A COMPUTER SIMULATION ANALYSIS OF
ALTERNATIVE METHODS OF ACCOUNTING
FOR UTILITY CONSTRUCTION WORK IN PROGRESS
GORDON DOUGLAS QUICK
A DISSERTATION PRESENTED TO THE GRADUATE
COUNCIL OF THE UNIVERSITY OF FLORIDA IN PARTIAL
FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
UNIVERSITY OF FLORIDA
3 1262 08SS2 8502
This dissertation is dedicated to--
My parents, who have always given unselfishly,
Jeanelle, who means everything to me.
I would like to thank the University of Florida and countless out-
standing professors for making this dream a reality. There are many
professors that provided instruction, stimulation, and friendship who
cannot be named individually; to them I express my most sincere apprecia-
I am most grateful to the members of my supervisory committee, Dr.
Eugene F. Brigham, Dr. Russ Fogler, Dr. Sanford Berg, Dr. Rick Jesse, and
Dr. Milford Tysseland for their time and effort on my behalf. A most
respectful and sincere special thanks is extended to Dr. Brigham, who as
chairman provided the guidance, criticism, and encouragement necessary
to the preparation of a dissertation.
The Public Utility Research Center and its members are gratefully
acknowledged for their financial support and the facilities made avail-
able. I am indebted to Messrs. James Taggart and Dick Koble of the Tampa
Electric Company for their most capable assistance in the developmental
stages of the project.
I would like to thank Dr. Normal Keig who, perhaps unknowingly, has
been so helpful. His frank opinions, advice, and friendship have been
A special note of thanks goes to Barbara Brown, for her excellent
typing and her thoughtful assistance in so many ways.
My acknowledgments would be incomplete unless three other people
are mentioned. To my parents I owe more than I can express. Were it
not for their constant prompting and encouragement, understanding and
support, I would not have aspired to, let alone achieved, this degree.
Finally, and most important, is my wife, Jeanelle, who read drafts,
edited, and helped in numerous other ways. Her greatest contribution,
however, was her tolerance, understanding, and encouragement, not only
in the preparation of the dissertation, but in the years before the dis-
sertation was even considered.
TABLE OF CONTENTS
LIST OF TABLES vii
LIST OF FIGURES xi
1 INTRODUCTION 1
Objective of the Study 2
Scope of Interest 3
Plan of the Study 4
2 PROVIDING A RETURN ON CONSTRUCTION WORK IN PROGRESS 6
The Return on Construction Work in Progress 6
The Increasing Interest in Construction Ac-
Providing a Return on CWLP 14
Problems Encountered in Providing a Return on
3 THE PROTOTYPE FIRMS 33
Derivation of the Alternative Firms 33
The Firm as an Aggregation of Assets 42
Illustration of the Alternative Firms with the
Single Asset Model 45
The Continuous Growth Model 64
Examination of the Results 66
Depreciation Policies 76
4 THE SIMULATION MODEL 82
General Features of the Model 83
Input Options 93
Output Variables 99
APPENDIX A 114
Variable Definitions 115
TABLE OF CONTENTS (CONTINUED)
APPENDIX B 125
Input Value Combinations 126
Validation of Subroutines 127
Validation of Equations 130
APPENDIX C 137
5 ANALYSIS OF THE RESULTS 169
The Typical Firm Case 170
The Atypical Firm Case 247
APPENDIX A: THE SENSITIVITY ANALYSIS 259
APPENDIX B: THE EARLY PERIODS 269
APPENDIX C: THE LARGE FIRM CONSTRUCTION PROGRAM 275
APPENDIX D: INCLUDING THE FUNDS GENERATED BY
DEPRECIATION IN THE CONSTRUCTION BUDGET 285
6 CONCLUSIONS AND IMPLICATIONS 288
A Summary of the Results, their Significance,
and Some Conclusions 288
BIOGRAPHICAL SKETCH 315
LIST OF TABLES
1: RATIO OF IDC TO EARNINGS BASED ON 114 ELECTRIC
2: SUMMARY COMPARISON OF METHODS RB AND IDC 20
3: NET INCOME AND INTEREST CAPITALIZED. FOR-PRIVATE
OWNED ELECTRIC UTILITY INDUSTRY 1965-1971 28
4: RELATIONSHIP OF INTEREST DURING CONSTRUCTION TO
EARNINGS AND DIVIDENDS FOR SELECTED UTILITY
COMPANIES IN 1970 29
5: ILLUSTRATION OF THE INTEREST-TAX BENEFIT 36
6: SUMMARY OF THE KEY DIFFERENCES BETWEEN THE FIRMS 42
7: LIFE CYCLES OF ASSETS FOR A FIRM OVER A THREE-
YEAR PERIOD OF GROWTH 43
8: SINGLE ASSET: BALANCE SHEETS AND INCOME STATE-
MENTS FOR FIRM RB 48
9: SINGLE ASSET: BALANCE SHEETS AND INCOME STATE-
MENITS FOR FIRMS RB-IDC AND IDC 52
10: T-ACCOUNTS FOR FIRM RB-IDC FOR 1971 TRANSACTIONS
REGARDING IDC 54
11: SINGLE ASSET BALANCE SHEETS AND INCOME STATE-
MENTS FOR FIRM RB-IDC-ITB 56
12: COMPARISON OF RESULTS WHEN THE RATE FOR COMPUTING
IDC DIFFERS FROM THE ALLOWED RATE OF RETURN FOR
FIRM RB-IDC 61
13: SINGLE ASSET: BALANCE SHEETS AND INCOME STATE-
MENTS FOR FIRM IDC-ITB 62
14: REVENUE REQUIREMENTS: SINGLE ASSET MODEL 67
15: TIMES-INTEREST-EARNED : SINGLE ASSET MODEL 72
LIST OF TABLES (CONTINUED)
16: REVENUES: CONTINUOUS GROWTPIH MODEL 74
17: TIMES-INTEREST-EARNED : CONTINUOUS GROWTH MODEL 76.
18: ILLUSTRATION OF THE EFFECTS OF IDC AND THE ITB
FOR FIRMS IDC-ITB-G AND IDC-ITB-N 80
19: COMPARISON OF DEPRECIATION OF FIRMS RB AND IDC 86
20: GROWTH RATE PATTERN 95
21: VALUES DERIVED USING THE NFCP 128
22: VALUES DERIVED USING TH1E LFCP 129
23: VALUES DERIVED FROM THE BOOK DEPRECIATION SUB-
24: VALUES DERIVED FROM THE ACCELERATED DEPRECIA-
TION SUBROUTINE 130
25: INPUT VALUES FOR THE TYPICAL FIRM CASE 171
26: REVENUE REQUIREMENTS: SINGLE ASSET MODEL 176
27: SUMMARY OF TYPICAL FIRM CASE RESULTS FOR DIF-
FERENT ASSUMPTIONS 239
28: INPUT VALUES FOR THE ATYPICAL FIRM 249
29: UTILITY RATES 251
30: IDC TO EARNINGS RATIOS 251
31: IDC TO DIVIDENDS RATIOS 252
32: CASH FLOW AS A PERCENT OF THE CASH FLOW OF FIRM
33: CASH FLOW PER SHARE AS A PERCENT OF CFS OF FIRM
34: TIMES-INTEREST-EARNED RATIOS 254
35: PERCENTAGE CHANGE IN VARIABLES AS A RESULT OF 10
AND 20 PERCENT CHANGES IN THE EQUITY RATE OF
LIST OF TABLES (CONTINUED)
36: PERCENTAGE CHANGE IN VARIABLES AS A RESULT
OF 10 AND 20 PERCENT CHANGES IN THE GROWTH
37: PERCENTAGE CHANGE IN VARIABLES AS A RESULT
OF 10 AND 20 PERCENT CHANGES IN THE DEBT
38: PERCENTAGE CHANGE IN VARIABLES AS A RESULT
OF 10 AND 20 PERCENT CHANGES IN THE INTEREST
39: PERCENTAGE CHANGE IN VARIABLES AS A RESULT
OF INCREASING THE CONSTRUCTION PERIOD TO
SIX AND SEVEN YEARS 268
40: ONE COMPLETE CYCLE OF CONSTRUCTION PERIOD
41: PERCENTAGE DIFFERENCE IN UTILITY RATES BETWEEN
THE EARLY AND STABLE PERIODS 271
42: PERCENTAGE DIFFERENCE IN IDC/EARNINGS RATIOS
BETWEEN THE EARLY AND STABLE PERIODS 272
43: PERCENTAGE DIFFERENCE IN CFS AS A PERCENT
OF FIRM RB BETWEEN THE EARLY AND STABLE
44: PERCENTAGE DIFFERENCE IN TIMES-INTEREST-
EARNED RATIOS BETWEEN THE EARLY AND STABLE
45: PERCENT OF THE ACTIVITIES .ALLOCATION DEVOTED
TO THE VARIOUS PROJECTS 276
46: STABILIZED IDC/EARNINGS RATIOS 280
47: STABILIZED IDC/DIVIDENDS RATIOS 280
48: CFS AS A PERCENT OF FIRM RB 282
49: STABILIZED TIMES-INTEREST-EARNED RATIOS 284
50: VALUES OF OUTPUT VARIABLES WITH DEPRECIATION FUNDS
INCLUDED IN THE CONSTRUCTION BUDGET 286
LIST OF TABLES (CONTINUED)
51: PERCENTAGE CHANGE IN OUTPUT VARIABLES FROM THE
TYPICAL CASE WHEN DEPRECIATION FUNDS ARE IN-
CLUDED IN THE CONSTRUCTION BUDGET 287
52: IDC TO EARNINGS RATIOS OF 116 FIRMS IN 1972 300
LIST OF FIGURES
1: DERIVATION OF ALTERNATIVE FIRMS 34'
2: REVENUE REQUIREMENTS: SINGLE ASSET MODEL 69
3: TIM~ES-INTEREST-EARNED RATIOS: SINGLE ASSET
4: REVENUE REQUIREMENTS: CONTINUOUS GROWTH
5: TIMES-INTEREST-EARNED RATIOS: CONTINUOUS
GROWTH MODEL 75
6: PATTERN OF CWIP FOR THREE CONSTRUCTION AC-
7: CWIP TO TOTAL ASSETS RATIOS (FIRM RB) 92
8: UTILITY RATES (NORM.) 179
9: FUTILITY RATES 182
10: IDC TO EARNINGS RATIOS (NORM.) 186
11: IDC TO DIVIDENDS RATIOS (NORM.) 188
12: CASH FLOW OF EACH FIRM AS A PERCENT OF FIRM
RB (NORM.) 190
13: CASH FLOW PER SHARE AS A PERCENT OF FIRM RE
14: CASH FLOWS COMPARING DEPRECIATION POLICY 193
15: ADDITIONAL STOCK (NORM.) 194
16: ADDITIONAL DEBT (NORM.) 195
17: ADDITIONAL STOCK COMPARING DEPRECIATION
18: TIMES-INTEREST-EARNED RATIOS (NORM.) 198
LIST OF- FIGURES (CONTINUED)
19: TIMIES-INTEREST-EARNED RATIOS COMPARING DEPRECIA-
TION POLICY 200
20: TIMES-INTEREST-EARNED RATIOS COMPARING DEPRECIA-
TION POLICY 201
21: RATE OF RETURN ON EQUITY SETTING THE UTILITY
RATE IN PERIOD ZERO (NORM.) 224
22: RATE OF RETURN ON THE RATE BASE SETTING THE
UTILITY RATE IN PERIOD ZERO (NORM.) 227
23: TIMES-INTEREST-EARNED RATIOS SETTING THE UTILITY
RATE IN PERIOD ZERO (NORM.) 229
24: RATE OF RETURN ON EQUITY SETTING THE UTILITY RATE
IN PERIOD ONE (NORM.) 232
25: RATE OF RETURN ON THE RATE BASE WITH ONE PER-
CENTAGE POINT LIMITS 255
26: UTILITY RATES (LFCP) 277
27: IDC TO EARNINGS RATIOS (LFCP) 279
28: CASH FLOWS PER SHARE (LFCP) 281
29: TIMIES-INTEREST-EARNED RATIOS (LFCP) 283
30: PROFITS WITH A FIXED UTILITY RATE (NORM.) 297
Abstract of Dissertation Presented to the
Graduate Council of the University of Florida in Partial
Fulfillment of the Requirements for the Degree of
Doctor of Philosophy
A COMPUTER SIMULATION ANALYSIS OF
ALTERNATIVE METHODS OF ACCOUNTING
FOR UTILITY CONSTRUCTION WORK IN PROGRESS
Gordon Douglas Quick
Chairman: Eugene F. Brigham
Major Department: Management
Are the quality and stability of utilities' earnings significantly
influenced by the choice of the construction accounting method? Are
cash flows and coverage ratios affected by construction accounting?
These and many related questions are being raised with increasing fre-
quency. The purpose of this investigation was to examine the impact of
several construction accounting methods in order to provide some of the
information necessary to begin answering these and related questions.
Hopefully, this information will help utilities and their regulators
choose the "best" construction accounting method.
Under regulatory theory, service rates should be set at levels
which allow investors a fair rate of return on the investment needed to
provide that service. The theory also states that if a part of the plant
is not being used for the benefit of current customers, i.e., it is not
currently "used and useful," then current customers should not be re-
quired to provide a return on this unused plant. Since plant under con-
struction is not directly serving customers, it does not meet the used-
and-useful rule. Yet, investors have funds tied up in construction pro-
grams, and both fairness and the need to acquire capital require that
such funds earn competitive rates of return.
Several construction accounting methods were investigated, including
two basic methods which represent the "polar cases" between which the re-
sults of the other cases lie. Under one of these basic methods (referred
to as method RB), construction work in progress (CWIP) is included in the
rate base and earns a rate of return equal to that earned on used and use-
ful assets. Under the other basic method (referred to as method IDC),
CWIP is excluded from the rate base, and to provide a fair return on CWIP,
an imputed return (IDC) is calculated and reported as income. IDC income
is not cash income, but a credit added to the cost of the asset and even-
tually recovered over the useful life of the asset.
Accounting practices almost always involve compromises; rarely is
one method superior in all respects. Method IDC is clearly superior on
theoretical grounds, but may have serious practical drawbacks. Method
RB is not as theoretically appealing, but it has many practical advantages.
To illustrate the long-run effects of the alternatives, a computer
simulation model was developed and run for a wide range of assumptions
and conditions. Some of the significant findings include the following:
1. In theory, method IDC avoids the need for current
customers to pay the capital costs of assets under
construction. Given real-world conditions of rela-
tively steady growth, this result may not hold in
2. Method RB minimizes fluctuations in utility rates
and, thus, may provide customers with the best informa-
tion about future price trends.
3. Method RB has no IDC earnings and, thus, none of the
undesirable implications of high IDC to earnings
and IDC to dividends ratios.
4. Method RB produces higher cash flows and cash flows
5. Method RB produces higher coverage ratios.
Several more general results are:
1. The more discontinuous the construction program, the
greater is the fluctuation in all the variables for
all the methods.
2. With a discontinuous construction program, the
capitalization of IDC increases the fluctuation in
all the variables such that the higher the IDC rate,
the greater the fluctuation.
3. With respect to the differences found: a) the higher
the growth rate, the greater the differences and b)
the longer the average construction period, the
greater the differences.
4. The results are more significant for some firms than
for others. The higher the IDC to earnings ratio,
the greater the differences are likely to be. For
some firms, the differences will be quite substan-
Is the quality of earnings of many utility companies comparable to
the quality of earnings of nonregulated firms? Are utility companies'
earnings subject to instability because of construction accounting? Are
the cash flows and coverage ratios of the utilities affected by con-
struction accounting? These are a sample of the questions which are
being raised by financial analysts, academicians, regulators, and others
concerning the construction account-ing method currently used by most
utilities. Although this method is generally regarded as the "theoret-
ically correct" approach to providing a return on funds invested in con-
struction, these questions are being raised with increasing frequency.
While the investment community, utility managements, regulators, and
others each have their own feelings regarding these questions, definite
answers cannot be found. Furthermore, little has been done to evaluate
alternatives to the traditional construction accounting method.
Fortunately, much of the criticism has been followed by the sugges-
tion of alternative methods which may eliminate, or at least alleviate,
some of these criticisms. Much of the attention has been directed to-
ward one alternative which represents a sharp departure from the tradi-
tional procedures for accounting for construction. In theory, the meth-
od employed by most utilities and this often-mentioned alternative are
the two basic approaches to providing a return on funds invested in con-
struction. In practice, they represent two extremes along a continuum
and, consequently, a number of alternative procedures have been sug-
gested which lie between the two extremes. These intermediate alterna-
tives are either derived from one of the two basic approaches or repre-
sent a combination of them.
Objective of the Study
The objective of this study is to examine alternative construction
accounting methods for providing a return on public utility construction
work in progress (CHIP). In light of the current criticisms of the con-
struction accounting method most firms now use, four questions are posed
which suggest a format for exploring these alternatives:
1. Do the alternatives result in differences with respect
to certain variables?
2. If they exist, are the differences significant enough
to be of concern?
3. Are these differences such that the cost of capital
could be affected by the choice of the construction
4. What impact will the complexities of utility regula-
tion and the economic environment have upon the basic
This study is aimed primarily at answering the first two questions. To
definitively answer the third would require an extensive empirical in-
vestigation; consequently, only a tentative answer to this question is
offered. The final question could also be the subject of a separate in-
vestigation, and we merely offer some insights into this area based on
the understanding of the alternatives developed in this study.
Although this is primarily a theoretical study rather than an em-
pirical investigation, our primary concern is with the real-world impli-
cations of the alternatives for public utilities. Thus, a secondary ob-
jective of this study is to provide enough information to allow the
reader to draw his own conclusions about the alternatives, depending on
his environmental expectations at any point in time.
Scope of Interest
This study will examine the factors which cause differences between
firms using different construction accounting treatments and the key
variables which create these differences. Regulators, customers, and
management will be concerned with these differences because they may in-
fluence the firm's cost of capital. Investors will be interested in the
differences between firms because of the potential impact on the riski-
ness of the firm. The differences will be of concern to academicians
inasmuch as different accounting treatments may alter the homogeneity of
subgroups used for study.
The monopolistic nature of the utility industry implies the ab-
sence of the competitive forces that tend to eliminate less than optimal
construction accounting methods. Thus, regulatory bodies must thorough-
ly understand the impact of construction accounting practices on the
firm. This is necessary to assure that investors are fairly compensated
if risk differentials exist and that consumers are not required to pay
for operating or financial inefficiencies.
The customer's desires are adequately expressed in terms of two ob-
jectives: 1) Other things being equal, the customer would like service
provided at the lowest cost and 2) he would like rates established on an
equitable basis. Unfortunately, these may be conflicting goals, with
one being attainable only by partially sacrificing the other.
Utility managements do not want to adopt practices which may create
risk differentials and which may raise their cost of capital. This
would place some managements at a disadvantage in the capital markets.
Capital-intensive utility companies which require large sums for capital
investment could experience serious problems in attracting capital.
Investors will be interested in whether significant differences
exist between the construction accounting methods which might alter the
riskiness of the firm. Such a risk differential would justify a differ-
ence in the rate of return allowed on equity funds. Hopefully, investors
will better understand why differences occur, and this will assist them
in making the proper assessment of the importance of these differences.
Academicians are often in need of homogeneous samples for testing
hypotheses about cost of capital and other variables. Different con-
struction accounting methods may result in significant differences be-
tween the firms which would alter the comparability of the firms.
Plan of the Study
Chapter 2 is designed to provide the reader with a brief but fairly
complete background to the problem of providing a return on CHIP. This
is accomplished by examining the return on CWIP, briefly introducing and
comparing the two basic alternatives, and by exploring some of the prob-
lems which have provided the impetus for this investigation.
In Chapter 3 the five alternatives are introduced and then explain-
ed in detail using a highly simplified model. By reviewing the results
of this simplistic model and a slightly more realistic model, we see
that these alternatives will result in differences with respect to cer-
tain variables. However, the simple models are not detailed enough for
us to assess the magnitude of these differences. Thus, Chapter 4 des-
cribes a computer simulation model which provides enough realism to al-
low these magnitudes to be considered. This detailed explanation is
necessary, since the model is the tool used to evaluate the alternatives,
and the credibility of the model determines the credibility of the
Chapter 5 presents a detailed explanation of the results of the
model and explains the causative factors affecting the variables. In
addition to the basic results, several other cases are examined to pro-
vide a broader base from which conclusions may be drawn. The final chap-
ter contains the summary and conclusions of Chapter 5, along with the im-
plications of the findings.
PROVIDING A RETURN ON CONSTRUCTION WORK IN PROGRESS
Providing a return on the funds invested in construction is nothing
new in public utilities. Since the early 1900's, regulators in some
jurisdictions have provided some form of compensation to investors for
funds allocated to non-revenue producing construction activities. Why,
then, has construction accounting recently become of interest to all
those concerned with public utilities? Two factors have combined to
create this current interest. However, before we examine these two
factors, it is necessary to present some general background information
concerning the return on construction work in progress (CWIP).
The Return on Construction Work in Progress
A brief discussion covering four topics will provide the necessary
background concerning the return on CWIP. We first discuss the rationale
for providing a return on CWIP and then describe it as consisting of two
components, relating to the sources of capital used to finance construe-
tion. These components are then used in exploring the foundation of a
return in regulatory history. Finally, we indicate why a separate pro-
vision for the return on CWIP is necessary.
Is a Return on Funds Allocated to Construction Warranted?
The primary justification for a return on CWIP is founded in basic
economic theory. Throughout the construction period, funds are invested
in assets which produce no current revenues. However, the individuals
who supplied the funds must be compensated. There is an opportunity
cost long recognized in economics which is based on the fact that these
funds could have been otherwise invested in current revenue-producing
activities. This foregone return should be recognized as a legitimate
cost of converting capital into a productive asset. This cost is as
valid as if funds were actually paid out to acquire the asset.
A commercial lender, such as a bank or savings and loan, that makes
a construction loan charges interest during the construction period. The
funds which utilities use for construction must be provided by someone,
whether it be equity holders, bond holders, or banks. These investors
must be compensated for providing these funds. Without this compensation
these sources of funds will seek other revenue-producing investments.
Having established that compensation must be provided, it is useful
to think of the return as consisting of two components, debt and equity.
The following discussion of these components further supports the conten-
tion that compensation must be provided.
The Components of the Return on Funds Invested in CWlP
Debt Component. There is little controversy surrounding the debt
component of the return on CWIP. This is because out-of-pocket cash ex-
penditures result from the contractual interest on borrowed funds anid,
perhaps, the dividends on preferred stock. These are real obligations
which the company must meet as a result of some of the sources of capital
used during construction.
Equity Component. Some feel that a real distinction should be made
between the debt and equity components.' With debt there is a contrac-
tual obligation, while with equity no such contractual obligation exists.
The equity component is merely an "imputed" return to common equity funds
and some have questioned the legitimacy of "creating" such a return.
This practice is not as questionable as it may first appear. Rare-
ly, if ever, can debt be obtained without some form of collateral to
guarantee repayment of the loan. In a going concern, the "collateral"
is provided by the equity holders. In the event of default or business
failure, the bond holders are paid in full before the equity holders are
even considered. In addition to providing some of the capital, equity
holders bear nearly all of the risks of construction. Without the equity
holders to assume the risks, debt capital would not be available. It
seems obvious that equity holders must also receive fair compensation.
The preceding discussions have illustrated the need for a return on
CWIP. Initially, regulatory agencies allowed only a return on the debt
component. The following discussion indicates that some provision for
the equity component has been allowed by most regulators only in the last
The Foundation of a Return in Regulatory History2
In 1909 uniform accounting systems for electric and gas utilities
IFor example, see John H. Bickley,"CnostuinCrteIcm
and Profit?", Public Utilities Fortnightly, Vol. 88, No. 2 (July 22,
1971), pp. 32-33.
2Litke, Arthur L., "Allowance for Funds Used During Construction,"
Public Utilities Fortnightly, Vol. 90, No. 7 (September 28, 1972), pp.
were adopted in New York and Wisconsin. Both systems provided for capi-
talizing interest paid on debt attributable to CWIP. However, until
1914 no provision was made for the equity component of CWIP. At that
time the Interstate Commerce Commission first released the Uniform Sys-
tem of Accounts for Steam Railroads. In 1922 the state regulatory com-
missions, through the National Association of Railroad and Utilities
Commissioners and in cooperation with the utilities, developed a classi-
fication of accounts which recognized an allowance for both debt and
equity capital devoted to construction. Also in 1922, the Federal Power
Commission (FPC) prescribed an accounting system which provided for an
allowance for both debt and equity components. Similar provisions were
made in the system of accounts adopted in 1936 by the FPC for electric
utilities and in 1939 for gas companies.
The relatively late acceptance of a return on CWIP is perhaps part
of the reason why a separate provision for the return is necessary. The
following discussion will briefly discuss why funds invested in CWIP are
not compensated directly, thus forcing regulators to use another method
to compensate investors for funds allocated to CWIP.
Why a Separate Provision for the Return on CWIP is Necessary
Because of the monopolistic nature of the industry, utilities are
regulated by public authorities. These regulators govern the prices
utilities may charge for their services, thereby regulating the profit,
within limits, which utilities may earn. Historically, regulators have
used the net amount of plant currently in service as an indicator of the
funds "prudently" invested. These "used and useful" assets are desig-
nated as the base upon which investors are entitled to a return. The
regulators also specify a rate of return which the utility company may
earn on the base. The "fair rate of return" and the "rate base" are
then used to determine the rates utility companies may charge and, con-
sequently, the profits they earn. By definition, "used and useful"
plant excludes plant under construction. Therefore, the funds devoted
to construction are not being offered any return. If a return is to be
provided on CWIP, some other allowance must be made.
The Increasing Interest in Construction Accounting
Having provided the reader with some background concerning the re-
turn on CWIP, we can now explore the reasons construction accounting has
generated so much interest in the last several years. There are essen-
tially two reasons: 1) The return on construction has become a signifi-
cant component of the total earnings of many utilities and 2) the method
currently used by most utilities has apparently created some problems.
Each factor considered individually would not present a problem for
utilities. The problems resulting from the construction accounting meth-
od used by most firms have always existed. However, as the size of the
return on construction increases relative to the total earnings of the
firm, the resulting problems become increasingly significant. As the
following discussion reveals, the return on CWIP has become an increas-
ingly large proportion of the total earnings of utilities.
The Increasing Significance of the Return
on CWIP Relative to Total Earnings
The ratio of the return on CWIP to total earnings is directly re-
lated to the bases upon which these earnings are computed; that is, the
ratio CWIP to total assets. If CWIP is large relative to total assets,
then the return on CWIP is large relative to the total earnings of the
firm. While aggregate statistics on CWIP as a percentage of total as-
sets are not currently available, there is every reason to believe that
the percentage has been increasing in recent years, and that this in-
crease will continue in years ahead.
The size of CWIP relative to total investment depends primarily
upon two factors: 1) the annual investment in construction relative to
total assets and 2) the average length of the construction period. Sup-
pose annual construction expenditures are 5 percent of net assets, and
the construction period is one year. In this case CWIP will represent
about 5 percent of total assets and the income from CWIP will be about
5 percent of total income.
Now suppose increased demand, stepped up environmentally-related
expenditures, and inflation combine to double investment as a percent of
net assets, to 10 percent. If the construction period remains constant,
then CWIP as a percent of total assets will double, as will CWIP income.
However, if the construction period also doubles, to two years, the com-
bined effect will be a four-fold increase in CWIP, and income from CWIP
will increase from 5 to 20 percent of total income. Thus, CWIP and its
related income increase exponentially with increases in construction ex-
penditures and in the length of the construction period.
Construction expenditures have increased dramatically in recent
years; for the electric companies, construction expenditures as a percent
of net assets rose from about 7 percent in 1962 to almost 14 percent in
1972.3 In view of the fact that the industry's reserve ratio was lower
3These percentages are based on 104 electric companies listed on
the S&P Compustat tapes.
at the end than at the beginning of the period, the forecast of much
higher environmentally-related expenditures, plant modifications made
necessary by the energy crisis, and so on, it seems safe to assume that
construction expenditures as a percent of net assets will remain high
if not actually continue to rise.
We have no aggregate statistics on the length of the construction
period, but from all reports this period has risen dramatically and will
continue to rise. The construction period for distribution facilities
has increased significantly, largely because of a trend toward under-
ground lines. Nuclear plants represent an increasing percentage of
generation expenditures, and these plants have very long construction
periods. Fossil fuel plants today are larger than in the past, and
larger plants take longer to build. Both increased plant sizes and en-
vironmental considerations are causing plants to be located further from
power users, and this factor and the trend toward interties has caused
an increase in transmission construction. Finally, and probably most im-
portant of all, is the shift in construction expenditures from distribu-
tion to generation:'
Percentage of construction
budget devoted to: 1962 1971
Generation 35% 56%
Transmission 24 21
Distribution 41 23
Since generating plants take much longer to construct than transmission
lines and distribution facilities, this shift in the construction budget,
Source: Statistical Year Book, Edison Electric Institute, 1972,
even without the changes in the individual construction categories,
would have lengthened the construction period appreciably. When all
factors are combined, it would be easy to visualize a doubling or even
tripling of the average construction period.
As noted earlier, direct evidence on the trend in the ratio of CWIP
to total assets is not currently available. However, the evidence pre-
sented indicates that the primary determinants of this ratio, construc-
tion expenditures as a percentage of total assets and the length of the
construction period, have both increased substantially. As a result, it
seems clear that the ratia CWIP to total assets must have risen dramat-
ically. And as the CWIP/assets ratio rises, so must the ratio of CWIP
income to total income.
If CWIP is excluded from the rate base, and IDCs is computed and
used as compensation for funds invested in CWIP, then the ratio of IDC
to operating income provides a direct measure of the increasing impor-
tance of construction accounting. The average ratio of IDC to earnings
for the electric companies for the period 1964 to 1972 is shown in Table
1. These figures also suggest that the importance of construction ac-
counting has increased markedly in recent years.
These facts have demonstrated that the return on CWIP has become an
increasingly significant component of the total earnings of utilities.
If the construction accounting method used by most utilities produced
satisfactory results, there would still be little interest in construc-
tion accounting. However, this widely-used method has some shortcomings,
and as the return on CHIP becomes more important relative to the total
A formal explanation of IDC is provided later, but at this point
it is sufficient to say that IDC is the return on CWIP.
RATIO OF IDC TO EARNINGS BASED ON 114 ELECTRIC COMPANIES
Source: S&P Compustat tapes.
earnings of the firm, the shortcomings of this method become much more
noticeable. But before we can explore these problems, it is necessary to
explain the construction accounting procedure used by most firms. Thus,
the next section provides a brief introduction and comparison of the
basic methods for providing a return on CWIP.
Providing a Return on CWIP
There are three basic ways a return on CWIP may be recognized. The
first method would be to allow a rate of return higher than would nor-
mally be allowed, after the plant has been placed in service. The second
way would be to allow a return currently, by including CWIP in the rate
base as the investment in CWIP is made. The third method would be to al-
low the return to be capitalized during the construction period, with it
being included as part of plant cost when computing allowable deprecia-
tion and as part of the rate base in future rate determinations.
The first method would require the determination of an increment to
the normally allowed rate of return. This higher return would be main-
tained until adequate to provide a return on funds invested during con-
struction. The objective is simply to allow a higher return in the
future; the increment would presumably be equivalent to the return fore-
gone during the construction period, adjusted for the time value of
money. While this is a feasible alternative, it has rarely been sug-
gested for use by regulated companies. As a result, a detailed proposal
of this nature is not available. Thus, this method is excluded from
further consideration in this study.
The second method has been advocated quite frequently. By including
the costs of construction in the current rate base, investors immediately
receive full fair compensation on funds allocated to construction The
rate of return provided on assets "used and useful" is also provided on
The third method is the one most commonly used today. This method
provides for an immediate return, in the form of a credit, which may or
may not represent a full fair return to investors. Whether or not it
does depends on the rate at which the return is to be calculated. This
return is also a part of the cost of the asset, which is eventually re-
covered through depreciation.
Thus, we are left with two basic procedures for providing a return
on CWIP. These two procedures are briefly explained and compared in the
discussions which follow.
Method One: Include CWIP in the Rate Base
This method is the easier of the two remaining basic methods to
Tiasueof course, that the currently allowed rate of return
understand. CWIP is included in the "rate base" which is used by regu-
lators to determine the rates necessary to provide a "fair rate of re-
turn" to investors. The same rate of return allowed on "used and use-
ful" assets is also allowed on CWIP. To the extent that the fair rate
of return is adequate, investors are fully compensated for their invest-
ment in CWIP. Under this method, current customers bear the capital
costs of construction. Under current regulatory interpretations, this
could be justified if CWIP could be defined as "used and useful." De-
fining CWIP as "used and useful" would be possible if the construction
were: 1) to lower operating costs, 2) to provide assurance of continued
service, or 3) to meet expanded needs of old customers. Making the cur-
rent rate payer bear the costs of capital would appear to be unjustified
if the construction were primarily to meet the needs of new customers.
It can be argued that current rate payers should not have to pay for as-
sets which will be "used and useful" to others in the future. This is
often countered by noting that construction is seldom for entry into
completely new areas previously without service. New construction is
mainly to replace or upgrade service in existing areas. Then, it is ar-
gued, there is little distinction between present and future customers.
This argument must assume that people do not move, which is clearly an
invalid assumption. However, a short construction period will minimize
the problem since the return would be small.
The question of shifting the burden is an important consideration
where CWIP is included in the rate base. To avoid this problem, the
capital costs of construction may be capitalized so that payment of these
costs is deferred to future periods. This is the essence of the other
basic approach to providing a return on CWIP.
Method Two: Capitalize the Return on CWIP
As an alternative to the previous method, CWIP could be segregated
and excluded from the rate base. In this way current customers will be
charged only for the investment in plant currently in service. To meet
the obligation to investors, a return could be calculated at some rate
(probably the "fair rate") and this amount would be capitalized. This
return is commonly referred to under this method as "interest during con-
struction," or IDC.7 The essence of this approach is as follows. First,
the amount of funds invested in CWlP must be determined. Then the re-
turn, or IDC, is computed as the product of CWIP and the rate of return
allowed on CWIP. This return is then credited to income on the income
statement o In addition, at the end of the period IDC is added to the
cost of construction (CWIP). Therefore, when construction is completed
and the asset is included in the rate base, the total cost of the asset
exceeds the actual dollar outlay by the amount of IDC accumulated over
the construction period. The total cost of the asset (including IDC) is
finally depreciated over its useful life. Obviously, the total cost of
the asset, or book value, is greater under this method than when CWIP
goes directly into the rate base.g
'The term "interest" in IDC is unfortunate and has created numerous
misconceptions about IDC. A more descriptively appropriate term is "al-
lowance for funds used during construction," or AFDC. "Interest" implies
that the construction program is debt financed; in reality, it is finan-
ced by the same debt-equity mix as the "used and useful" plant.
BTwo methods have been employed in the past to account for IDC. One
is to add (credit) IDC to income as mentioned above. The other is to use
the interest credit to reduce the interest expense. The net result, in
terms of earnings, is the same,
SThe cost of the asset, including IDC, is presumably close to what
it would be if the asset was purchased on a "turn-key" basis. Here the
A Comparison of the Two Basic Methods
In comparing these two methods we will examine the effect of each
on three key variables: 1) the size of the rate base, 2) customer
charges (revenues), and 3) investors' returns. This comparison is pro-
vided to begin to familiarize the reader with the basic construction ac-
counting methods which are fully developed in the next chapter.
The rate base. Given the same initial plant, then during the con-
struction period the rate base will be higher for the method which in-
cludes CWIP in the rate base (hereafter the rate base method is referred
to as method RB). This is because this method includes CWIP in the rate
base as soon as the funds are used in construction. Conversely, where
IDC is capitalized (hereafter referred to as method IDC), no costs re-
lating to construction are included in the rate base during the construc-
tion period. Therefore, for method IDC, the rate base will be lower by
the amount of CWIP throughout the construction period. The situation is
reversed once construction is completed and the asset is placed in ser-
vice. For method RB, there is no change in the rate base. But for meth-
od IDC, the total amount of CWIP, which includes accumulated IDC, is now
included in the rate base. Therefore, for the period following construc-
tion and until the asset is fully depreciated, the rate base will be
larger for method IDC. The difference will always be the net amount of
contractor would add a profit, as well as a cost of debt used during the
construction period, that would be roughly equivalent to IDC.
Customer charges (revenues). The cost to the customer is deter-
mined, in part, by multiplying the rate base by the allowed rate of re-
turn. Since the rate base is higher when CWIP is included in it, cus-
tomer charges will also be higher during the construction period for
method RB. Once the asset goes "on line," customer charges will be
higher for method IDC, in direct proportion to the amount of IDC includ-
ed in CWIP. From this we see that current customers bear part of the as-
set cost under method RB, while future customers pay the full cost under
Investors' returns. Investors will be fairly compensated by either
method as long as these methods are handled properly.10 Using method
RB, investors receive an immediate return at a rate identical to the fair
rate of return on funds invested in "used and useful" assets. Under
method IDC, investors receive a deferred return, or credit, at the rate
specified for calculating IDC. If this rate is equal to the fair rate
of return, the compensation under both methods is the same.
Table 2 presents a summary of the comparison and is sufficient to
illustrate that significant differences exist between the two basic meth-
ods of providing a return on CWIP. We shall now explore the usage of
these methods by regulated and nonregulated firms.
Use of the Two Basic Methods
This section explores the available data on the actual usage of the
As long as these methods are handled properly" is a critical
qualification which may not be met in the real world. In later chapters
we discuss some of the implications when this qualification does not
Include CWIP in the
rate base during
construction Yes No
Capital costs borne by Present Those who benefit.
customers from the asset
The rate of return on
CWIP is equal to Rate of return The IDC rate
on the rate base
basic methods of providing a return on CWIP by regulated firms. Unfor-
tunately, complete and timely data are not readily available. However,
these data provide some idea of the relative use of the two methods.
A study by Goodbody and Company in 1964 indicated that approximately
90 percent of the gas and electric utility companies in this country use
method IDC." Another study by Haskins and Sells found similar results:
Of a group of 138 gas and electric companies with revenues in excess of
$20 million each, 124 companies, or 89.9 percent, capitalized interest
during construction in 1963 and 1964.' After a detailed examination of
reports filed by 212 electric utilities with the Federal Power Comrmission
in 1966, another writer reported that 171, or 80.7 percent, of these
companies capitalized IDC, while 41 did not.' In 1968, of these same
Sayad, Homer E., "An Accountant Looks at Capitalized Interest,"
Haskins and Sells--Selected Papers, 1966, p. 122.
M3orris, Everett L., "Capitalization of Interest on Construction:
Time for Reappraisal?", Public Utilities Fortnightly, Vol. 87, No. 5
(March 4, 1971), p. 24.
SUMMARY COMPARISON OF METHODS RB AND IDC
212 companies, the number which capitalized IDC had increased to 174, or
These studies indicate that the large majority of utility companies
follow method IDC. None of the studies reported mentioned how many
companies followed method RB; most of those which did not capitalize IDC
apparently did not make any provision for a return on CWIP. Presumably
most of these are smaller companies which do not have substantial con-
struction programs. However, a few may have used method RB.
It is interesting to contrast these findings with similar statistics
on industrial companies. Of a sample of 77 companies from Fortune's 500,
61 responded and not one reported any capitalized interest in 1968."
The accounting procedures applicable to nonregulated firms do not permit
an investment to generate a reported profit, so this finding is not sur-
prising. Industrial companies can adjust prices, hence rates of return,
within the limits of the free market, to compensate investors for funds
employed during construction. Because of the restrictions imposed on
regulated firms, they cannot adjust prices and rates of return to recoup
the return foregone during construction. Consequently, regulated firms
are permitted to capitalize the return and report it as profit.
Problems Encountered in Providing
a Return on CWIP
Among utility companies, IDC is the most widely used method for pro-
viding a return on CWIP. Unfortunately, the use of IDC is beset with
several problems, some concerning implementation and others relating to
the firm's performance. There are three problems in implementing method
IDC which are best expressed as questions: 1) What is the proper rate
of return on CWIP?, 2) What is the proper base for computing IDC?, and
3) Are the goals of financial reporting helped or hindered by the use of
method IDC? While these problems have arisen primarily in connection
with the use of IDC, when appropriate we also briefly indicate how meth-
od RB compares with regard to each issue.
What is the Proper Rate of Return on CWIP?
As a practical matter, determining a specific rate of return on
CWIP is difficult enough even if the theoretical rate is agreed upon, and
although the IDC method has been in use for many years, no agreement has
been reached on what the theoretical rate of return on CWIP should be.
This issue must be resolved before a comprehensive, final evaluation of
the methods can be achieved.
Arguments have been made in favor of a rate as low as the after-tax
cost of short-term debt up to a rate as high as the current (or marginal)
weighted cost of capital. Since all funds have a cost, the lower limit
is set by the least expensive source of funds. This lower limit can gen-
erally be regarded as the after-tax cost of short-term debt. Conversely),
equity is usually the most expensive source of funds, and equity is fully
taxable. Since common equity holders assume most of the risks of con-
struction, they must also be compensated. In general, most people argue
for a rate which is a weighted average of the after-tax cost of debt and
the equity rate.Is
Method RB does not require the specification of a separate rate of
"Discussion of a more specific rate of return on CWIP is deferred
to the next chapter.
return on CWIP since CWIP is included in the rate base and earns the
same rate as funds invested in "used and useful" plant.
What is the Proper Base for Computing IDC?
This issue is related to the previous discussion in that they both
affect the compensation investors receive on CWIP. Logically, it would
appear that all construction investment not included in the rate base
should be included in the base for computing IDC. However, this is not
always the case. IDC is generally computed only on projects which ex-
tend over a specified period of time (e.g. a month) or exceed a certain
dollar amount. If a large number of projects do not meet these specifi-
cations, a significant amount of CWIP will not earn a return.
Are the Goals of Financial Reporting Helped or Hindered by
the Use of Method IDC?
The accounting procedures currently specified by regulatory agencies
fail to separate the impact of construction activities from the operating
results of the company.' This makes it difficult to assess the opera-
ting results of the company apart from its construction activities. The
separation of utility and construction results is even less clear under
method RB than under method IDC.
Another accounting principle, matching costs and revenues, is ac-
complished through the use of method IDC but not RB. Under the IDC meth-
od, the capital costs of construction (IDC) are deferred until the plant
is placed in service and revenues to offset these costs are being gener-
Frazer, R. E. and R. C. Ranson, "Is Interest During Construction
'Funny Money'?", Public Utilities Fortnightly, Vol. 90, No. 13 (December
21, 1972), pp. 23-27.
ated. Method RB would not satisfy this requirement, since the capital
costs of construction are paid out of current revenues and are not de-
rived from the plant under construction.
Concerning the performance of utility companies, four areas have
been mentioned where problems might exist: 1) cash flows, 2) times-
Lnterest-earned ratios, 3) earnings instability, and 4) the quality of
earnings. Although the implementation problems are quite important,
they are not as amenable to objective evaluation as are utility perfor-
mance measures. Thus, this study is more concerned with the problem
areas which can be explored objectively. Although these problems have
also arisen primarily from the use of method IDC, we again provide some
comments, where appropriate, about the impact of method RB on these prob-
Reduced Cash Flows
During periods of large, sustained construction activity, a cash
flow problem may develop. This is caused by a number of related factors,
the primary one being that IDC is non-cash income. When dividends are
paid based on total earnings (including IDC credits), the size of the
dividend and thus the cash outflow are larger because of IDC. There is
not a corresponding inflow of cash from IDC to offset the outflow. In
addition, construction itself results in increased cash expenditures.
Finally, increased interest payments result in reduced taxes, which in
turn cause operating income to be overstated." To avoid this over-
statement, utility rates might be reduced, further reducing cash flow.
"A more thorough discussion of this syndrome is provided in the
As a consequence of these factors, cash flow is reduced more by
construction programs if the IDC method is used. When projects go "on
line," full revenues are generated (given adequate demand) and cash flow
is increased. Under method RB, operating revenues are provided on both
"used and useful" plant and CWIP, and the result is that the cash flow
problem is eased.
Lower Times-Interes t-Earned Ratios
Since IDC earnings are "imputed" credits which do not result in
cash inflows, they are frequently excluded when computing coverage ratios
on bond interest. If these ratios become too low, the utility's bonds
may be downgraded, and this would raise the rates for future bond issues,
thus raising the overall cost of capital to the company.
The use of method RB will not result in a similar interest coverage
problem. Provided rates are set at the appropriate levels, full cash
revenues are generated on CWIP investment causing no reduction in fixed
charge coverage ratios.'
Financial analysts are particularly concerned over the potential in-
stability of earnings when IDC is capitalized. Earnings instability may
occur because IDC credits boost reported earnings as long as CWIP exists,
'However, lags or other problems may prevent rates from being set
at the appropriate levels resulting in a fixed charge coverage problem
similar to that encountered when using method IDC.
For a good discussion of impact of earnings instability on the
firm, see Richard Schramm and Roger Sherman, "Profit Risk Management and
the Theory of the Firm," Southern Economic Journal, Vol. 40, No. 3 (Jan-
uary 1974), pp. 354-357.
but when the project is completed and IDC is no longer being earned,
earnings fall until demand is sufficient to provide full use of the new
plant. If demand builds up slowly, as often occurs, sufficient revenues
will not be generated immediately to offset the loss of IDC credits.
Earnings could be maintained through a rate increase, but this is un-
likely if demand is expected to rise enough to provide adequate returns
in the future, and in any event, lags will be encountered. The problem
can become quite severe if construction is discontinuous or "1umpy." On
the other hand, this problem does not exist if new construction projects,
and thus IDC credits, arise continually.
Method RB may result in a different earnings instability problem.
Under this method, CWIP is included immediately in the rate base. With
significant amounts of investment in construction, the rate base will in-
crease to the point where the earnings from current operations are no
longer sufficient to provide a fair return on the rate base (including
CWIP). To generate a fair return from the existing operating base,
customer rates may have to be increased. After construction is complete
and the plant goes "on line," earnings will increase since the operating
base is larger and demand is growing. Since the rate base does not in-
crease when the plant goes "on line," the increased revenues from grow-
ing demand provide a return in excess of the fair return. This necessi-
tates a reduction in customer rates. This procedure will produce a rise
in earnings when demand is constant and reduction in earnings when de-
mand is expanding. Again, this potential instability problem will be
minimal if the construction program ib stable and continuous.
Whether or not an instability problem will occur depends upon a
number of other factors, including the pattern of growth in demand and
economies of scale. It is quite possible that the pattern of demand
grDwth COuld differ from firm to firm. As a result, one firm could have
an earnings instability problem while another, using the same construc-
tion accounting method, may not. For example, if demand grew at a pace
identical to the growth in investment, revenues would grow at a rate
sufficient to provide a fairly constant return on investment, using
method RB. The same could be true depending on the economies of scale
the firm achieves during expansion.
Reduced quality of Earnings
Some have argued that the quality of earnings when IDC is capital-
ized is inferior to that of companies with no IDC credits. The reasons
for this assertion include several of the issues previously discussed,
such as cash flow or earnings instability. But the real criticism is
directed primarily at the fact that IDC earnings are merely "imputed"
credits. These credits are claims to future revenues and as such are
not as highly regarded by investors and analysts as are operating reve-
nues. There is no absolute guarantee, even with regulated industries,
that future revenues will be generated. The railroad industry provides
the best example of this fact.
Several statistics reflect the severity of this problem and indicate
why analysts are questioning the quality of the earnings of many utility
companies. Table 3 presents figures on privately owned electric utilities
that report to the Federal Power Commission (FPC). Between 1965 and 1971,
net income increased 49 percent ($2.6 billion to $3.8 billion). During
this same period, capitalized interest increased 863 percent ($93.7 mil-
lion to $808.8 million). In other words, 57 percent ($715.1 million) of
NET INCOME AND INTEREST CAPITALIZED FOR PRIVATE
OWNED ELECTRIC UTILITY INDUSTRY 1965-1971*
Reported % Net
Year Net Income Amount Income
1965 2,580.7 93.7 3.6
1966 2,749.1 127.5 4.6
1967 2,908.3 186.3 6.4
1968 2,995.5 274.7 9.2
1969 3,196.0 402.9 12.6
1970 3,407.5 588.4 17.3
1971 3,842.7 808.8 21.0
*Source: Based on E. L. Morris, "The Interest on Construction
Dilemma--A Proposed Solution," Investment Dealers'
Digest--Public Utility Survey, Section Two, Vol. 38
(August 29, 1972), p. 20.
the $1.262 billion increase in net income was attributable to capitalized
interest. While reported earnings grew at an annual compound rate of 7
percent, the rate was only 3.2 percent when capitalized interest is ex-
cluded.2o Looking at another set of data, this same result can be viewed
in another way. Table 1, presented earlier, listed the aggregate IDC/
earnings ratios of 114 companies listed on the Standard and Poor Compus-
tat tapes. Between 1964 and 1972 the ratio of IDC to earnings rose con-
tinuously from 4 percent to over 29 percent. Thus, in the aggregate,
Morris, Everett L., "The Interest on Construction Dilemma--A Pro-
posed Solution," Investment Dealers' Digest--Public Utility Survey, Sec-
tion Two, Vol. 38 (August 29, 1972), p. 20.
Based on Richard Walker, "The Capital Cost of Utility
Construction," Arthur Andersen Chronicle, Vol. 31, No.
4 (September 1971), p. 34.
capitalized IDC has become a very significant part of the utility indus-
try's total reported earnings.
The problem is even more severe for many individual companies.
Table 4 shows the percentage of earnings per share represented for 20
selected companies. The portion of earnings represented by IDC credits
ranges from 17 percent all the way up to 60 percent. And the situation
continues to worsen as indicated by the increase in the average IDC to
RELATIONSHIP OF INTEREST DURING CONSTRUCTION TO EARNINGS
AND DIVIDENDS FOR SELECTED UTILITY COMPANIES' IN 1970*
B C D E F
Dividends as a
3r Share % of Column B
as a % of
earnings ratio for the industry.
These non-cash earnings may also cause firms to be unable to meet
their preferred and common stock dividends. Although earnings per share
(EPS) are unaffected by IDC,2 more shares of stock will be outstanding
when construction is partially equity financed. In addition, part of the
earnings are now in the form of IDC credits. Companies may find that
they are unable to earn adequate cash income from operating revenues to
cover the payment of cash dividends on preferred and common stock. This
could seriously impair the ability of these companies to raise sufficient
capital for construction to satisfy public demand. The severity of this
problem is illustrated by columns E and F of Table 4 above.
For these reasons, investors may value IDC earnings, and the value
of the company in general, less highly than when IDC credits are not
present. To the extent that this is true, a company with large IDC
credits will have to offer a higher return on its common equity to have
securities as attractive as those of firms which do not capitalize IDC.
This results in a higher overall cost of capital and, thus, higher costs
Issues Relating to Method RB
There are also several issues relating to method RB which are impor-
tant and should be considered. One writer has argued that since we are
currently facing increasing costs in the production of utility services,
The impact of IDC on EPS is often misunderstood. For a thorough
discussion of this subject, see R. E. Frazer and R. C. Ranson, "Is
Interest During Construction 'Funny Money'?", Public Utilities Fortnight-
ly, Vol. 90, No. 13 (December 21, 1972), pp. 20-27.
.using method RB is preferable because it results in higher prices ini-
tially.2 These higher prices more accurately reflect the current trend
which may serve as a signal to consumers who can plan their expenditures
Method RB includes CWIP in the rate base as investments are made.
If these investments are being made continually throughout the construc-
tion period, which is usually the case, the rate base changes continual-
ly. A changing rate base may require changing utility rates if a fair
return on CWIP is to be assured. This would require a rapid rate adjust-
ment process. With the usual regulatory lag, the actual compensation
ultimately received by investors may be insufficient.
Finally, with method RB, rates may increase at a time when no new
services are being provided. Given the political nature of regulation,
raising rates as this method requires may be difficult. With the in-
creasing costs of construction and the lengthening construction periods
faced by most utilities, this problem could be severe.
This chapter has provided a brief but fairly complete introduction
to the issues involved in providing a return on CWIP. Two conclusions
were reached early in the discussion. First, it was shown that investors
must be compensated for funds allocated to CWIP. Second, current regula-
tory procedures require that the provision of a return on CWIP must be
considered separately. We also discussed the reasons construction ac-
-r01son, Charles E., "Interest Charged Construction: Economic,
Financial, and Regulatory Aspects," Public Utilities Fortnig~htly, Vol. 88,
No. 4 (August 19, 1971), pp. 31-32.
counting has generated so much interest in recent years. We then exam-
ined the two basic construction accounting methods which form the basis
for the evaluation presented in later chapters. The final section in-
troduced a number of problems which have evolved mainly from the use of
capitalized IDC. These problems are important in that they provide some
framework around which the alternatives are evaluated. Subsequent chap-
ters explore these issues more thoroughly and further define the specific
alternatives for providing a return on CWIP.
THE PROTOTYPE FIRMS
Chapter 2 introduced the two basic methods for providing a return on
construction work in progress (CWIP). By altering and combining these
two basic methods, several additional methods for providing a return on
CWIP are available. These alternatives require different accounting pro-
cedures which are explained in considerable detail using a simplified set
of assumptions. Once the reader has become familiar with the alterna-
tives, a more complex model is developed, and the results of both models
are examined. These limited results are sufficient to demonstrate that
the different alternatives will produce different results.
Depreciation policy also influences many of the variables which are
examined in the analysis of the firms. Thus, we also differentiate be-
tween straight line depreciation accounting and two procedures for hand-
ling accelerated depreciation. In effect, 15 prototype firms are estab-
lished which are analyzed in subsequent chapters.
Derivation of the Alternative Firms
A number of suggestions have been made which represent variations on
the two basic methods. Figure 1 illustrates the derivation of six alter-
native firms from the two basic methods and the mnemonic symbols that
will be used when referring to each of the firms.' The derivation of
these firms is presented below.
IIt seems most useful to think of a number of similar firms which
differ only in the method chosen for handling the return on CWIP. Thus,
rather than speak of alternative methods for providing a return on CWIP,
we shall speak of alternative firms which employ these methods.
Include CWIP in
the Rate Base
RB RB-IDC RB-%IDC-ITB
DERIVATION OF ALTERNATIVE FIRMS
Firm RB employs one of the two basic methods for providing a return
on CWIP. This firm includes the full amount of CHIP in the rate base
and, therefore, theoretically earns the same rate of return on CWIP as
is earned on used and useful assets. There is no capitalization of IDC,
and, thus, only the actual dollar investments are included in CWIP and,
eventually, the rate base.
Like Firm RB, Firm RB-IDC includes CWIP in the rate base and earns
a rate of return on CWIP equivalent to the rate of return on used and
useful assets. However, this firm also capitalizes IDC at the gross
rate and includes IDC "above the line" for rate making purposes. "Above
the line" means that IDc: is included as a part of operating income for
the purposes of determining the revenues required to provide an adequate
return on the rate base. For example, if net operating income of $1,000
were required to provide the allowed rate of return on the rate base and
the amount of IDC was $200, then only $800, rather than $1,000, would
have to be generated from the firm's customers. This does not provide a
double return on CWIP, since the income generated in the form of IDC re-
duces what must be generated through customer revenues.
Like the previous two firms, this firm includes CWIP in the rate
'The term "gross," as used in connection with the IDC rate, means
that IDC is capitalized at the before-tax cost of capital, however de-
termined. The "net" rate is the after-tax cost of capital.
base and earns a rate of return on CWILP equivalent to the rate of return
on used and useful assets. This firm also capitalizes IDC and, like
Firm RB-IDC, includes capitalized IDC above the line. However, this
firm capitalizes IDC at a very low rate, at a rate equal to the debt
component of the IDC rate. For example, if CWIP were financed equally
by debt and equity, and the interest rate was 8 percent, then the debt
component of the IDC rate would be 4 percent (8 percent x .5). This
figure would be the rate at which IDC is computed. But Firm RB-%IDC-ITB
also allocates the interest-tax benefit (ITB) which results from con-
struction. Table 5 illustrates the origin of the interest-tax benefit.
Part A of Table 5 shows the income statement of a firm which has no CWIP.
Assets are $1,000, financed equally by debt and equity, with debt requir-
ing 8 percent interest and equity earning a 12 percent return. Part B
shows the same firm with $100 of CWIP, financed equally by debt and
ILLUSTRATION OF THE INTEREST-TAX BENEFIT
Firm With Firm With
No Construction Construction
Revenues $410.00 $410.00
(including depreciation) $250.00 $250.00
Taxes 60.00 58.00
Interest-Tax Benefit* -0- 310.00 2.00 310.00
Net Operating Income $100.00 $100.00
Adjusted NOI $110.00
Interest Expense 40.00 44.00
Net Income 1-99122 A Afri
*Normally the interest-tax benefit is simply a debit to the income
tax expense. It is shown this way merely for emphasis.
equity. With increased debt due to construction, interest payments have
risen $4, and as the tax calculation reveals, taxes are $2 less. Since
nothing else changed in the example, the reduction in taxes must be
caused by construction. It can be argued that this tax saving, which is
caused by the interest payments on funds borrowed to finance construc-
tion, should be allocated (credited) to the CWIP account. Since the tax
saving results from the construction program, it should be used to reduce
the cost of construction. The effect of the procedures followed by this
firm is to capitalize IDC at an effective rate equivalent to the after-
tax cost of debt times the debt ratio, or 2 percent for the example used
above. The return on CWIP financed by debt is provided by IDC credits,
while the return on CWIP financed by equity is generated from operating
The first basic method and two variations on it have yielded three
alternative firms. These firms have significant differences which will
become more clear when each firm is fully illustrated using the single
asset model. The final three firms discussed in this section derive
from the procedure whereby IDC is capitalized as a means of providing a
return on CWIP.
Firm IDC includes no CWlP in the rate base, but computes IDC on CWIP
at the "gross"3 IDC rate. This imputed amount is the only means by which
a return is provided on CWIP. IDC also becomes a part of the cost of the
asset and is, therefore, included in CWIP, The most difficult and still
As defined in footnote 2.
unresolved question which arises for this and the next two firms is the
rate at which to capitalize IDC.
IDC rate. From the viewpoint of the stockholders, the IDC rate is
a crucial issue for Firms IDC, IDC-ITB-G, and IDC-ITB-N. In these firms
the amount of capitalized IDC is the only return investors receive.
Firms RB, RB-IDC, and RB-%IDC-ITB include CWIP in the rate base, and in
doing so earn a rate of return on CWlP equal to the rate of return earn-
ed on used and useful assets. For these firms the amount of IDC reduces
merely the amount of revenues which must be generated through customer
payments. Therefore, the IDC rate is crucial mainly to Firms IDC,
IDC-ITB-G, and IDC-ITB-N, for the rate at which IDC is computed deter-
nines the rate of return earned on CWIP. If IDC is computed at a rate
of 10 percent of CWIP, investors earn a rate of return of 10 percent on
The IDC rate is important for another reason. The proper rate is
necessary to accurately reflect the cost of the plant under construction.
IDC becomes part of the base upon which future rates are determined and
is part of the base to be depreciated and charged against future reve-
As indicated in Chapter 2, regulators have allowed a rate of return
on CWIP large enough to cover debt payments on funds borrowed for con-
struction and to provide some return on equity funds invested in con-
struction. However, they have not systematically specified, in theory,
what the rate of return on equity funds allocated to construction should
be. There are two systematic approaches to determining the IDC rate.
Both approaches have considerable merit, and both implicitly assume that
the construction program does not alter the riskiness of the entire in-
One approach to determining the IDC rate is to use the current, or
marginal, cost of capital. The current cost of capital is computed as
the weighted average cost of those funds used specifically for construc-
tion. The cost of debt sources is the effective interest rate, and the
cost of equity sources is the equity rate of return allowed on used and
useful assets. Assuming the firm's currently allowed rate of return pro-
vides the proper return to investors before construction, then the re-
turn for funds invested in CWIP should be set at the marginal cost of
those funds. The marginal rate may be above or below the currently al-
lowed rate of return, but in either case the return would be just suffi-
clent to meet the debt payments and provide a return to equity holders
at the specified rate.4 At some future time when the sources of capital
used to finance CWIP are included in the determination of the allowed
rate, the rate of return on CWIP should be set equal to the rate allowed
on used and useful assets. This approach requires the assumption that
the sources of capital used for construction can be segregated for the
purposes of determining the marginal cost of capital. For many firms
this would be a difficult and arbitrary process. Thus, a more practical
approach may be necessary.
Another approach is to set the IDC rate equal to the average cost
of capital for the firm, or the allowed rate of return on used and use-
ful assets. From a practical standpoint, this approach has considerable
Current conditions make it likely that the marginal cost will ex-
ceed the allowed rate of return for a utility in a net investment rate
base jurisdiction. This is because the embedded cost rates of senior
security capital may well be below those in current capital markets.
However, should rates begin falling, marginal costs could be below em-
merit. Regulatory bodies would not be burdened with specifying two
rates, one for used and useful assets and another for CWIP. The same
rate would apply to CWIP as well as used and useful assets. Also, firms
would not have to be able to identify the sources of capital used for
construction, thereby eliminating a possible source of conflict between
management and regulators. However, one disadvantage of this approach is
that the rate of return on CWIP would be incorrect when the current cost
of capital differed from the embedded cost. This situation could be rem-
edied by initiating rate hearings if the difference in cost rates became
There appears to be no justification for a rate which does not pro-
vide a return to equity holders equal to the rate of return earned on
used and useful assets. If regulators allow 12 percent for equity hold-
ers for investment in used and useful assets, it would be difficult to
justify a lower rate of return for that portion of CWIP financed by eq-
uity. Using the allowed rate of return on used and useful assets as the
IDC rate has two advantages. First, it requires no additional work for
regulators or management. Second, this amount is the return effectively
provided for firms which include CWIP in the rate base. Use of the al-
lowed rate provides consistency among the different methods for providing
a return on CWIP. However, firms should be able to petition for rate
changes any time the marginal cost of capital increases the embedded cost
For the illustrations in this chapter, the IDC rate is set equal to
the allowed rate of return on used and useful assets.s This equality
In the simplified examples used in this chapter, the costs of debt
and equity are constant, making the marginal cost and the embedded cost
makes the firms which only capitalize IDC comparable to the firms which
include CWIP in the rate base for the analysis provided later. The last
two firms, which are variations of Firm IDC, are discussed next.
Firm IDC-ITB-G follows the same procedures as Firm IDC but also al-
locates the ITB. The arguments brought out earlier regarding the ITB
apply here as well. It can be argued that since IDC is used to place the
burden of construction on those who benefit from it, any saving which re-
sults from construction should be passed on to those who bear the cost.
The allocation of the ITB makes the effective IDC rate lower for this
firm than for Firm IDC.
Firm IDC-ITB-N is similar to the previous firm except that the net
rate for capitalizing IDC is used. The ITB is automatically allocated
to CWIP since by using the net rate, it is never added into the construc-
tion cost. The tax saving is listed on the income statement under other
income as an income tax credit.
The two basic methods and the variations of them have yielded five
firms which employ different methods for providing a return on CWIP.
Table 6 summarizes the key differences between these firms. It is the
different formulas for the variables shown in this table which create the
differences in revenues, taxes, profits, and the other variables. These
formulas will be useful when the procedures followed by each firm are
presented in detail. But first it will be helpful to think of the firm
from the perspective described below.
SUMMARY OF THE KEY DIFFERENCES BETWEEN THE FIRMS
RB RB-IDC RB-%IDC-ITB IDC IDC-ITB-G'
Construction Work in
in the Rate Base Yes Yes Yes No No
Rate for Computing
Construction' IDC-1 IDC-2 IDC-3 IDC-2 IDC-2
Rate for Computing the
Benefits ITB-1 ITB-1 ITB-2 ITB-1 ITB-2
Revenue Requirements4 RR-1 RR-2 RR-2 RR-1 RR-1
Notes_: 1. As demonstrated later, the last two firms pro-
duce identical results and it is only necessary
to consider one of them.
2. IDC Rate Formulas
IDC-1: zero (do not capitalize IDC)
IDC-2: (debt ratio x interest rate) + (1.0 debt
ratio) x equity rate
IDC-3: debt ratio x interest rate
3. ITB Rate Formulas
ITB-1: zero (do not allocate the ITB)
ITB-2: debt ratio x interest rate x tax rate
(allocate the ITB)
4. Formulas for Revenue Requirements
RR-l: Firms which include IDC below the line for
rate making purposes:
1/1 TX [DBK + AIR x RB) + ITB TX (DTX + INT)]
RR-2: Firms which include IDC above the line for
rate making purposes:
1/1 Tx [DBK + (AR x RB) IDC + ITB TX (DTX
where DBK = book depreciation, AR =- allowed rate of
return, RB = rate base, ITB = interest-tax benefit, IDC
= interest during construction, DTX = tax depreciation,
and INT = interest expense.
The Firm as an Aggregation of Assets
To facilitate the explanation of the alternative procedures for
providing a return on funds invested in CWIP, it is useful to think of
the firm as an aggregation of numerous individual assets, or "'units of
plant." Every unit of plant must be financed, and they all generate
revenues, depreciation, and output over their useful lives. In every
case a unit of plant constructed by the firm goes through a cycle of
construction and then depreciation.
Table 7, which depicts the growth of a hypothetical firm over a
three-year period, is founded on the following two assumptions: 1) The
firm has no construction work in progress prior to the first year shown
and 2) each asset requires an investment of $100 during each year of its
construction period. In the first year, construction is initiated on
four assets, A, B, C, and D. The first asset (A) requires two years for
construction and has a useful life of four years. The other three as-
sets (B, C, and D) begun in year one have construction periods of two,
three, and four years, and useful lives of six, eight, and ten years,
respectively. With an investment of $100 required for each asset, the
LIFE CYCLES OF ASSETS FOR A FIRM
OVER A THREE-YEAR PERIOD OF GROWTH
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
A C C DDD D
B C CD DD D DD
C CC CD DD DD DD D
D C C C C D D D D D D D D D D
E C CD DD D
F C CD DD DD D
G C C C D D D D D D D D
H C C C D D D D D D D D)
I C C C C D D D D D D D D D D
J C C DD DD DD
K C C C D D D D D D D D
L C C C C D D D D D D D D D D
Notes: 1. The "C's" represent the years in which the asset
is under construction and costs are being incurred.
2. The "D's" represent the years in which deprecia-
tion occurs and inflows accrue.
total investment for year one is $400.
In the second year, construction is begun on five additional assets.
These assets have construction periods and useful lives similar to those
assets begun in year one. Thus, nine assets are under construction in
year two. With a $100 investment required for each asset, the total in-
vestment for year two is $900. At the end of year two, the amount of
CWIP accumulated in each asset begun in year one is $200, and the amount
of CWIP accumulated in each asset begun in year two is $100. The total
amount of CHIP at the end of year two is $1,300.6
At the beginning of the third year, construction begins on three
additional assets. At the same time, two of the assets begun in year one
are completed and go on line. This leaves ten assets still under con-
struction in year three, and will again require an additional investment
of $100 for each of them. Thus, the total investment outlay for the year
is $1,000. The two remaining assets under construction from year one
each represent $300 of accumulated CWIP, while each asset begun in year
two represents $200 of accumulated CHIP, and each asset begun in year
three represents $100 of accumulated CWIP. Therefore, the total CWIP at
the end of year three is $1,900. The $200 invested in each of the com-
pleted assets begun in year one is now included in the rate base and no
longer represents CWIP.
This hypothetical example was designed to illustrate the indepen-
dence of each asset. As the preceding discussion revealed, to determine
the total CWIP for the firm as a whole, it is only necessary to aggregate,
bEhroughout this illustration we ignore IDC since it would compli-
cate the example and add nothing to the point we are trying to make. The
procedures for dealing with IDC are explained later.
or sum, the amounts of CWIP associated with each asset. Similarly, the
total investment required by the firm in any period is simply the sum of
the investments required for each asset. Though not illustrated, this
independence also holds for items such as depreciation and maintenance
costs. More importantly, for regulated companies, this independence
exists for items such as revenues and IDC. Thus, to determine the value
of any of these variables for the firm, it is only necessary to sum the
value of the variables for each of the individual assets. Given this in-
dependence of assets, the procedures followed by the various firms can be
illustrated using a single asset. Consequently, to get a complete pic-
ture of each firm, a single asset which goes through the complete cycle
of construction and depreciation is all that is necessary.
Illustration of the Alternative Firms
With the Single Asset Model
Using the single asset model greatly simplifies the explanation of
the alternatives and should allow the reader to more easily follow the
presentation. The illustrations which follow consist of a series of
beginning and ending balance sheets and income statements for selected
years during the asset's life cycle. These illustrations require a num-
ber of assumptions.
1. On January 1, 1971, each firm is incorporated and as-
signed a group of customers for whom it must provide
service beginning January 1, 1973. The service re-
quirement for each firm is identical, and service must
be provided for four years, from January 1, 1973 to
December 31, 1976.
2. Customers may be billed beginning January 1, 1971,
depending on the requirements of the individual firm.
3. The construction period is two years, from January
1, 1971 to December 31, 1972. A total investment
of $200 is required for each firm to complete
construction. Half of the investment is made on
January 1, 1971, and half on January 1, 1972.
4. The depreciable life is four years, and the de-
preciation policy followed is straight line. Cash
generated by depreciation is used to retire debt
and equity in equal proportions.
5. Revenues are based on the rate base at the beginning
of the year,
6. The return on equity is determined on equity held
at the beginning of each year. The return on equity
is always 12 percent. Equity holdings may be in-
creased or reduced with no flotation costs.
7. Operating expenses for each firm are identical
and, therefore, omitted from the income statements.
8. All profits are paid out in dividends.
9. The capital structure is always maintained at 50
percent debt and 50 percent equity.
10. The interest rate on all debt is 8 percent. Debt
may be increased or reduced at no cost, other than
the interest expense.
11. The average cost of capital is 10 percent, which is
also the allowed rate of return. The gross IDC rate
is 10 percent, and the net rate is 8 percent.
12. The tax rate is 50 percent. IDC is not taxable in-
These assumptions result in firms with identical commitments, finan-
cial requirements, and service requirements. They differ only in the
method of providing a return on CWIP. For each firm, beginning and end-
ing balance sheets and income stateadnts are provided for 1971, 1972,
1973, and 1976. The first two years represent the construction period,
the third year is the year the asset goes on line, and 1976 is the last
year of the useful life of the asset. Nothing uniquely significant oc-
curs during 1974 and 1975; consequently, the data for these years are
not provided. The important procedures followed by each firm are fully
revealed in the years for which data are presented. For each year, the
beginning balance sheet, income statement, and ending balance sheet are
discussed and significant changes noted. The objective is to illustrate
the complete cycle of an asset and show how the different policies of
each firm are reflected in the balance sheets and income statements.
By the end of 1976, the asset constructed by each firm during 1971-1972
is fully depreciated, the service requirement has been fulfilled, and the
cycle is complete.
Firm RB invests $100 at the beginning of 1971, as shown in the bal-
ance sheet for January 1, 1971, in Table 8. Since Firm RB includes CWIP
in the rate base, the firm has a base upon which operating revenues may
be earned. The customers of Firm RB begin making utility payments in
1971, though service has not yet begun. This simplified example illus-
trates that, under the procedures followed by Firm RB, current customers
must pay part of the cost of an asset from which they do not currently
benefit. As the income statement shows, sufficient operating revenues
must be generated to cover taxes, interest, and profit. Net operating
income (NOI), as always, is 10 percent of the rate base. The balance
sheet at the end of the year is the same as the balance sheet at the be-
ginning of the year.
In 1972, another $100 is invested, financed equally by debt and
equity, raising CWIP to $200. The full $200 is included in the rate
base, and sufficient revenues must be generated to cover interest, taxes,
and profit. As the income statement for 1972 shows, NOI is 10 percent of
BALANCE SHEETS AND INCOME STATEMENTS FOR FIRM RB
Invest $100 for Construction
Invest $100 for Construction
$200 00 Total Liab.
Year Ending 12/31/72
S100.00 Total Liab.
Year Ending 12/31/71
$100 00 Total Liab.
$2000 Total Liab.
TABLE 8 (Continued)
Total Liab. $200 00
Total Liab. $50 00
$20 0 0
Year Ending 12/31/73
Year Ending 12/31/76
& 3 00
Oper. Plant 200.00
Debt $ 75.00
Total Liab. $150.00
the rate base. Customers are again required to make payments in 1972
though service has still not begun. The balance sheet at the end of 1972
At the beginning of 1973, the asset which had been under construc-
tion goes on line. The balance sheet for January 1, 1973 no longer lists
the assets as CWIP, but as operating plant. The rate base is unchanged
since the investment, as CWIP, was included before, and, as operating
plant, is included in the rate base now. The income statement for 1973
indicates that sufficient revenues must be generated to cover not only
taxes, interest and profits, but depreciation as well. NOI is again 10
percent of the rate base. The balance sheet at the end of 1973 includes
some depreciation which accumulates over the life of the asset. Debt and
equity are reduced as the net asset value falls.
By the start of 1976, accumulated depreciation has reduced the net
plant to $50, which is also the rate base. Debt and equity are both $25.
The income statement is generated as usual, with NOI 10 percent of the
rate base. By the end of 1976, the asset is fully depreciated, and the
cycle is complete.
Firm RB-IDC, like Firm RB, includes CWIP in the rate base, but also
capitalizes IDC at the gross rate of 10 percent. The amount of IDC is
then included above the line for rate making purposes. Since IDC is in-
cluded above the line, adjusted net operating income, rather than net
operating income, must always be 10 percent of the rate base. In 1971,
$100 is invested in construction, which is financed equally by debt and
equity. As shown in Table 9,7 IDC is calculated as 10 percent of CHIP,
or $10, Since adjusted NOI must be equal to 10 percent of the rate base,
NOI must be zero. With NOI zero, profits, for tax purposes (i.e., ex-
cluding IDC), are minus $4, or a $4 loss. Consequently, taxes are minus
$4, or a $4 credit, and revenues are minus $4.8 At the end of 1971, the
$10 of IDC is added to CWIP and debt and equity are both $5 larger.
Table 10 illustrates the "T-accounts" which help explain what actually
occurs with the creation of IDC. First, IDC is credited to other income
and simultaneously added to CWIP as a construction expense. Next, $4 in
cash is used to pay the interest expense and another $6 in cash is used
to pay dividends. Finally, $10 in cash is raised by increasing debt and
equity by $5 each to meet the $10 cash expenditure for interest and divi-
dends. All the T-accounts balance, and the balance sheet at the end of
1971 is as shown in Table 9. Similar transactions occur in every in-
stance in which IDC is capitalized.
At the beginning of 1972, another $100 is invested in construction.
The rate base is now $210, so that adjusted NOI must be $21. IDC is cal-
culated as 10 percent of CWIP, or $21 and, therefore, NOI is again zero.
For tax purposes, an $8.40 loss results, since IDC is not taxable in-
come. An $8.40 tax credit is incurred, making revenues minus $8.40 for
1972. At the end of 1972, IDC is added to CWIP and debt and equity are
Table 9 shows the balance sheets and income statements for Firm
RB-IDC and Firm IDC. As explained after the discussion of Firm IDC, the
results for these two firms are identical, and Table 9 is used for the
illustration of both firms.
8In a more realistic situation with an on-going firm, revenues
would be $4.00 less, rather than negative. Also, taxes would simply be
lower, rather than having a credit,
BALANCE SHEETS AND INCOME STATEMENTS FOR FIRMS RB-IDC AND IDC
Invest $100 for Construction
Invest $100 for Construction
$100.00 Total Liab.
Year Ending 12/31/71
Total Liab. $210 00
Year Ending 12/31/73
Net Oper. Income (NOI)
Debt $ 55.00
Total Liab. )]JE X)
$231 00 Total Liab,
- --- -
TABLE 9 (Continued)
Total Liab. $231.00
Total Liab. $57.75
Year Ending 12/31/73
Year Ending 12/31/76
$ 7 .2
AD usted NOI
$173.25 Total Liab.
Oper. Plant 231.00
$-O- Total Liab.
Other Income Dividend Interest
(IDC) CWIP CashExesEpee
10 (1) (1) 10 (4)(5 :5 4 (3)2)(36(2
At the beginning of 1973, the asset goes on line with a book value
of $231. This amount is higher than the book value of the asset of Firm
RB by the amount of accumulated IDC. The income statement is derived in
the usual manner, with one exception. For tax purposes, depreciation is
calculated only on the net-of-IDC value of the asset, or $200. To allow
the IDC part of the asset to be depreciated for tax calculations would
make the IDC income of 1971 and 1972 tax free. This procedure would pro-
vide unjustifiable benefits to firms which capitalize IDC. The tax cal-
culation for Firm RB-IDC for 1973 is illustrated below:
T-ACCOUNTS FOR FIRM RB-IDC FOR 1971
TRANSACTIONS REGARDING IDC
Taxes @ 50 percent
Less Accumulated IDC
$200.00 over four years equals
'$ 50.00 per year
Straight line depreciation, for tax purposes, is $50 per year, not $57.75,
which is book depreciation. This procedure for calculating taxes is used
for all firms which capitalize IDC. The income statement for 1973 shows
that adjusted NOI is 10 percent of the rate base. The balance sheet at
the end of 1973 reveals that net plant is reduced by accumulated deprecia-
tion as it accrues over the life of the asset. Debt and equity are corres-
At the beginning of 1976 net plant is $57.75, as is the rate base.
Debt and equity have been periodically reduced, corresponding to the net
value of the asset. The income statement is derived as usual, with NOI
10 percent of the rate base. The final balance sheet shows the asset
fully depreciated, and the cycle is complete.
Firm RB-%IDC-ITB also includes CWIP in the rate base. Like Firm
RB-IDC, this firm capitalizes IDC and includes it above the line. This
again requires that adjusted NOI be equal to 10 percent of the rate base.
Firm RB-%IDC-ITB differs from Firm RB-IDC in two important respects.
First, Firm RB-%IDC-ITB calculates IDC at a lower rate of 4 percent.
Second, the ITB is allocated to the CWIP account. In 1971 the firm in-
vests $100 in construction, as shown in Table 11. As shown in the income
statement, IDC is computed at 4 percent. Since adjusted NOI is 10 per-
cent of the rate base, of $10, NOI must be $6. With taxes of $4 and the
ITB of $2, revenues are $12. This firm, like Firm RB, requires its cus-
tomers to make payments in 1971, even though service has not yet begun.
BALANCE SHEETS AND INCOME STATEMENTS FOR FIRM RB-IDC-ITB
Invest $100 for Construction
Invest $100 for Construction
$100 00 Total Liab
Year Ending 12/31/71
Total Liab. $202.00
Year Ending 12/31/72
TABLE 11 (Continued)
Oper, Plant 206.04
Net Plant $206.04
Total Assets $206 0
Total Liab. $20604
Total Liab. $5 .51
Year Ending 12/31/73
Year Ending 12/31/76
$154.53_ Total Liab.
Oper. Plant 206.04
$-0- Total Liab.
The payments required are not as large as those required by Firm RB since
some IDC is capitalized and included in earnings. The balance sheet at
the end of 1971 reveals that IDC is added to CWIP and the ITB has also
been credited to CWIP. With IDC of $4 and an ITB of $2, the net addition
to CWIP is $2. Consequently, debt and equity are increased by $1 each.
At the beginning of 1972, another $100 is invested in construction.
The rate base is now $202, and adjusted NOI is 10 percent of this, or
$20.20. As shown, revenues are again positive, requiring customers to
make payments in 1972 even though service has not yet begun. IDC is cal-
culated at 4 percent, and the ITB is allocated to CWIP. Rather than the
CWIP increasing by the total IDC of $8.08, the allocation of the ITB of
$4.04 makes the net increase in CWIP only $4.04. As a result, debt and
equity are increased by $2.02 each.
In 1973 the asset goes on line and depreciation begins. The book
value of the asset is greater than for Firm RB since some IDC is capital-
ized, but the book value is less than for Firm RB-IDC since IDC is
capitalized at a lower rate. The rate base is $206.04, NOI is $20.60,
and there is no IDC or ITB since CWIP is now zero. Depreciation, for
tax purposes, is determined as explained earlier. The balance sheet at
the end of 1973 shows the net value of the asset being reduced by de-
preciation. Debt and equity are reduced correspondingly.
By the start of 1976, the net plant has been reduced substantially
by accumulated depreciation. Adjusted NOI is 10 percent of the rate base,
and there is no IDC or ITB. By the end of 1976, the asset is fully de-
preciated and the cycle is complete.
Firm IDC does not include any part of CWIP in the rate base, but
capitalizes IDC at the gross rate as a means of providing a return on
funds invested in CWIP. Given this procedure, NOI must always be equal
to 10 percent of the rate base. In 1971, $100 is invested in construction
as shown in Table 9.9 The rate base is equal to zero, since CWIP is
excluded from the rate base. With NOT equal to 10 percent of the rate
base, or zero, and IDC being non-taxable, profits, for tax purposes, are
minus $4. Taxes are also minus $4, or a $4 credit, and, therefore, rev-
enues are minus $4. At the end of 1971, IDC has been added to CHIP and
debt and equity have beenl increased.
In 1972 another $100 is invested in construction. The rate base is
still zero, making NOI zero. IDC is 10 percent of CWIP, or $21, but not
taxable income. For tax purposes, profits are minus $8.40, and the tax
credit for 1972 is $8.40. IDC is again added to CWIP, and debt and equity
have both been increased by $10.50. For both 1971 and 1972, the customers
of Firm IDC are not required to make any utility payments. The total re-
turn to funds allocated to CWIP is provided by the IDC credit.
In 1973, the asset goes on line and the rate base is $231. The book
value of the asset for Firm IDC is higher than the book value for Firm RB
by the amount of accumulated IDC. The income statement for 1973 reveals
that NOI is 10 percent of the rate base and IDC is zero. Here again, as
with earlier firms, depreciation, for tax purposes, is allowed only on the
net-of-IDC cost of the asset. Depreciation accrues this year, and by the
end of 1973 the net plant has been reduced. Debt and equity are reduced
As noted earlier, Table 9 illustrates the results for Firm RB-IDC
and Firm IDC.
As with the other firms, by 1976 the net plant has been reduced by
accumulated depreciation. NOI is 10 percent of the rate base, and IDC is
zero. By the end of 1976, the asset is fully depreciated and the cycle
for the asset of Firm IDC is complete.
As mentioned earlier, Firms RB-IDC and IDC produce identical re-
sults. This is only true, however, when the IDC rate is equal to the al-
lowed rate of return on used and useful assets, as is true in the illus-
trations of this chapter. Any difference between the IDC rate and the
allowed rate will produce different results for the two firms. This is
illustrated in Table 12, where the IDC rate for Firm RB-IDC is now 8 per-
cent rather than the original 10 percent. Using the data for 1971, Table
12 shows that taxes and revenues now differ for the two firms. Both are
still providing a 12 percent return on equity. Firm RB-IDC still has an
adjusted NOI equal to 10 percent of the rate base, but $8 is in the form
of IDC credits and $2 is generated from revenues after taxes. Previous-
ly, the entire $10 was in the form of IDC credits. While the two firms
produced identical results for the assumptions specified earlier, the two
firms could produce quite different results under different assumptions.
In addition to capitalizing interest at the gross IDC rate, Firm
IDC-ITB allocates the interest-tax benefit to the construction account.
"Thug the_ accounting procedures_ differ slightly Firms IDC-ITB-G
and IDC-ITB-N produce identical results. This is illustrated in the ap-
pendix to the chapter. Therefore, we shall deal only with Firm IDC-ITB-
G, and it will be referred to as Firm IDC-ITB in the remainder of this
COMPARISON OF RESULTS WHEN THE RATE FOR COMPUTING IDC
DIFFERS FROM THE ALLOWED RATE OF RETURN FOR FIRM RB-IDC
Firm RB-IDC Firm IDC
Allowed Rate 10% 10%
IDC Rate 8% 10%
Year Ending 12/31/71
Revenues $ -0- -$ 4.00
Depreciation -0- -0-
Taxes -2.00 -2.00 -4.00 -4.00
NOI 2.00 -0-
IDC 8.00 10.00
Adjusted NOI 10.00 10.00
Interest Expense 4.00 4.00
Net Income $ JE.000 $ 6.00
In 1971, $100 is invested in construction and is financed equally by debt
and equity. Table 13 shows that NOI is 10 percent of the rate base,
which is zero since CWIP is excluded from the rate base. IDC is calcu-
lated as 10 percent of CHIP, and the ITB is allocated as illustrated in
the income statement. Taxable income is -$4.00, which makes taxes -$2.00.
Adding the ITB to the tax expense makes net taxes zero and thus, revenues
are zero also. The ending balance sheet shows that $10 of IDC has been
added to CWIP, which, with the ITB allocation of $2, makes the net addi-
tion to CWIP $8. Debt and equity are both increased by $4.
In 1972, another $100 is invested in construction. The rate base is
still zero, making NOI zero. IDC is again computed as 10 percent of CWIP
and the ITB is allocated to the income statement. Taxable income is
-$8.32, which makes taxes -$4.16. The allocation of the ITB of $4.16
makes net taxes zera and again revenues are zero. The procedures followed
BALANCE SHEETS AND INCOME STATEMENTS FOR FIRM IDC-ITB
Invest $100 for Construction
Invest $100 for Construction
$100 00 Total Liab.
Year Ending 12/31/71
Total Liab. $208 00
Year Ending 12/31/72
Debt $ 54.00
Total Liab. $108.00
$224.64 Total Liab.
TABLE 13 (Continued)
$5 6 1
Total Liab. $56 16
Oper. Plant 224
$ 24 64 Total Liab.
Year Ending 12/31/76
Year Ending 12/31/73
( 3 7
Oper. Plant 224.64
$168_48 Total Liab.
by Firm IDC-ITB do not require customers to make payments during the
1971-1972 period. The entire return for funds invested in CWIP is pro-
vided by IDC credits. The ending balance sheet shows that IDC has been
added to CWIP and the ITB has been credited to CWIP, making the net addi-
tion $16.64. Debt and equity increase by $8.32 each.
When the asset goes on line in 1973, the book value of the asset for
Firm IDC-ITB is less than for Firm IDC since a smaller net amount of IDC
was added to CWIP. The rate base is now $224.64, and NOI is 10 percent
of the rate base, or $22.46. Depreciation begins this year and, for tax
purposes, is computed on the book value of the asset, excluding accumu-
lated IDC. The balance sheet at the end of 1973 shows the declining
value of net plant as depreciation accumulates.
By the beginning of 1976, net plant has been reduced to $56.16, due
to .accumulated depreciation. NOI is again 10 percent of the rate base.
By the end of 1976, the asset is fully depreciated and the cycle is again
While the explanation of the procedures followed by each firm re-
quires only a single asset model, the evaluation of the different methods
for providing a return on CWIP requires that the firm as a whole be
examined. It is the cumulative effects of rapid, fluctuating growth
which create problems. In the next section, another model is briefly
described which is somewhat more realistic and begins to demonstrate these
The Continuous Growth Model
The continuous growth model is simply an extension of the single as-
set model. Most of the assumptions of the single asset model are re-
tained, but a few modifications are required. First, the service re-
quirement, which is identical for all firms, is such that capacity must
be expanded by a constant amount. After the first two years, an asset
similar to the asset in the single asset model must go on line each year.
This assumption requires construction to begin on a new asset each year.
With a two-year construction period, an investment of $200 must be made
in each year after 1971, while the first year still requires only a $100
investment. Of the $200, $100 is allocated to the construction of an
asset begun in the previous year, and $100 is allocated to the construc-
tion of an asset begun in the current year. This process is assumed to
go on indefinitely.
Another major change is that the funds generated by depreciation are
now used to maintain the asset at full use and value. The maintenance
cost of each asset is $50 per year, w~thich is the same for each firm.
This change means that the "IDC part" of the asset is depreciated and
"retired" over the asset life. An example should make this result clear.
The book value of an asset, excluding IDC, is $200. Thus, over four
years, depreciation, excluding IDC, is $50. The book value of an asset
for each firm differs, depending on the amount of IDC capitalized. Using
Firm 1DC for the illustration, an asset has a book value of $231, $31 of
which is IDC. Book depreciation, calculated on $231, is $57.75 each
year. Of the $57.75, $50 is used to maintain the full usefulness of the
asset and $7.75 is depreciated and "r-etired." The funds generated by
depreciated IDC are used to reduce debt and equity in proportion to the
debt ratio. At the end of four years, an asset maintained in this way
will have a book value of $200, rather than $231, since $7.75 per year,
or $31 over four years, is written off. The part which is written off
is always the amount of accumulated IDC for each firm. This procedure
allows the rate base to be maintained, without altering the comparability
of the various firms.
The final modification for the continuous growth model specifies
that each firm begins with an asset base of $1,000, rather than zero.
Thus, each firm has a beginning rate base of $1,000.
As demonstrated earlier, the firm can be viewed as an aggregation
of assets. This model illustrates the impact of adding a series of sin-
gle assets, similar to the asset in the single asset model, to an exist-
ing firm. This model is one step closer to the more complex computer
nodel which is described in the next chapter. Since the procedures fol-
lowed by each firm were adequately demonstrated by the single asset
model, only the results are presented for this model.
Examination of the Results
This section reviews the results of the two models with respect to
two variables, revenues and times-interest-earned ratios. The single
asset model indicates how the variables change over the complete life
cycle of a single asset. The continuous growth model illustrates the
cumulative effects on each variable of adding a series of such assets to
an existing firm. Each model provides unique insights into the various
methods for providing a return on funds invested in CWIP.
The Sinele Asset Model Results
Revenues. The single asset model best illustrates two facts re-
garding the revenues generated by each firm. First, the revenues re-
quired over a complete life cycle indicate the cost to the customer of
the procedures followed by each firm. Second, revenues indicate the
timing of payments required by each firm, since the service requirement
(or customer group size) is the same for all firms. Table 14 shows the
revenues required by each firm over the life cycle. Of particular in-
terest to customers is the overall cost they incur under the procedures
followed by each firm. Since the timing of payments differs for each
firm, it is necessary to compare the net present value (NPV) of the
revenue requirements of each firm." As Table 14 reveals, the NPV of
revenue requirements for each firm is identical. This equivalence means
that the cost to the customers is the same over a complete life cycle.
Consequently, the importance of cost considerations in evaluating the
SINGLE ASSET MODEL
RB RB-IDC RB-%IDC-ITB IDC IDC-ITB
1971 16.00 -4.00 12.00 -4.00 -0-
1972 32.00 -8.40 24.24 -8.40 -0-
1973 82.00 102.46 85.98 102.46 98.26
1974 74.00 93.22 77.74 93.22 89.28
1975 66.00 83.98 69.50 83.98 80.29
1976 58.00 74.74 61.26 74.74 71.31
Net Present Value of Revenue Requirements at 8 Percent
243.20 243.20 243.20 243.20 243.20
The rate used to discount the stream of revenues is the net (af-
ter-tax) rate of return. The allowed after-tax rate' is 8 percent in
firms is secondary. The timing of payments, cash flows, and other vari-
ables become more important.
The timing of payments is important to individual customers. Cur-
rent customers want to pay only for those assets which benefit them. Two
of the firms examined, Firms RB and RB-%IDC-ITB, require customers to
nake payments during the construction period. In an on-going firm, this
result means that current customers pay part of the cost of assets which
are not currently used to their benefit. On the other hand, Firms RB-
IDC and IDC both show negative revenues during the construction period.
In an on-going firm, present customers would be receiving a benefit from
construction. This benefit results from the failure to allocate the tax
savings, or ITB. This reduces revenues currently required, which bene-
fits present customers, and raises the revenues required in later years,
to the detriment of future customers. Firm IDC-ITB, which capitalizes
IDC and allocates the ITB, has revenues of zero during construction.
This result means that present customers pay no part of the cost of as-
sets that are to be used in the future. Firm IDC-ITB is the only firm
which fully reflects the exact capital costs of construction in the cost
.of the asset.
Figure 2 is a graph of the revenue requirements of each firm from
the single asset model. Firm RB produces the least extreme changes in
revenue requirements. During construction revenues for Firm RB are the
highest, and after construction they are the lowest. Firm RB-%IDC-ITB,
which capitalizes a small amount of IDC, produces slightly more extreme,
results. Firm IDC-ITB, which, in effect, capitalizes IDC at the net
rate, produces still more extreme results. Finally, Firms RB-IDC and
IDC, which capitalize the most IDC, produce the widest fluctuations .in
1971 1972 1973 1974 1975 1976
REVENUE REQUIREMENTS: SINGLE ASSET MODEL
revenue requirements. In general, the more IDC capitalized, the greater
the fluctuation in revenue requirements. The importance of this fluctua-
tion depends on many other factors, particularly demand and other cost
considerations. At this point, we merely recognize this fluctuation,
leaving its evaluation to subsequent discussions.
Times-Interes t-Earned Ratios.1 2 In addition to indicating the cov-
erage of interest charges, the times-interest-earned ratio provides a
rough indication of cash flow. Table 15 summarizes these ratios for each
firm in each year. Figure 3, which graphs these ratios, indicates that
they respond in a manner similar to revenues. Firm RB has a constant
coverage ratio. Firm RB-%IDC-ITB, which capitalizes a small amount of
IDC, has relatively lower ratios during construction and somewhat higher
ratios after construction. Firm IDC-ITB, which capitalizes more IDC, has
more widely fluctuating ratios before and after construction, while Firms
RB-IDC and IDC have ratios which fluctuate the most. Like revenues, the
fluctuation increases as the amount of capitalized IDC increases. This
fluctuation results from the fact that during construction, cash revenues
are replaced by IDC credits, in a proportion of two-to-one. That is, for
every dollar of IDC, revenues can be reduced by two dollars, since IDC is
not taxable. After construction, or during the asset's useful life, an
opposite effect occurs. The larger rate base of the firms which capi-
talize IDC cause larger cash flows relative to interest payments and the'
ratios are higher. Consequently, for the single asset model, more IDC
"Times-interest-earned is calculated as the ratio of income before
taxes to interest charges. Since IDC is not cash income, it is excluded
from the calculation of the coverage ratio.
7.0 r //IDC-ITB
1971 1972 1973 1974 1975 1976
TIMES-INTEREST-EARNED RATIOS: SINGLE ASSET MODEL
SINGLE ASSET MODEL
Year ; Firm
RB RB-IDC RB-IDC-ITB IDC IDC-ITB
1971 4.00 -1.00 3.00 -1.00 -0-
1972 4.00 -1.00 3.00 -1.00 -0-
1973 4.00 4.84 4.18 4.84 4.69
1974 4.00 5.12 4.24 5.12 4.91
1975 4.00 5.68 4.37 5.68 5.37
1976 4.00 7.35 4.73 7.35 6.74
during construction makes the times-interest-earned ratios relatively
low during construction and relatively high after construction ends.
The examination of these two variables has revealed that the larger
the amount of IDC, the greater is the fluctuation in both revenues and
times-interest-earned ratios. These results, particularly for the period
after construction, are somewhat exaggerated by the use of a short de-
preciable life. They are further exaggerated by the absence of continued
construction and growth. The results for the continuous growth model,
though still not entirely realistic, more accurately depict what might
be found in the real world.
Continuous Growth Model Results
Revenues. Table 16 summarizes the revenues generated in the con-
tinuous growth model, and Figure 4 illustrates them graphically.- The
pattern during the initial construction period resembles the pattern
established in the single asset model. Also like the single asset model
results, Firm RB has the highest revenues initially but the lowest later
'71 '72 '73 '74 '75 '76 '77 '78 Year
REVENUE REQUIREMENTS: CONTINUOUS GROWTH MODEL
CONTINUOUS GROWTH MODEL
RB RB-IDC RB-IDC-ITB IDC IDC-ITB
1970 410.00 410.00 410.00 410.00 410.00
1971 426.00 406.00 422.00 406.00 410.00
1972 458.00 397.60 446.24 397.60 410.00
1973 540.00 500.06 532.23 500.06 508.26
1974 622.00 601.28 617.97 601.28 605.54
1975 704.00 701.26 703.47 701.26 701.83
1976 786.00 800.00 788.74 800.00 797.14
1977 868.00 882.00 870.74 882.00 879.14
1978 950.00 964.00 952.74 964.00 961.14
on. The other firms follow the same relative pattern. But with con-
tinued investment, the shifts in relative position are less extreme and
take longer to occur. Once a full cycle has been completed, by 1976,
the annual increase in revenues is $82 for all firms. This constant in-
crease in revenues is due to the pattern of constant growth specified in
Times-Interest-Earned Ratios. Table 17 summarizes the times-
interest-earned ratios, which are graphed in Figure 5. Firm RB still
has a constant ratio of income to interest charges. The ratios for each
of the other firms decline during the first two years in proportion to
the amount of IDC capitalized by each. The ratios for these firms then
begin to rise and asymtotically approach the level of Firm RB as the
percentage of .CWIP to other assets falls. That is, as the percentage of
CWIP to other assets falls, IDC as a percentage of earnings falls, and
times-interes t-earned ratios approach the level of Firm RB, where no IDC
'71 '72 173 '74 '75 '76 '77 '78
TIMES-INTEREST-EARNED RATIOS: CONTINUOUS GROWTH MODEL
CONTINUOUS GROWTH MODEL
RB RB-IDC RB-IDC-ITB IDC IDC-ITB
1970 4.00 4.00 4.00 4.00 4.00
1971 4.00 3.55 3.91 3.55 3.64
1972 4.00 2.82 3.77 2.82 3.06
1973 4.00 3.12 3.82 3.12 3.30
1974 4.00 3.34 3.87 3.34 3.47
1975 4.00 3.51 3.90 3.51 3.61
1976 4.00 3.65 3.93 3.65 3.72
1977 4.00 3.68 3.93 3.68 3.74
1978 4.00 3.70 3.94 3.70 3.76
The examination of the results of both models has revealed that sig-
nificant differences will result for each firm. The simplified models
examined in this chapter were designed to illustrate the various firms
and provide some insight into the results which might be expected. The
final section of this chapter adds another degree of realism before the
full computer simulation model is examined in the next chapter.
Under the assumptions specified earlier, the firms examined followed
a straight line depreciation policy. While some firms use a straight
line depreciation policy for both book and tax purposes, the majority of
utility companies use some form of accelerated depreciation for tax pur-
poses. The benefits of accelerated depreciation can be handled by two
basic accounting procedures. Under the flow through method, the tax
savings which occur early are flowed through to reported profits. Under
the normalization method, the savings are not included in reported prof-
its but are segregated in a reserve for deferred taxes. Arguments have
been made for both procedures, and both are widely used.l The importance
of depreciation policy to this study stems mainly from the differences in
the patterns of cash flow which result. Since the timing of cash flows
differs due to depreciation policy,' and also differs depending on the
policy followed for providing a return on CWIP, it will be useful to
examine whether these factors are compounding and create a serious cash
flow problem or, perhaps, are offsetting and minimize any potential cash
flow problem. Consequently, the combination of three depreciation poli-
cies with the five methods for providing a return on CWIP produces 15
firms which are analyzed using the computer simulation model.
This chapter explained each of the various methods for providing a
return on funds invested in CWIP. This was done using the single asset
model, which was shown to be sufficient for the purpose of illustrating
these procedures. In addition to the single asset model, a continuous
growth model was presented, and the impact on revenues and times-interest-
earned ratios was examined for each firm. The results indicated that
significant differences between the various firms could be expected. The
last section indicated that different depreciation policies are followed
by firms, and these have different impacts on variables of interest to
Brigham, E. F. and J. L. Pappas, Liberalized Depreciation and the
Cost of Capital, 1970 MSU Public Utilities Studies, MSU, 1970, pp. 1-2 and
"Ibid, Chapter 4.
this study. The result of this chapter is to suggest 15 prototype firms
which are to be analyzed by the computer simulation model.
As pointed out in Chapter 3, Firms IDC-ITB-G and IDC-ITB-N produce
identical results and, therefore, only one of these firms needs to be
analyzed. This is demonstrated below using the values of the gross and
net IDC rates as specified for the single asset model.
To illustrate the equality of these two firms, Table 18 outlines the
accounting treatment, the income statement effects, and the effect on the
construction account, of IDC and the ITB. Part A of Table 18 illustrates
the accounting treatment. For Firm IDC-ITB-G, construction is debited 10
percent and IDC income is correspondingly credited 10 percent. For the
ITB, income tax expense is debited 2 percent,' and the construction ac-
count is credited 2 percent. For Firm IDC-ITB-N, construction is debited
at the net rate of 8 percent and IDC income is credited 8 percent. The
income tax expense is again debited 2 percent, but for Firm IDC-ITB-N,
the offsetting transaction is a 2 percent credit to income taxes, listed
under other income.
Part B of Table 18 shows the income statement effects. Income taxes
are raised by 2 percent for both firms as illustrated by the accounting
transactions. For Firm IDC-ITB-G, other income is raised 10 percent due
to the IDC credit of 10 percent making net income 8 percent. Firm IDC-
ITB-N raises other income by 10 percent also, but only 8 percent is IDC
"The income statements shown in Tables 10 and 13 of Chapter 3 list
the ITB separately from the income tax expense. This is done merely to
emphasize the ITB.
ILLUSTRATION OF THE EFFECTS OF IDC AND THE ITB
FOR FIRMS IDC-ITB-G AND IDC-ITB-N
Debit Construction 10 Percent
Credit IDC Income 10 Percent
Debit Income Tax Expense 2 Percent
Credit Construction 2 Percent
Debit Construction (Net) 8 Percent
Credit IDC Income 8 Percent
Debit Income Tax Expense 2 Percent
Credit Income Taxes-0ther Income 2 Percent
Firm IDC-ITB-G Firm IDC-ITB-N
Income Tax 2 Percent 2 Percent
IDC 10 Percent 8 Percent
Income Tax Credit 2 Percent
Total 10 Percent 10 Percent
Net Income 8 Percent 8 Percent
Firm IDC-ITB-G Firm IDC-ITB-N
IDC 10 Percent 8 Percent
Income Tax Credit (2 Percent)-
Net IDC Cost 8 Percent 8 Percent
income while the other 2 percent is in the form of an income tax credit.
Net income is again 8 percent, the same for both firms.
Looking next to the construction account, Part C shows that the cost
of construction for both firms rises by 8 percent. For Firm IDC-ITB-G,
IDC of 10 percent is added to construction, but the allocation of the 2
percent tax credit makes the net addition 8 percent. Firm IDC-ITB-N only
adds IDC of 8 percent to the construction account, making the net increase
the same for both firms.
As shown, the net addition to the income statement is 8 percent for
both firms, as is the net addition to the construction account. Thus,
both firms produce the same net results, making it necessary to examine
only one of them.
THE SIMULATION MODEL
The models of the previous chapter are too simplified to provide
the kind of information necessary to evaluate the alternatives. A more
realistic model which tracks the effects of the alternatives over an ex-
tended period is too cumbersome to deal with manually. Consequently, a
computer simulation model is used which provides the necessary realism
and flexibility for a thorough analysis.
The validity of the results and the conclusions subsequently drawn
are directly related to the assumptions used in the development of the
model. Rather than simply list the assumptions, this chapter provides a
broad overview of the computer model, allowing the assumptions to be dis-
cussed as they arise. The chapter itself is primarily a verbal descrip-
tion of the model, with the important equations provided in Appendix A.
The first discussion examines some of the important considerations
designed to assure that the model is both realistic and valid. Along
with the assumptions, the input variables determine the usefulness of the
results. The discussion of the input options illustrates the flexibility
and range allowed in evaluating the alternatives. The output variables
are quite numerous and indicate the measures of performance to be used in
the analysis. Appendix B is the result of efforts to validate the model,
and Appendix C provides a listing of the actual computer program.
General Features of the Model
This section discusses four important features of the simulation
model. The model incorporates a start-up period which~is necessary to
avoid cycling and unrealistic beginning values. Another important as-
pect is the establishment of the proper timing of events to avoid a "re-
turn lag" or a built-in construction period. It is also crucial that
investment is determined in a manner which maintains the comparability of
the firms. Finally, and perhaps the most important aspect, is the model-
ing of the construction program. The realism of this feature greatly
affects the usefulness and validity of the model.
The Start-Up Period
The model has been designed to allow a start-up period of 100 years.
The start-up period assumes a construction period of zero,' which re-
sults in all firms beginning with the same initial values. The only dif-
ference in the firms at the beginning of the first year of interest is
that the normalizing firms have a reserve for deferred taxes included in
their capital structure. With depreciation being reinvested each year,
the start-up period minimizes the cycling that would otherwise be sig-
nificant. The initial 100 years also allows the cumulative reserve for
deferred taxes to be established, rather than unrealistically starting it
at zero. Beginning in the first year of interest, the construction ac-
tivities generally assume non-zero construction periods, and this change
causes the differences in firms. With respect to providing a return on
CWIP, a construction period of zero would produce identical results for
This is equivalent to saying that all facilities are purchased on
a turn-key basis throughout the start-up period.
all firms since there would not be any construction work in progress.
Having a start-up period eliminates some potential problems and
makes the model more realistic, thereby improving the quality of the re-
sults. Establishing the proper timing of events is also crucial to a
All of the stock variables represent either~ beginning-of-period
values or end-of-period values. Beginning-of-period values always re-
flect the investment made during the period. That is, all investment is
made at the beginning of each period. End-of-period values reflect all
the changes which occurred during the period, particularly with respect
to the flow variables associated with the income statement. While flow
variables represent flows which occur throughout the period, these are
accounted for at the end of each period, and, for convenience, may be
referred to as end-of-period values. Thus, we focus on two points in
time in each period, the beginning of the period and the end of the
period. With regard to the computation of such variables as interest,
rate of return on equity, rate base, and others, beginning-of-period
values are used. Since investments are made at the beginning of the
period, there is no "return" lag built into the model. That is, all
funds invested receive a full return from the moment of their commitment.
The Determination of Investment
Having assured that investments receive a full return, it is neces-
sary to examine the procedure which determines the size of the investment
made by the firms in each period. The amount of investment must be the