to the propriety of accounting procedure.3 His main point appeared to be that cost of redeeming the old issue was a legitimate cost of the new financing, analogous to the cost of removing old debris as part of the cost of a new building.4 To reinforce his view, Gushee cited the case 11bid., 209. 2 Charles H. Gushee, "Bond Discount or Premium at Refunding," The Accounting Review, XXI (January, 1946), 61. 3Ibid., 63. 41bid.. 64. 57 of a firm refunding at an even higher rate of interest than borne by the retired issue, the action being taken for the specific purpose of extending the time of the loan.1 Similarly, when the Committee on Accounting Proce- dure reconsidered the matter in 1953, Perry Mason contended that write-off over the life of the new issue was justified in instances of refunding to extend the present rate of interest in anticipation of higher future rates.2 Mason was the only outright dissenter to reapproval of the stand essentially taken by the 1939 committee membership. In its 1965 Opinion, the Accounting Principles Board indicated that the circumstances of refunding because of lower current interest rates or anticipated higher future rates appeared to reasonably relate benefits to the life of the new issue.3 Thus, prOper matching would require costs to be written off over that period. But no mention was made of refunding to effect an extension of the loan for purposes other than interest savings. Bierman reasoned that unaccumulated discount on lIbid., 65. 2A. I. C. P. A., Committee on Accounting Procedure, Accountinngesearch Bulletin No. 43, p. 133. 3A. I. C. P. A., Accounting Principles Board, Opinion No. 6, p. 43. 58 bonds refunded at a higher rate could be considered a loss on retirement that could be avoided by not redeeming the issue prematurely.1 He somehow saw fit to claim that the discount is not being carried forward; instead, the amount of the carryover was labeled a "cost factor" associated with the loss, the future interest benefits, and the lengthening of the debt period.2 Thus, write-off over the life of the new issue was held to be reasonable. Until very recent years, the history of regulatory sanction of gradual write-off of unamortized financing costs related to refunded bonds seems to indicate about as much support for write-off over the life of the new issue as for any procedure except immediate write-off. Alterna- tives to immediate write-off were usually allowed through special rulings. No national body Specifically prescribed any gradual write-off procedure at the time of the Institute's attention to the matter in 1939. At the state level in the mid-thirties, two appli- cations of the concept of conservatism seemed to prevail. Immediate write-off was based on balance sheet conservatism, while gradual write-off emphasized conservatism with respect 1Bierman, Op. cit., pp. 89-90. 21bid., p. 90. 59 to future earnings measurement and reporting. Wisconsin, in 1934, required write—off over the life of the new issue up to the annual limit of the effec- tive rate on the old issue but stipulated that the excess over that level be written off directly as a loss.2 The apparent intent of such a requirement was to insure against consumer rate disturbances in years following the refunding. The Wisconsin requirement was included in that state's prescribed system of accounts for electric utili- ties. But Michigan, having no specific provision for such items in its system at the time, handed down an order around 1935 that write-off was to be over the life of the new issue. In considering a direct exchange of new for old bonds in the Great Western tax case, the Supreme Court seemed to view the new issue as a substitute contract merely establishing a different time period to which costs originally assignable to the life of the old issue should be applied.4 Accountants accepting the same view with 1Freeman, "The Treatment of Unamortized Discount and Premium on Retirement in Refunding Operations," 256. 21bid., 255. 31bid., 256. 4Great Western v Commissioner, Op. cit. 60 respect to retirements from cash proceeds of new bonds appear to regard the difference between the two types of transactions as a matter of form rather than substance. Objections to write-off over the life of the new issue as viewed by the Committee on Accounting Procedure have already been mentioned. Its list seems to be fairly representative of the Opposition to the procedure. Other Proposed Procedures The Committee on Accounting Procedure published only passing comment on alternatives other than the three apparently believed to be most commonly advocated and practiced. But certain others have been commented upon to some degree in this chapter. Additional ones are dis- cussed below. Most of the national agencies have allowed accelera- tion of the write-off of financing costs while the bonds were outstanding.1 The relevance of the acceleration pro- cedure, sometimes referred to as premature write-off, for refunding is that it is one way that discount and issue cost may be eliminated prior to retirement. The amounts involved may be disposed of immediately upon issuance of 1N. A. R. U. C., Uniform System of Accounts for .Qlectric Utilities (1937), p. 14. Systems of member com- missions show a similar provision. 61 bonds or at such other time of management's choosing. The balance prematurely charged off, however, has been required to be of "insignificant" amount in most cases. A variation of accelerated write-off subsequent to refunding but prior to the normal maturity date of the retired issue was deemed an acceptable middle course by the 1939 Committee on Accounting Procedure.1 Conservatism and convenience to management seem to be the only points men- tioned in the literature in support of this procedure. Klug advocated a hybrid procedure involving write- off of discount and issue cost over the remaining life of the old issue but call premium over the life of the new. His reasoning was that call premium should be related to the benefits of a later maturity date.2 In 1958 national regulatory authorities made the first major change in accounting procedures adOpted for general use in connection with refundings since World War II. The National Association of Railroad and Utility Com- missioners approved revisions of its model system to permit 1A. I. C. P. A., Committee on Accounting Procedure, Accounting Research Bulletin No. 2 (Reprint 1—37, p. 21). 2V. Childs Klug, "The Accounting Treatment of Unamortized Discount and Premium in Utility Refunding Operations," The JOurnal of Land and Public Utility Economics, XII (November, 1936), 411. 62 write-off of items applicable to a refunded issue over the period required for interest savings associated with the refunding to just equal the amount to be written off.1 This version of payback time, currently in effect, may be adapted by a utility without explicit prior permission pro- vided that three important conditions are met. These are 2 as follows: 1. Any balance of items applicable to "'grandfather'" issues previously re- funded by the bonds being currently refunded must be charged off entirely to retained earnings. 2. An amount equal to the tax saving related to the refunding must be charged against income in the year of refunding. 3. Accumulation or amortization must be by equal monthly charges of the remaining balance, after giving effect to the above and any additional income taxes attribut— able to the net annual saving in interest and related charges, over the period so determined. The procedure was adOpted officially by the Federal Power Commission to take effect in 1961.3 Many states also 1National Association of Railroad and Utility Com- missioners, uniform System of Accounts for Class A and Class B Utilities (Washington: N. A. R. U. C., 1959), pp. 59-60. 2Ibid., p. 60. 3U. S. Federal Power Commission, Uniform System of Accounts Prescribed for Public Utilities and Licensees, effective January 1, 1961 (Washington: U. 8. Government Printing Office, 1960), p. 32. 63 follow the N.A.R.U.C. system or allow similar procedures, at least in the regulation of electric, gas, and water utilities. This is indicated by Exhibit I (p. 65). As a practical matter, use of the payback procedure results in nullification of the effects of the refunding until the write-off has been completed, provided that tax allocation based on the refunding deductions is confined to the net-of—tax approach described in the above instructions. The prescribed treatment of the tax-saving equiva- lent in the N.A.R.U.C. system is in accordance with the previously cited A.I.C.P.A. recommendations in Accounting Research Bulletin No. 18 and No. 43 for write-off procedures other than immediate charge of the entire balance to income. Present State of Authoritative Sanction In Spite of much agitation in recent years to find the "one best way" in accounting or at least to narrow the areas of differences, of which bond refunding is only one, wide latitudes of choice remain. The philoSOphy expressed by George 0. May in 1940 still prevails. May likened accounting procedure to the law in the sense that it is determined by a combination of many forces—-including his- tory, custom, tradition, and concepts of public interest. 1May, "Unamortized Discount and Premium on Bonds Refunded," 51. 64 He suggested that it is seldom possible to designate a single procedure as the only prOper one or, conversely,as entirely unacceptable.l In Exhibit I on the following page is shown a sum— mary of procedures currently required or expressly approved by national accounting and regulatory authorities and that were in effect in various states in 1968. The procedures listed are officially approved for use by firms under the regulatory jurisdiction indicated without obtaining explicit prior permission. Interestingly, no national uniform system of accounts examined specifically prescribes write-off over the life of the old issue or of the new. But the Federal Power Commission announced in August, 1971, that it prOposes to adOpt both of these procedures in addition to immediate write—off.2 The commission also prOposed adOption of com- prehensive tax allocation in accordance with Accounting . . . . 3 . . . PrinCiples Board Opinion No. 11. ThlS lS apparently in- tended to replace the net-of-tax approach currently lIbid. 2U. S. Federal Power Commission, "Notice of Pro— posed Rule-making," Docket No. R-424 (August 6, 1971), pp. 13-14. 3Ibid., p. 5. 65 EXHIBIT I ACCOUNTING PROCEDURES FOR ITEMS RELATED TO BONDS REFUNDED AS APPROVED OR REQUIRED BY REGULATORY AUTHORITIES AND PROFESSIONAL ACCOUNTING ORGANIZATIONS 1968 Procedure Sanctioning Organization 1. Immediate write-off to income or retained earnings in the year of refundinga Write-off over remaining life to normal maturity of the refunded issue Write-off over the life of the refunding issue Write-off over period of equalization of savings and amount to be written off (payback time) Accelerated write—off No formal regulation American Institute of Certified Pub- lic Accountants; American Accounting Association; National Association of Regulatory Utility Commissioners;b Federal Power Commission; Interstate Commerce Commission; Internal Revenue Service; Federal Communications Com- mission; forty state commissions and the District of Columbia. American Institute of Certified Public Accountants. American Institute of Certified Pub- lic Accountants; Internal Revenue Service;c one state commission (Maine). National Association of Regulatory Utility Commissioners; Federal Power Commission; thirty-one state commissions. American Institute of Certified Pub- lic Accountants;d National Associa— tion of Regulatory Utility Commis— sioners; Federal Power Coumission; Federal Communications Commission; Interstate Commerce Commission; thirty-two state commissions. One state commission (Alaska). aA.I.C.P.A., F.P.C., and I. C. C. guidelines currently indi— cate a charge to income; F.C.C. guidelines indicate a charge to retained earnings for telephone companies but to income for certain other media. bFormerly National Association of Railroad and Utility Commissioners. cFor direct exchange of issues only. (Continued on page 66) 66 prescribed for refunding accounting. Stated reasons for the prOposed changes, which had not been officially acted upon at the date of this writing, include bringing utility accounting for long-term debt more into line with generally accepted accounting principles. The Federal Communications Commission indicated that write-off over the life of the old issue may be per— mitted on the grounds that "true" interest cost for rate dOver period shorter than remaining life of old bonds. Sources: Manuals and other publications previously cited plus additional sources as follows: 1) American Institute of Certified Public Accountants, Accounting Principles Board, Opinion No. 9 (December, 1966). 2) State of Michigan, 1960 Annual Administrative Code Supplement. 3) U. S. Federal Communications Commission, Rules and Regulations, VIII (Washington: U. 8. Government Printing Office, 1965). 4) U. S. Federal Communications Commission, Rules and Regulations, IX (Washington: U. S. Government Printing Office, 1968). 5) U. S. Federal Power Commission, Uniform System of Accounts Prescribed for Public Utilities and Licensees (Washington: U. 8. Government Printing Office, 1970). 6) U. 8. Interstate Commerce Commission, Uniform System of Accounts for (lass I and Class II Common and Contract Metor Carriers of Property (Washington: U. 8. Government Printing Office, 1969). 7) Mail survey of state commissions, 1968. Responses were received from forty-nine states and the District of Columbia. Iowa did not respond to initial and follow-up requests. 1Ibid.. p. 3. 67 purposes should include a prorata portion of discount and call premium related to a refunded issue over the period to its normal maturity date.1 At the state level, Maine indicated approval of write-off over the life of the new issue without qualifica- tion. But California and Maryland mentioned approval of this procedure under Special circumstances. California indicated occasional approval for a firm's accounting pur- poses but not for rate proceedings. Maryland said that the procedure was allowed when the refunding was to modify underlying mortgage requirements and not considered a normal refunding. A "normal" refunding was inferred to be for the usual purpose of securing interest savings. The payback write-off has not been specifically discussed in Institute pronouncements on bond refunding. But acceptance of accelerated write—off over a period shorter than the normal life to maturity of the retired issue apparently would permit use of the payback procedure. Accelerated write-off is indicated to be an acceptable pro- cedure in a provision of Accounting Research Bulletin NO. 43 which has not been changed by Board Opinion.2 1Letter from Kelley E. Griffith, Chief, Domestic Rates Division, Federal Communications Commission, November 17, 1969. 2A. I. C. P. A., Committee on Accounting Procedure, Accountinngesearch Bulletin No. 43, p. 132. 68 As indicated in Footnote "a", some differences in requirements concerning the immediate write-off charge still exist. In Accountinngesearch Bulletin No. 43 the position taken was that the charge should be to income rather than to retained earnings unless a misleading dis- tortion of net income would be the result.1 Support for the all-inclusive income statement contained in Accounting Principles Board Opinion No. 9, issued in 1966, appears to require the charge to income.2 The inferred view is that the gain or loss arising in connection with a refunding should be considered a result of a transaction of the cur- rent period rather than a prior period adjustment. Sources cited for Exhibit I indicate that the charge to income has been prescribed in systems of accounts issued by the Federal Power Commission and the Federal Communications Commission since publication of Opinion No. 9. Including accelerated write-off, four alternative procedures of accounting for financing costs related to bonds refunded are acceptable to the American Institute of Certified Public Accountants at the present time. But only two of these are significantly favored in published lIbid., p. 142. 2A. I. C. P. A., Accounting Principles Board, Opinion No. 9, pp. 112—116. 69 regulations of utility commissions currently in effect. Whether the changes prOposed by the Federal Power Commis- sion are indicative of a future trend toward closer agree- ment between the accounting profession and regulatory authorities remains to be determined. At any rate, the historical evidence appears to clearly support a conclusion that bond refunding is an area in which differences in accounting procedures have not been narrowed.' Summary The interest rate is viewed by economists as a price of capital. In accounting it is regarded as the price of borrowed capital. Varying interest rates result in the issuance of bonds at prices differing from face value. Thus, premium and discount enter into the deter- mination of the effective rate of interest or yield rate. Physical prOperty accounts were often charged with bond discount during the early 19005. But by the 19305 bond premium and discount were well accepted as components of the effective interest rate and, hence, of the effective bond liability. Before 1958 most of the developments concerning accounting for bond refundings revolved around three principal procedures of writing off financing costs related 70 to the refunded issue. These were: 1. Write-off immediately at refunding, either against income or directly to retained earnings. 2. write-off over the remaining period to normal maturity of the refunded issue. 3. Write-off over the life of the refunding issue. Theoretical argument in support of the first proce- dure has been based largely on the concept of conservatism and the view of the refunded issue as a completed trans- action. Advocates of immediate write-off hold the differ- ence between redemption cost and the carrying value of the liability to be a gain or loss realized upon retirement of the bonds. Consistent with this view, current A. I. C. P. A. guidelines and those of some utility commissions require that the immediate write-off charge be made to income rather than to retained earnings. Advocates of the second procedure view the period to normal maturity of the refunded issue as the objec— tively determinable period of interest savings. Hence, prOper matching of costs and benefits is said to result from write-off over the life of the old issue. Some prOponents of write-Off over the life of the new issue regard financing costs applicable to the old issue as capital—retention costs prOperly assignable to 71 the old issue but recommend write-off of call premium over the life of the new issue on the basis that it is related to extending the time of the loan. Refunding in anticipation of higher future interest rates is held by some authorities to justify write-off over the life of the new issue. Uhder these circumstances, the expected benefits are assumed to result from refunding at a lower rate than would be incurred at maturity of the old issue. The American Institute of Certified Public Account- ants approved the first two of the above procedures in 1939 and the third in 1965. Historically, write-off in the year of refunding has been in accordance with tax requirements and the basic rule of utility regulatory commissions. But departures from the general rule have often been allowed by the com- missions. Since 1958, write-off over the payback period has been adOpted by several commissions. The two gradual write-off procedures emphasized and approved by the American Institute of Certified Public Accountants are not specifically stated as acceptable in regulatory uniform systems of accounts currently in effect. But the Federal Power Commission has prOposed future adOption of the pro— cedures approved by the Institute. 72 In most theoretical discussion of accounting for bond liabilities, the investor's historical yield rate is advocated for use in the accrual process. But Hendriksen proposes a "full-accrual" approach based on the current market rate. Complete application of this procedure to the date of call leaves no remaining balance of premium or discount to be written off at refunding. CHAPTER III TRENDS AND EFFECTS OF ACCOUNTING PRACTICES In general, the purpose of this chapter is to show trends in accounting treatment of unamortized financing costs related to bonds refunded and to compare size- relationships of amounts involved as to possible effects on retained earnings, net income, and dividends. Specifically, the chapter contains the following: 1. A brief review of practices related to bonds outstanding. Comparative data concerning write-off prac— tices of public utilities and industrials related to refunding during the periods 1936-1945 and 1956-1965. An analysis of size-relationships of financing costs applicable to bonds refunded as per cents of REBCO and NIBCO for firms writing these costs off immediately and for those deferring write—off.1 An analysis of dividend size relative to REBCO and NIBCO for firms writing refunding-related items off immediately and for those deferring write-off. 1REBCO: Retained Earnings Before Charge-Off (of both dividends and refunding-related items). NIBCO: Net Income Before Charge-Off (of refunding- related items deducted in the income statement). 73 74 5. Certain statistical tests of representative- ness of data and of relationships between dividends and charges related to refunded bonds. The results provide some evidence of earnings and dividend effects as a basis for deferred write—off, espe- cially during the earlier period, but show increased use of immediate write-Off by public utilities refunding during the later period. General Practices Related To Bonds Outstanding A review of practices related to bonds oustanding is considered necessary for complete coverage of the sub- ject of accounting for refunding. This is because pre- refunding disposition of discount, issue cost, and premium received upon issuance may have a bearing upon the proce- dures used when refunding occurs. The era of charging discount to physical plant accounts, mentioned in the previous chapter, was apparently followed by a period of divided practices of either charging the amounts directly against retained earnings upon issu- ance of the bonds or deferring them in the balance sheet to 1 be gradually written off over the life of the bonds. 1American Institute of Certified Public Accountants, Committee on Accounting Procedure, "Unamortized Discount and Redemption Premium on Bonds Refunded," Accounting Research Bulletin No. 2 (New York, 1939) (As reprinted in indexed volume of numbers 1 through 37, 1949, p. 12). 75 Gradual write-Off was practiced by a majority of firms in- cluded in a survey by Clendenin in 1941, approximately two years after publication of Accounting Research Bulletin Np;_2 on the refunding problem.1 Most published references to balance sheet treat— ment of deferred debit balances related to bonds report the carrying of the amounts on the asset side. Among those reporting this as almost universally practiced, especially with respect to bond discount, were Paton in 19372 and Lemke in 1946.3 Financial statement data gathered for the period 1956-1965 indicate that little change has occurred in accounting for financing costs applicable to outstanding bonds since the 1936-1945 period. Data con- cerning the practices of 66 firms during the more recent period are summarized in Table 1 below. 1J. C. Clendenin, "How 118 Major Corporations Account for Bond Discount," The Journal of Accountancy, LXXI[(Ju1y, 1941), 38. 2W. A. Paton, "Presentation of Bond Discount," Paton on Accounting (Reprinted from The Accounting Review, September, 1937), ed. H. F. Taggert (Ann Arbor: Bureau of Business Research, University of Michigan, 1964), p. 335. 3Bernhard C. Lemke, "The Treatment of Bond Discount and Premium in Connection with Refunding," (unpublished Ph. D. Thesis, University of Minnesota, 1946), p. 43. 76 TABLE 1 PRACTICES OF ACCOUNTING FOR FINANCING COSTS RELATED TO BONDS OUTSTANDING 1956-1965 Public Write-off Procedure Utilities Industrials Number Per Cent Number Per Cent Straight-line accumulation or amortization 19 54 8 26 Straight-line adjusted for sinking fund retirements 7 20 —- -- Bond-month-dollars 2 l 3 Bond-year—dollars 2 —- -- Premature write-offa l 2 6 Deferred; basis undeterminedb 4 ll 20 65 Totals 35 100 31 100 aDiscount, premium, or issue cost written off immediately upon issuance of the bonds. bAll firms included in the "undetermined" category were inferred from remaining balances to have used some form of gradual write-off up to the time of refunding. Sources: Moody's Manuals, Questionnaires, and S. E. C. Reports. Practices Related to Refunding During the Period 1936-1945 The Clendenin Report The Clendenin study in 1941 was apparently the first documented report of practices embracing a broad range of firms. The study was based on a questionnaire survey of 118 firms. Accounting practices were reported 77 but not the amounts involved in refunding transactions.1 The findings of the study for public utilities and unregu- lated industrials are shown in Table 2 (p. 78). The cases tabulated represent those for these two groups responding as to treatment of unaccumulated discount and issue cost upon refunding. Fourteen railroads were included in the survey, but these are omitted from the present study for reasons discussed in Chapter I (see "limitations" pp. 20—22). The most notable results of the Clendenin report were the reverse pattern of preferences between public utilities and industrials for immediate write—off and gradual write-off and the emphasis upon write-off over the life of the retired issue among those deferring write-off. Clendenin summarized reasons mentioned by firms in support of the various procedures to some extent. Those favoring immediate write-off were mentioned primarily by industrials and railroads and included the following: 1. The retired issue was viewed as a completed transaction. 2. The procedure was considered conservative with respect to the balance sheet. 3. Immediate write-off conformed to income tax requirements. 1Clendenin, Op. cit., 37. 21bid., 41. 78 TABLE 2 ACCOUNTING TREATMENT OF UNAMORTIZED FINANCING COSTS APPLICABLE TO BONDS REFUNDED AS REPORTED BY CLENDENIN 1941 Public Write-off Procedure Utilities Industrials Number Per Cent Number Per Cent Immediate charge to- retained earnings 19 35 33 8O income -- —- 3 8 19 35 36 88 Deferred over- life of retired issue 23 42 4 10 life of new issue 4 7 l 2 payback period 28 4 -- -- life of old issue or life of new, whidhever is shorter 7 12 -- -- 36 65 5 12 Totals 55b 100 416 100 aTime required for cumulative interest savings to equal the total amount to be written off. bTreatment of call premium in a manner different than accorded discount and issue cost was indicated by 10 firms, of which 9 deferred write-off of call premium over the life of the new issue. cSix additional industrials, or a total of 47, replied as to treatment of call premium. Three treated this item in some manner different than that applied to discount and issue cost. Source: Clendenin, Op. cit., 40-43. 79 Concerning the last item, Clendenin reported that one-third of the industrials indicated that their accounting was influenced by law or regulation, apparently meaning tax requirements and possibly Securities and Exchange Commission preferences.1 Specific distribution of mentions of the other items was not given. The author noted that the firms mentioning the advantages of immediate write-off usually had retained earnings balances of sufficient size to absorb the charge without serious results. Reasons in support of gradual write-off,inferred to have been given largely by public utilities, included the following:3 1. The notion that stockholders' equity is, in effect, increased by a profitable refunding. 2. The view of funded debt financing as an on- going transaction until ultimate discharge with equity funds. 3. Authoritative support found in Accounting Research Bulletin No. 2 and regulatory commission rulings. Regulation was mentioned as an influence by one- half of the utilities.4 1Ibid. 21bid. 3;p;g.. 42. 4Ibid., 43. 80 Clendenin reported that some firms using gradual write-off made a charge in the income statement equal to the tax saving, as later recommended in Accountinngesearch . l . . . Bulletin No. 18. The remaining balance of unamortized financing costs related to the retired issue was then written off over subsequent periods. The Lemke Report The Lemke study in 1946 reported practices of 30 public utilities and 30 industrials refunding bonds at various times during the decade 1936-1945. Including multiple refundings by some firms, 31 utility and 34 indus- trial cases were reported upon.2 The findings of this study are summarized in Table 3 (p. 81). Similar to the Clendenin study, the Lemke report showed a majority of utilities deferring write-off and the Opposite for industrials. But the results differed some- what as to specific procedures within the two main write- off categories. The Lemke study reported refundings up to four years later than the Clendenin study and showed a larger proportion of industrials making the immediate 'write-off charge to income rather than to retained earnings. lIbid., 42. 2Lemke, Op. cit., pp. 180, 193. 81 TABLE 3 ACCOUNTING TREATMENT OF UNAMORTIZED FINANCING COSTS APPLICABLE TO BONDS REFUNDED AS REPORTED BY LEMKE FOR THE PERIOD 1936-1945 Public Write-off Procedure Utilities Industrials Number Per Cent Number Per Cent Immediate charge to- retained earnings 10 30 14d 41 income 1 3 11d 32 capital surplus -- -- l 3 11 35C 26 76 Deferred over- .life of retired issue 4 l3 2 6 life of new issue 6 19 --d -- payback period 4 13 —- —— other 5a 16 -- -- 19 61 2 6 Premature write-off lb 3 6 18 Totals 31 100C 34 100 aVarious periods shorter than the remaining normal life of the retired issue. bDiscount and issue cost prematurely written off; call premium deferred over life of new issue. cRounded dApparent hybrid procedures involved charging call premium to retained earnings in 18 cases (53%), to income in 12 cases (35%), and writing it off over the life of the new issue in one case (3%). Source: Lemke, op. cit., pp. 180, 193. 82 On the other hand, public utilities continued to charge retained earnings. Within the deferred write-off category, Lemke showed a much smaller percentage of utilities using write-off over the remaining life to normal maturity of the retired issue and somewhat greater percentages for the other procedures. Having compiled his data largely from prospectuses rather than by contact with the companies, Lemke did not obtain stated reasons for the use of particular procedures for most of the cases. Like Clendenin, however, Lemke called attention to the relatively small retained earnings balances of many of the utilities compared to those of the industrials.1 Both writers indicated that deficits would have existed for some of the deferring firms if unamortized financing costs had been written off immediately upon refunding.2'3 The Lemke study included greater attention to the tax effects of refunding than did the Clendenin report. Many of the cases reported by Lemke occurred during the war years when there was emphasis on the excess profits tax. Also, many occurred following publication of Accounting 1Ibid.. pp. 148-149. 2Ibid. 3Clendenin, Op. cit., 41. 83 Research Bulletin No. 18 in 1942, whereas Clendenin's study was conducted the year before. Lemke's findings concerning treatment of the tax saving are reported in Table 5 and related comments (pp. 87- 88) for comparison with 1956-1965 data. The relevance of the discussion of this matter lies in the effect upon the amount of the write-off. Reducing the amount of unamortized financing costs by the credit offsetting the charge in the income statement for the tax-saving equivalent can make a substantial difference in future charges to income where a large balance is involved. Practices Related to Refunding During the Period 1956-1965 The decade 1956-1965 was quite different than 1936- 1945 as to general conditions in the economic environment. Rather than depression followed by war, peacetime (or cold- war) prosperity prevailed for the most part. Interest rates were relatively higher than in the thirties and forties and generally rising. Downward fluctuations occurred sometimes but not often. Nevertheless, bond refunding continued to be practiced, especially by public utilities and to some extent by industrials. By tracing details of issues reported in lists of bonds redeemed in Moody's Public Utility Manual and 84 Industrial Manual, 69 utilities and 38 industrials were determined to have engaged in refunding during 1956-1965. From the investment manual reports and other sources tabu- lated in Table 7 (p. 93) of the statistical analysis sec— tion, adequate data for use in the study were obtained for 36 utility and 27 industrial refundings during the period. Regulated Public Utilities Practices of public utilities compared to indus- trials during the 1956-1965 period are presented in Table 4 (p. 85). In comparison to the findings reported by Lemke and Clendenin for the 1936-1945 period, the 1956- 1965 data indicate a definite increase in the practice of immediate write-off by the public utilities, 50 per cent of the sample against 35 per cent found by each of the two previous writers. Including the two hybrid cases in which one of the items was charged to retained earnings upon re— funding, cases of immediate write-off were actually in the majority in the sample for the more recent period. Within the deferred write-off category, substan- tially less use of write-Off over the life of the retired issue is shown by the 1956-1965 data than indicated by Clendenin in 1941. But a similar result was shown by Lemke's report covering 1936—1945. Greater use of payback 85 TABLE 4 ACCOUNTING TREATMENT OF UNAMORTIZED FINANCING COSTS APPLICABLE TO BONDS REFUNDED DURING THE PERIOD 1956-1965 Public Write-off Procedure Utilities Industrials Number Per Cent Number Per Cent Immediate charge to- retained earnings 18 50 -- -- income -- -- 20 71 18 50 20 71 Deferred over— life of retired issue 7 l9 1 3 life of new issue -- -- 2 7 payback period 9 25 -- -— period undetermined -- -- 3 11 16 44 6 21 Hybrid treatment 28 6 2b 7 Totals 36 100 28 100° aIssue costs prematurely written off and call premium charged to retained earnings in one case; issue costs charged to retained earnings and call premium deferred over the remaining life of the retired issue in one case. In the analysis of quantitative data hybrid cases were classified according to dispo— sition of the largest amount remaining at refunding. bIssue costs prematurely written off and no other items involved in one case of a direct exchange of issues. Issue costs prematurely written off and call premium charged to retained earnings in one case. cRounded Sources: Moody's Manuals, Securities and Exchange Commission reports, and questionnaires. 86 time, or the cumulative interest-savings period, for the write-off is shown by the 1956-1965 survey than by either of the two former studies. This is possibly related to recent official sanction of the procedure by utility com- missions as discussed in Chapter II. In View of the recent approval of write-off over the life of the new issue by the A. I. C. P. A. Accounting Principles Board, Hypothesis No. 2 held that a trend toward greater use of that procedure would be shown by the public utility statistics. Such was not the case, since no in- stance of write-off over the life of the new issue was found among the utility firms included in the 1956-1965 sample. Unregulated Industrial Corporations Comparing Table 4 with Table 2 and Table 3 shows that a clear majority of industrials practiced immediate write-off during 1956—1965 as well as 1936-1945. During the later period, the charge was made against income exclu- sively, whereas the earlier studies showed the write-off as being made to retained earnings in many cases. The almost complete switch since 1941 may possibly be attributable to the Institute's advocacy of the charge to income in Accountinngesearch Bulletins Nos. 2, $8, and 43.1 lSee Ch. II, pp. 41. 45, and 68, supra. 87 In contrast, utilities continued to charge retained earn— ings, although some commissions have recently prescribed the charge to income. Tax Effects The Lemke report included a tabulation of public utility cases in which the tax—saving equivalent was charged to income. A comparison of the findings of that study and those of the 1956—1965 survey is provided in Table 5 (p. 88). In the first three categories, the off— setting credit was made to the balance of unamortized financing cost in the year of refunding. Although the Lemke study did not give details for firms not charging the tax-saving equivalent to income, the closeness of the percentages for that procedure seems to just indicate continuation of approximately the general pattern reported in that study as becoming pOpular in the 2 . 19405. The impetus given to the practice by Accounting Research Bulletin No. 18 was perhaps later reinforced by the several additional Institute publications dealing with the subject of tax allocation.3 1Ibid., p. 65. 2Lemke, op. cit., pp. 152, 193. 3American Institute of Certified Public Accountants, Committee on Accounting Procedure, Accounting Research Bulletins N95. 23 (1944);.43 (1953: No. 43(1954) (Revised, 1958): and g; (1959). 88 TABLE 5 ACCOUNTING TREATMENT OF TAX EFFECTS RELATED TO REFUNDINGS BY PUBLIC UTILITIES DURING THE PERIODS 1936-1945 AND 1956-1965 Accounting Charged or 1936-1945 Other Dis sition (Lemke Report), 1956'1965 p0 Number Per Cent Number Per Cent Income 19 61 20 55 Retained earnings -- -- 6 l7 Accrued taxes —- -- 2 6 Deferred over- life of retired issue -- —- 3 8 life of new issue -- —- l 3 Not available 12 39 4 11 Totals 31 100 36 100 Sources: Lemke, Op. cit., p. 193; Moody's Manuals, and questionnaires. Since little information was available concerning the practice by industrials not using immediate write-off and the more recent data show the entire write-off to in- come in the first place for most cases, no similar analysis is presented for these firms. Procedure-Selection Reasons In the questionnaire used in the 1956—1965 survey, firms were asked to indicate the primary reasons for the choice of procedure used. The results are shown in Table 6 (p. 89). The relatively large prOportion of the utility 89 TABLE 6 STATED REASONS FOR CHOICE OF PROCEDURES 1956-1965 Public Utilities Reasons Deferred Immediate Write-off Write-off Industrials Deferred Immediate Write-off Write-off Amortization over period of interest savings due to issuance of refunding bonds at rate actually lower than rate on old bonds 4 -- View of refunding as pri- marily having the effect of extending maturity of the original debt to new due date -- -- Procedure approved by the American Institute of Certified Public Accountants 3 -- Procedure approved for Federal income tax purposes 2 2 Regulatory commission requirements 11 9 View of unamortized items as gain or loss in year of refunding -— 3 Materiality of amount 1 2 Convenience or expediency l -- Other -- 2 Totals 22 18 aDirect exchanges of issues in 2 cases; apparent misinterpre- tation of the question in 1 case. Deferred write-off is not accept- able for tax purposes except in connection with a direct exchange. bMultiple reasons indicated by 7 utilities and 3 industrials; totals include responses by 29 utilities and 8 industrials. Source: Questionnaire reSponses. 90 sample indicating regulatory requirements to be the basis of choice appears to be in line with the Clendenin findings. Among the list of possible reasons was included "effect upon dividend pattern." Interestingly, no firm checked this item. Also, there were fewer mentions of materiality, convenience, and expediancy than might have been expected because of the small amounts of financing costs relative to retained earnings and net income of firms refunding in recent years. Firms were asked to indicate whether write-off was in accordance with the period of expected realization of savings according to capital budgeting calculations used in making the refunding decision. An implied assumption of Hypothesis No. 2 was that some rate of return over the period to maturity of the new issue might have been used and the write—off basis determined accordingly. This was not indicated by the responses received. But the notion that capital budgeting considerations played some part in selection of accounting procedure seems borne out to some extent by the prOportion of firms shown by the practice statistics to have used a variation of payback time as the write-off period. In view of the small number of industrials that answered the questionnaire, the responses by this group 91 might be considered relatively inconclusive. But since Table 4 shows that most industrials used immediate write- off, the observations may be made that: l. The procedure is approved by the American Institute of Certified Public Accountants. 2. The procedure is approved for tax purposes. 3. Quantitative data subsequently presented show that industrials using immediate write-off had lower median balances of unamortized financing costs relative to net income and retained earnings before charge-off than either utilities or industrials using deferred write-off. Comparative Statistics of Net Income and Retained Earnings Relationships of Bond Financing Costs and Dividend Payouts It is the purpose of this section to present data related to Hypothesis No. l, which held that a pragmatic basis for gradual write-off of financing costs of refunded bonds could be inferred for public utilities. The grounds for such an inference were expected to be the greater rela— tive size of the unamortized balances for public utilities than for industrials. Hence, immediate write-off would impair earnings and dividend position of the utilities to a greater extent than of industrials. Investigation of this hypothesis included appli- cation of certain tests as follows: 92 1. The Wald-Wolfowitz runs test of representative- ness of the 1956-1965 data. 2. A percentage analysis of unamortized financing costs, the amount of such costs charged off in the year of refunding, and the annual dividend in relation to REBCO and NIBCO. 3. A median test of differences between dividends of firms deferring write-off of refunding— related financing costs and those of firms writing such amounts off immediately upon refunding. 4. A correlation test of association between dividends and refunding-related charges written Off in the year of refunding. The analysis subsequently presented shows some degree of support for the hypothesis. The amounts of bond financing costs used in this part of the study pertain solely to the issues retired by refunding and were obtained from sources indicated in Table 7 (p. 93). Although methods of selection of cases were not clearly disclosed by the Lemke and Clendenin reports, for purposes of the present study representativeness of the 1936-1945 data was assumed. Since the 1956-1965 data were obtained according to availability from the sources shown in Table 7 rather than by random sampling, nonparametric techniques were used for the statistical tests. These do not require specified conditions concerning the parameters of the pOpulation, e.g., a normal 93 TABLE 7 FREQUENCY OF DATA BY SOURCE AND PERIOD Period Source 1936-1945 1956-1965 Utilities Industrials Utilities Industrials Lemke study, supple- mented by investment manual reports 17 -- -_ __ Moody's Public Utility Manual 23 -- 4 _- Moody's Industrial Manual -- 33 -- Questionnaires -- -- 32 Securities and Exchange Commission files -— -- -- 17C Totals 40 33 36d 28" aLemke presented primarily balance sheet aggregates, with separate listing of amounts applicable Specifically to the refunded issues in 17 cases. Some of these were obtained from statements apparently issued some time in advance of actual retirement and did not reflect final amounts (see Lemke, op. cit., pp. 175, 184). Above totals include 28 utility and 27 industrial cases listed by Lemke. Amounts applicable to refunded issues were determined mainly from footnote data accompanying statements published in Moody's Manuals during this period of heavy refunding activity. bThe entire list of 1956-1965 refundings was initially obtained from the Manuals. Questionnaires were used to obtain details not adequately disclosed in the published data. cIn view of the limited responses by industrials, files were examined in the Washington office for data. dThese are the cases for which adequate data for the study were obtained out of 69 utility and 38 industrial refundings deter- mined from the Manuals to have occurred during the period. One industrial case involved no unamortized balance at refunding; hence, it is omitted from the analysis subsequently presented. 94 distribution.1 The Wald-Wolfowitz runs test was applied to estab- lish a basis as to representativeness of the cases for which data adequate for analysis in the study were obtained. Basically, this involves testing the null hypothesis (Ho) of no difference between two independent samples against the alternative hypothesis (H1) of a difference in any respect.2 Failure to reject the null hypothesis warrants a conclusion that the two samples have been drawn from the same pOpulation. For this purpose, cases for which data related to the refunded issues were available were viewed as one sample, and those for which such data were not available but were firms determined to have engaged in refunding during the period were viewed as another. Balance sheet aggregates of unamortized bond financing costs, retained earnings, and dividends (with one exception) were avail- able for all cases. The first two are affected by the accounting for refunding charges and the third is assumed lSidney Siegel, Nonparametric Statistics for the Behavioral Sciences (New York: McGraw-Hill Book Company, 1956), pp. vii, 31: W. J. Conover, Practical Nonparametric Statistics (New York: John Wiley and Sons, Inc., 1971), lpp. 93-94. 2Siegel, Op. cit., p. 136. 95 to be by Hypothesis No. l of this study. Thus, the public utility and industrial categories were each grouped as described above and tested on the basis of unamortized bond financing costs appearing in balance sheets at December 31 and the annual dividend for the year of refunding as per cents of retained earnings. REBCO could not be computed for the cases for which Charges applicable to the retired issues were not distinguished. Specific application involves combining the two variables in a single ranking in the order of increasing size. Then, the number of runs is determined, a run being defined as any sequence of values from the same group. Using the number of runs determined in this manner, Z values for samples larger than 20 are computed by the basic formula: 0r where r = number of runs r = mean of the pOpulation of r's 0r = standard deviation of the population of r's Rejection of Ho depends on the number of runs (r) being too small to generate a 2 that is significant at a lIbid., p. 140. 96 certain level. Significance of z is determined by refer- ence to a table of probabilities. If the observed 2 has an associated probability equal to or less than a pre- determined level of significance (a), then the null hypoth- esis may be rejected at that level of significance. Application of the test to the public utility and industrial variables yielded the results shown in Table 8 (p. 97) . Since the probabilities are greater than a - .01 for all of the 2 values, the null hypothesis is not rejected. No significant difference between the samples is indicated at a = .01. For the analyses other than the runs test, refunding—related variables and dividends were placed on common bases relative to REBCO. NIBCO, or both, depending on the test. The computations of REBCO and NIBCO are described and illustrated below. The purpose was to deveIOp figures such as might be examined by management in appraising the effects of accounting procedures, given that the desirability of refunding has been established on other grounds, such as interest savings. An assumption reflected in Hypothesis No. l is that the accounting pro- cedure to be used might be influenced by the relative size of balances prior to write-off. 97 TABLE 8 WALD-WOLFOWITZ RUNS TEST OF UNAMORTIZED BOND FINANCING COSTS AND DIVIDENDS AS PER CENTS OF RETAINED EARNINGS 1956-1965 Utilities Industrials Variables UBFC Dividends UBFC Dividends n1 Cases for which re- funding data were available 36 36 27 27 n2 Cases for which re- funding data were not available 33 33 11 10 N n1 + I12 69 69 38 37 2 Observed value .83 .35 .86 1.24 p Probability of 2 as extreme as observed value in normal distribution .21 .38 .21 .11 aDividend not available for one case. Sources: Methodology per Siegel, Op. cit., pp. 136-145, 247; Conover, Op. cit., pp. 349-356; raw data per sources tabulated in Table 7 (p. 93). REBCO (Retained Earnings Before Charge-Off) was computed by adding the bond-related charge-off and dividend payout back to the year-end retained earnings balance and subtracting the tax—saving equivalent, adjusted by the tax rate according to income statement treatment determined or inferred from the data examined. The actual effect of the saving obtained by deduc- tion of the aggregate balance of financing costs related 98 to a retired issue is present, regardless of statement treatment. Treatment of the tax—saving equivalent was not clear from available data for 14 utilities and 7 in— dustrials in the 1936-1945 group and for 4 utilities and 5 industrials in the 1956-1965 group. In the computations for these cases, deduction of a charge equal to the tax effect was assumed in accordance with the guidelines of Accounting Research Bulletin Nos. 18 (1942) and g; (1953).1 The tax rate used in the computations was the per cent borne by the tax—saving equivalent to the aggre- gate deduction in all cases in which this information was available. This was done because the tax provision shown in the income statement for many of the 1936—1945 cases was based on taxable income rather than book in- come. Inter-period tax allocation had not yet become popular at that time. Where neither the specific amount nor the tax rate was available, an estimate was made using the maximum rate to which the firm would be sub- ject in the year of refunding. The maximum rate was inferred to be more realistic than an average because of 1See p. 87, supra. 99 some evidence of the timing of refundings to take advantage of tax benefits in given years.1'2 The computation made for several utilities refund- ing after 1942 is illustrated below. Unamortized financing cost of $1,000 deducted in the tax return in the year of refunding results in a tax saving of $500 at an assumed 50 per cent rate. An amount equal to the tax saving is charged to income, and the balance of unamortized financing cost is reduced by the offsetting credit. The net-of-tax remainder is assumed to be written off on a straight-line basis over five years. Inter-period tax allocation is not used, at least with respect to the refunding effects. Hence, the actual tax charge in the income statement is $500 less than if the $1,000 deduction had not been avail- able. But net income is unaffected because of the off- setting charge equal to the tax saving. The refunding is assumed to occur on January 1, however, and the $100 amortization charge is reflected in net income transferred to retained earnings. NIBCO (Net Income Before Charge-Off) is determined by adjusting net income reported for the year for the 1Lemke, Op. cit., p. 152. 2Questionnaire comments of Puget Sound Power Com- pany regarding 1963 and 1964 refundings. 100 financing costs charged off and the tax effect. For the assumed case, the computation is as follows: Net income per statement $15,000 Add: Amortization $100 Tax-saving equivalent 500 600 $15,600 Deduct: Tax saving 500 NIBCO §15,100 The resulting figure would be reported net income in the absence of any write-off related to the retired bonds. It differs from reported income only by the amount of the annual amortization charge in the case assumed. If the refunding had been at the end of the year, NIBCO and reported net income would be the same. Assuming dividends of $4,400 were declared, REBCO is computed as follows: Retained earnings per statement $50,000 Add: Amortization per income statement $100 Tax-saving equivalent per income statement 500 600 Dividend charged to retained earnings 4,400 $55,000 Deduct: Tax saving 500 REBCO §54,500 The resulting figure would be the retained earnings balance in the absence of any write-off related to the re- tired bonds of any dividend charge. REBCO differs from reported retained earnings by the amount of the dividend 101 plus the amortization charge in this case. It is necessary to adjust retained earnings for the latter because of its reflection in reported net income as actually transferred to retained earnings. The observations of Clendenin and Lemke that public utilities tended to have greater balances of unamortized refunding-related items relative to retained earnings than the industrials during the 19305 and 19405 are supported by the general pattern of per cents shown in Table 9 (p. 102). The range and median per cents are considerably higher for utilities in both write-off categories than for industrials in the 1936-1945 group. Also, those for public utilities deferring write-off are higher than for public utilities writing the amounts off immediately in the year of refunding. The very high median for unamortized financing costs related to NIBCO for the deferring utilities indicates that more than half of these firms would have shown losses in the income statement had the entire charge been made to income in the year of refunding. During the 1956—1965 decade, the percentages in the various categories had declined to considerably lower levels compared to the 1936-1945 group. There were some excep- tions at the upper limits of the ranges of per cents of NIBCO for the industrials, but the median appears to best 102 TABLE 9 PERCENTAGE RELATIONSHIPS OF REFUNDING-RELATED ITEMS TO RETAINED EARNINGS AND NET INCOME BEFORE CHARGE-OFF Period and Write-off Public Utilities Industrials Procedure N Median Range N Median Range Immediate write-off 1936-1945 Unamortized financing costs- As per cents of REBCO 23 28 2 -71 30 4 l -36 As per cents of NIBCO 23 73 7-113 30 16 3 -55 Amounts charged off-(a) As per cents of REBCO 23 12 l -39 30 2 0 -24 As per cents of NIBCO 23 26 l -65 30 9 1 -39 Deferred write-off 1936-1945 Unamortized financing costs- As per cents of REBCO 17 53 lO-109 3 13 6 -29 As per cents of NIBCO 17 118 26—169 3 33 ll —61 Amounts charged off-(b) As per cents of REBCO l7 3 0 -l9 3 1 O -5 As per cents of NIBCO l7 5 0 -27 3 3 1 ~11 Immediate write-off 1956-1965 Unamortized financing costs- As per cents of REBCO l9 3 O -20 21 l 0 -19 As per cents of NIBCO 19 10 O -24 21 3 O —67 Amounts charged off-(a) (C) As per cents of REBCO l9 2 0 -ll 21 O 0 -9 As per cents of NIBCO 19 S O -11 21 l O —32 Deferred write-off 1956-1965 Unamortized financing costs- As per cents of REBCO 17 3 0 -72 6 2 l -27 As per cents of NIBCO l7 9 1 —60 6 12 3—150 (b) Amounts charged off- ( ) (c) As per cents of REBCO 17 0 C 0 —3 6 O 0 —1 As per cents of NIBCO 17 O 0 -3 6 l O -7 (a) Net of tax-saving equivalent. (b)Annua1 amortization of net-of—tax amount. (°)Less than 0.5% rounded to zero. Sources: Raw data from sources tabulated in Table 7 (p. 93). 103 represent the general pattern. Since potential effect on dividend paying ability was assumed to be an important influence in the selection of write-off procedures in formulating Hypothesis No. 1, tests for significant differences between the median dividends as a per cent of REBCO for firms deferring write-off of refunding—related balances and that of firms using immediate write-off were applied. The null hypothesis is that there is no difference between the medians of the populations represented by the two groups. The alterna— tive hypothesis for a one-tailed test is that the median dividend of the pOpulation of firms deferring write-off of bond charges is higher than that of the pOpulation of firms writing such costs off immediately upon refund- ing. The results of the test are shown in Table 10 (p. 104). The results shown in Table 10 are significant at a = .01 for the 1936—1945 utility group and a = .05 for the 1956-1965 utilities. The result for the latter is not significant at the .01 level. Hence, the result for the first group is stronger, although .05 might be regarded as a reasonable standard. The conclusion is that a differ- ence significant at the levels indicated existed between the median dividend of utilities deferring write-off of 104 TABLE 10 RESULTS OF MEDIAN TEST OF DIVIDENDS OF FIRMS USING IMMEDIATE AND DEFERRED WRITE-OFF PROCEDURES FOR FINANCING COSTS RELATED TO BONDS REFUNDED 2 X Test Fisher Test £82106 an: Observed 0 Observed o n ustry roup N Values Values 1936-1945 Public Utilities 40 8.59 .01 -- --b Industrials 33a -— —- 14 -- 1956-1965 Public Utilities 36 4.01 .05 -- -- Industrials 27 -- 9 __b aComputation included 30 cases. Median and values next above and below were drOpped to accommodate critical value table for a maximum N of 30. bMaximum level of significance shown by critical value table used is .05. Not significant at a g .05. Sources: Methodology per Siegel, 0p. cit., pp. 95-115, 249, 256-270; Conover, Op. cit., pp. 167-172, 367; raw data per sources tabulated in Table 7 (p. 93), refunding costs and that of utilities using immediate write-off in each of the two periods. The one-tailed test requires testing on the basis of probabilities equal to or less than one-half the value of alpha. The probabilities of occurrence of the values shown in the table under the condition of Ho (no signifi- cant difference) are p<.005 for the 1936-1945 utilities and p<.025 for the 1956-1965 utilities. Hence, the null 105 hypothesis (H0) is rejected in favor of the alternative hypothesis (H1) of higher dividends of utility firms using gradual write-off of unamortized financing costs of bonds refunded. Due to certain values too low for relevant use of the chi-square test, the Fisher Exact Probability Test was used for the industrial cases. The results shown in Table 10 are not significant at a:.05. The null hypothesis is not rejected for industrials in either period. Taking the median per cents shown in Table 11 (p. 107) as the most representative indicators of the general pattern, public utilities in all categories tended to pay higher dividends relative to both REBCO and NIBCO than the industrials. The seemingly very high upper limits of the ranges for the 1956-1965 industrial dividends repre— sent very few cases. Since results of the median test indicated signifi- cant differences between relative dividends of utilities in the two write-off categories, correlation coefficients were computed to test for association between dividends and refunding costs actually charged off. A question related to Hypothesis No. l is whether firms experience an inverse relationship between these two variables or adOpt accounting procedures having the effect of precluding such a 106 relationship. The latter point would be supported by failure to reject the null hypothesis of no correlation. The dividends and bond charges were tested as related to REBCO since retained earnings is ultimately affected by both variables. Dividends are not charged against income in the income statement, although the amount of net income may affect dividend payments. The results of the Spearman Rank Correlation test are shown with the percentages in Table 11. All are posi- tive. Those for industrials using immediate write—off are significant for both periods at g<.05. But for tests of significance of correlation coefficients to be relevant, the samples should be randomly drawn.1 Also, rejection of the null hypothesis on the basis of a one-tailed test re- quires observed values in the direction predicted by H1, negative coefficients in this case.2 Assuming randomness, partly supported by results of the runs test, the correla- tion results do not warrant rejection of HO. Although deliberate selection of accounting pro— cedures on the basis of effects on dividend position cannot be proved by the means used in this research, some measure 1Siegel, op. cit., p. 210. 21bid., p. 211. DIVIDENDS DECLARED IN THE YEAR OF REFUNDING AS PER CENTS OF REBCO AND NIBCO AND AS 107 TABLE 11 CORRELATED WITH BOND FINANCING COSTS CHARGED OFF AS PER CENTS OF REBCO Dividends and Category by Refunding— Dividends Related Write-off Procedure N Median Range Bond Charges Correlated and Period r s Dividends of Public Utilities Immediate write-off 1936-1945: As per cents of REBCO 23 23 11 ~59 .21 As per cents of NIBCO 23 59 23 -98 Deferred write-off 1936-1945: As per cents of REBCO 17 36 23 ~81 .03 As per cents of NIBCO 17 61 35 —93 Immediate write-off l956-l965: As per cents of REBCO 19 22 13 -56 .36 As per cents of NIBCO 19 71 23-133 Deferred write-off 1956-1965: As per cents of REBCO 17 38 0-111 .34 As per cents of NIBCO 17 72 0-120 Dividends of Industrials Immediate write-off 1936-1945: As per cents of REBCO 30 14 4 ~39 .55 As per cents of NIBCO 30 50 15~107 Deferred write-off 1936-1945: As per cents of REBCO 3 18 9 ~20 .54 As per cents of NIBCO 3 39 19 ~51 Immediate write-off 1956-1965: As per cents of REBCO 21 11 0 ~62 .38 As per cents of NIBCO 21 48 0-654 Deferred write-off 1956-1965: As per cents of REBCO 6 l 0 ~10 .33 As per cents of NIBCO 6 5 0-168 Sources: Methodology per Siegel, op. cit., pp. 202—213, 284; Conover, op. cit., pp. 243-249, 390; raw data per sources tabulated in Table 7 (p. 93). 108 of support for Hypothesis No. 1 appears to be established. On balance, the three-way analysis-~consisting of percent- age comparisons, the median test, and the correlation test-~seems to support an inference of a pragmatic basis for deferred write-off related to the dividend pattern of public utilities. The evidence, however, appears much stronger for the 1936-1945 period than for 1956-1965, when the percentages were much lower. Summary In this chapter, historical comparisons of account- ing practices related to bond refunding by public utilities and nonregulated industrial firms and certain analyses of size-relationships of items involved were presented. Public utilities practiced immediate write-off of refunding-related balances of discount, issue costs, and call premium to a greater extent during the 1956-1965 period than during 1936—1945. The charge was usually made to retained earnings directly during both periods. A majority of utilities sampled for earlier studies used deferred write off. The majority of industrial firms reported upon used immediate write-off during both periods. The 1956- 1965 group charged income exclusively, whereas the 109 1936-1945 group often charged retained earnings. Little use was made of write—off over the life of the new bond issue by either utilities or industrials during 1956-1965. This finding is in contrast to the ex— pectation of Hypothesis No. 2 for public utilities. Regulatory requirements were cited most frequently as reasons for the choice of accounting procedures by utilities during both periods. Reasons were stated by very few industrials in the l956~l965 group. Approval for tax purposes was mentioned in both periods. Approval by the American Institute of Certified Public Accountants was mentioned for 1956~l965. As per cents of REBCO and NIBCO, unamortized financing costs related to bonds refunded were consider- ably larger for public utilities than for industrials in corresponding write—off categories in the 1936-1945 samples. Utilities deferring write-off tended to have greater balances than those using immediate write-off. Differences existed in the 1956-1965 samples, but percentages in general were much lower than in the earlier period. Public utilities in all categories tended to pay higher dividends, as measured by the median per cents of REBCO and NIBCO, than industrials. By means of a chi-square median test, significant differences between dividends of 110 public utilities deferring refunding—related charges and those using immediate write—off were determined for both periods. Similar tests for industrials did not produce significant results. Correlation results for dividends and bond charges were not strong, and did not refute inferences based on the other tests. In general, the tests applied appear to support the assertion of Hypothesis No. l of a pragmatic basis for deferred write-off of unamortized financing costs of refunded bonds related to effects on earnings and dividend position of public utilities during the 1936-1945 period. Results for 1956-1965 were less conclusive in view of the small percentages involved. CHAPTER IV APPLICATION OF GUIDELINES FROM RECENTLY DEVELOPED MODELS OF ACCOUNTING THEORY The purpose of this chapter is to present the Hendriksen approach to accounting for bonds payable and related interest charges and to determine the degree of support for that or some other procedure that may be in- ferred from the guidelines contained in two other deduc- tive models. All of these models have been purported by the authors or by others to have been develOped through some variations of the methodology of deductive logic. All explicitly state or implicitly suggest that their aim is to contribute toward a logically unified body of accounting theory to facilitate realistic measurement and reporting of economic data at the firm level. The Hendriksen approach is described and illus- trated in the chapter. In addition, the supporting theoretical argument is summarized. The approach develOped in each of the other models is described, and guidelines prOposed that are related closely to accounting for long- term debt are summarized and discussed. Inferred 111 112 conclusions are stated as to whether the models support the Hendriksen procedure, referred to in this study as the "full-accrual" approach, or some other. Extensive deriva- tions will not be attempted, but the general methodology of each model will be summarized. In some cases, inconsisten- cies will be pointed out and modifications suggested to show how the argument might be reframed to logically lead to the full-accrual approach. The Hendriksen Model In his "eclectic" model, Hendriksen draws upon various forms of methodology but portends to emphasize the deductive process, which he says must accompany any com- plete theory formulation.l Hendriksen's view of the essen- tial elements of the deductive framework as applicable to accounting is summarized as follows: The structure of the deductive process should include the following: (1) the formulation of general or specific objectives of financial reporting; (2) a statement of the postulates of accounting concerning the economic, political, and sociological environment in which accounting must Operate; (3) a set of constraints to guide the reasoning process; (4) a structure, set of symbols, or framework in which ideas can be expressed and 1Eldon S. Hendriksen, Accounting Theory (Rev. ed.: Homewood: Richard D. Irwin, Inc., 1970), p. 18. 21bid., p. 4. 113 summarized: (5) the deve10pment of a set of definitions: (6) the formulation of principles or generalized statements of policy derived by the process of logic; and finally (7) the application of the principles to specific situations and the establishment of procedural methods and rules. Although Hendriksen does not show rigorous deriva- tion of principles in the sense of formal logic, he seems to have fairly observed the structure stated above in an informal but systematic and well—reasoned manner. Hendriksen claims that formulation of objectives is the most important element of the deductive system since different objectives may require different structures and 1 lead to different principles. He holds the recording of expenses and losses in the determination of current income to be perhaps the most important objective of both asset and liability valuation.2 He also says that liability measurement should provide information to investors and creditors that is useful for predictive purposes, for inter- period and intercompany income comparisons, and for com- 3 paring the claims of the various equity holders. He indicates that an evaluation of the user's needs should 2 . Ibid., p. 450. 3 Ibid. 114 be included in a logically derived structure for the deve10pment of sound accounting practices.1 The Full-Accrual Approach Hendriksen does not use the term "full—accrual" to refer to the procedure recommended for use in accounting for bond liabilities and interest charges. But the label is convenient and seems to fit the process that is described. Hendriksen advocates that the bond liability valu- ation should be the discounted present value of all future payments anticipated in the contract.2 The apprOpriate discount rate is held to be the yield rate currently pre- vailing in the market for bonds of similar risk and term. Discounting all future payments by the current market rate results in a value equal to the market price of the bond. In short, application of the full accrual procedure involves stating the liability at the current market price of the bonds and making the interest charge on the basis of the current market rate. The market has taken care of the discounting process at a given time inasmuch as there exists an inverse relationship between changes in the lIbid., p. 94. 21bid., p. 452. 31bid. 115 market yield rate and the market price of the security. Hendriksen regards the quantification of the bond liability and interest charge by discounting at the original yield rate as something similar to historical cost.1 In contrast, use of the full—accrual approach results in cur- rent cost valuation and recognition of holding gains and losses as changes in market yield rates and bond prices occur. Hendriksen considers gains to represent net favor- able results of transactions or events not related to normal Operations of the business and losses as the results of un— favorable occurrences of these.2 He emphasizes that losses are unrelated to the Operations of ppy period and that they are value expirations rather than cost allocations.3 In the absence of general price-level changes, differences between historical cost and specific prices arising during a period are regarded as holding gains and losses.4 That is, for bond liabilities and related interest charges the holding gains or losses would be the differences 1Ibid. 2Ibid.. pp. 190~l96. 3Ibid., p. 196. 4Ibid., p. 204. 116 between current market price and yield rate quantification and the corresponding figures which would result from use of the historical yield rate in the accrual process. The Egg; gains and losses related to the bond value and interest rate changes are represented by the effects on the assets of the entity. The loss resulting from in increase in the liability value is the decline in previously uncommitted assets, i.e., within the Hendriksen framework, some pgt asset value has eXpired. The expiration is viewed as a holding loss because the firm has held the debt during the increase instead of retiring it when the market price was lower. As the liability value increases, the corresponding decline in the market rate of interest results in a holding loss because the firm has remained committed to greater outlays for use of the funds than would be necessary to obtain them at the current rate. Similarly, decreases in bond values and increases in the market rate of interest result in holding gains. Recognition of holding gains and losses isolated as applicable to the interest rate changes may be regarded as reflecting the efficiency of financial management. Although the changes are market-determined, a holding gain may be interpreted to mean that management has borrowed 117 favorably. The commitment to the historical rate is management-determined. Stating the liability at its current value may be justified not only on the basis of the effects on income measurement but because it is necessary to reflect finan- cial position realistically in the balance sheet. The market value of the bonds simply represents the price at which the debt could be retired at the balance sheet date; i.e., it is the effective amount of the liability in the truest sense. And Hendriksen says that one of the objec— tives of price-level adjustments (presumably both general and specific) is to restore the balance sheet to the posi- tion of a respectable and useful financial statement. Making the expense charge in the income statement according to the current yield rate may be justified on the basis of several factors. Hendriksen indicates that accounting practice involves application of a combination of several concepts of income.2 No one concept seems to serve all purposes equally well. But he says that: While it may or may not be desirable, it is a realistic fact that users of accounting state- ments usually interpret net income to mean the return to shareholders. 11bid., p. 213. 2Ibid., p. 143. 31bid., p. 153. 118 Basing the interest charge on the current rate is in harmony with the economist's concept of imputed cost. The current rate represents the interest which the entity would have to incur if the funds were not already avail- able. Since the difference between historical interest and the charge based on the current rate is not actually paid to the bondholder, it accrues as an increment or a decrement to the entrepreneur (the shareholder group for the corporation). Accounting for the interest charge and the bond liability as well in current terms is in harmony with the all-inclusive concept of income, since the income under this concept includes gains and losses from all sources arising during the period. Hendriksen mentions that it is important to dis- tinguish between income and return of capital.1 The full— accrual procedure appears to be quite consistent with the capital-maintenance concept of income. While there are different views of capital—maintenance, one is that for a firm to maintain capital it must stand ready to Operate at the end of a period at a capacity as great as that at the beginning. For financial statements to Show whether or not this is the case, the capital must be stated in 1Ibid.. p. 126. 119 terms of values or prices required to measure such capacity according to conditions prevailing at the balance sheet date. If price changes have occurred during a period, the net increase or decrease in capital represented by the residual income statement figure must reflect such changes. Otherwise, capital may erroneously be construed to have been maintained. In such a case, outlays or commitments for, say, dividends may be based on an amount regarded as income when it actually includes recovery of capital, the expira- tion of which has not been accounted for in the income statement. Hendriksen also points out that capital maintenance is important to bondholders and other suppliers of capital as well as residual equityholders.l They are interested in adequate capital to protect their claims and anticipated return on investment. The implication of the full—accrual approach for refunding is that it results in stating the liability at the call price on the date of retirement. Discount and premium adjustment is automatically achieved through accrual according to the current yield rate and bond value. 1Ibid., p. 127. 120 The only write-off required in the year of refunding is that which would be necessary to adjust the accounts for changes to the date of call. At the date of retirement, the call price is the lowest amount at which the debt can be settled at the time.1 Hence, the call price becomes the relevant valuation at the retirement date. Even if the discounting process based on the current yield rate for bonds of similar risk and term should result in a figure that differs from call price, there is no need for the firm to pay more to retire the bonds. And the rational investor is presumed to be unwilling to settle for less than the call price. The process from issuance to retirement is illus- trated in Table 12 (p. 122). Purely for the purpose of illustrating the accounting procedure, certain assumptions are made as follows: 1. Bonds of $100,000 face value, due in 20 years and bearing interest at 5 per cent payable annually, are issued at 78.81, a price to yield 7 per cent. 2. The contract contains an Option permitting call of the bonds at 105 on or after December 31 of the third year. 3. The general price level remains constant during the periods covered in the example. 1Ibid., p. 453. 121 4. At the end of the first year the market rate of interest has risen to 9 per cent. 5. The market rate of interest has drOpped to 8 per cent by the end of the second year, and to 4% per cent by December 31 of the third year. 6. Allowing for possible tax effects and other relevant matters, management makes capital budgeting calculations indicating that re- funding at the end of the third year will result in a substantial saving. At that time the old issue is called and replaced with a new one at the reduced interest rate. Computation of the bond value at issuance is as follows: $100,000 of principal to be paid in 20 years, discounted to present value at 7 per cent $25,840 $5,000 of interest to be paid per year for 20 years, discounted to present value at 7 per cent 52,970 Current market value of bonds $78,810 At the end of the first year, the current bond value, computed via the discounting process at the new rate for 19 years, would be $64,200. Conceptually, accrual should be continuous as the rate changes during the year, but for convenience in the illustration the interest charge is com- puted on the basis of average rate and bond value as follows: $78,810 + $64,200 Average bond value = 2 = $71,505 Average interest rate = ’07 2 '09 = .08 Total interest charge = $71,505 x .08 = $ 5,720 122 .uSOEemo unencusfi Hmsfiw pom meson mo uooeouauom AwV .OHbNSma meson so mmOH no swam wcfioaom Amv .momumno ummuousa so mmoa no seem mswoaom on .uamo you n mo sneaked umououaa Hmsoc< on .uooo poo 5 mo dump madam Hmoauoumwn um ummuouafi mo Hmouoo< Anv .oocmsmma mo memo um wawouooou HmauHsH Amv "wmx omm Ase mom Ase newness nauseous as seam masseom ANH.mHon mabmmmm meson so Seem wswpaom ooo.m Ase ooo.m Ase ooo.m Ase ooo.m Ase ooo.n Ase ooo.m Ase neonate sensuous possess www.mm Amv mua.m on man ADV Non on ooo.moaAmV mmm on cam.wnAmv mma.maflmv oabmsma mpsom em Apv mowuoco ummuousfi so mmoH msfioaom mmm.~m Ase ame.a Ase seanssa asses as mass aaaseom omm Ape mom Ape so Ase Nem.m Ase mmm.m Ase aen.m Ase seasons assesses sass ooo.oaaawv ooo.m AOV ooo.m on oaw.mnAmv ammo .uu .un .uo .un .uo .un usoooo< m use» N “new H poor amaoua omHm ma H.0N0 unsouu< .mwumnu cowumuwuuoam uapoaumm mnu mm unsouum 08mm wnu Ou wpma ma mwumnu mmolmufina mamapmaaw 0:90 .mawufi :umnuo: aoum pmfimawmmao hawumummww mum nowna .mmwumao uwwumucfi Gnu ou uomwmm waH>Hm “mama ..m.H .waooaH xoon xmumua mo ucmu “we 00 ma mmsmmxm xmu maooaa HmuOH .N umm» OH 000.000 0am H Hmmw cw maoocfi mabmxmu 000.000 00 ucmu pmd on mm pwusdaoo mabmhma xmu Hmsuum saws pwwumnu ma H.000 unsoou< .cOHumOOHHm mo pozuma pmuumwmam 00m.5m m 00~.5N w 000.5N w 000.55 0 000.5N m 000mmm m msoocH umz HHHHHHII IIIIII 1111:: IIIIII IIIIII IIIIII mamuH mumawpuomuuxm 000.0 0 000.0 0 000.0 w 000.0 0 000.0 m 000.0H m 000 000 000.H 000.H Inuit: 000.0H AH.0NOV upon pmuasvummm so Ammo; mo coaumufiuuoa< ooo.m w ooo.m m ooo.m m ooo.n m ooo.m m ooo.m m Aemso upon as :0 ammumuaH mmwumno ummumuaH llllll 1111:: IIIIII 1111:: IIIIII Iluunn mcoauuspma 0cm mBOOaH umcuo 005.~m w 005.~m w 000.mm w 000.mm m 000.Nm w 000.5m w Asusumm huaawumv «EoosH wawumumoo huAHHuD umz 00~.50 m 00w.50 w 000.50 0 000.50 w 000.50 000.N0 w Aommv Aoomv A...HOV .uousHa .>opm 005.0 000.0 IIIII I 11:11: mAH.0H0v mmeH maooaH 0wuummmn How sOHmH>oum 000.5N 000.- 000.5N 000.- 000.5N 00m.- AH.0000 .pmmnmmxmy maoucH 000.00 0 000.00 0 000.00 0 000.00 0 000.00 0 000.00 0 msoaum> . ”mumsmdxm wswumumao 000.00Hw 000.00Hw 000.00Hw 000.00Hw 000.00Hw 000.00Hw wasm>om wswumumdo maoosH wcwumummo muHHHuD N umww H Haw» N Hum» H umm» N was» a Ham» mammH smz mo wmfiq mammH 0H0 mo mafia maoosH Ou nuclmufiuz aowunwuommn um>0 nuclmufiu3 Hmapmuo mumfiposEH onH