Essays in Honor of

William A. Paton
Pioneer
Accounting
Theorist
Edited by
Stephen A. Zeff
Joel Demski
Nicholas Dopuch

Division of Research
Graduate School of Business Administration
The University of Michigan
 nunc cocnosco ex parte

THOMAS J. BATA LIBRARY

TRENT UNIVERSITY
 Digitized by the Internet Archive
in 2019 with funding from
Kahle/Austin Foundation

https://archive.org/details/essaysinhonorofw0000unse_x1r7
1
Essays in Honor of
William A. Paton
WILLIAM A. PATON
Essays in Honor of
William A. Paton
Pioneer
Accounting
Theorist

Edited by
Stephen A. Zeff
Joel Demski
Nicholas Dopuch

Division of Research
Graduate School of Business Administration
The University of Michigan
Ann Arbor, Michigan
f-t r txpoM ,

Copyright © 1979
by
The University of Michigan

Printed in the United States of America

All rights reserved

Second Printing, 1981

ISBN 0-87712-183-4

Library of Congress Cataloging in Publication Data

Main entry under title:

Essays in honor of William A. Paton.

Bibliography: p.
CONTENTS: Greer, H. C. Greer on Paton.—Devine,
C. T. Observations on internal controls.—Solomons,
D. The politicization of accounting.—Bierman, H.,
Jr. [etc.]
1. Accounting—Addresses, essays, lectures.
2. Paton, William Andrew, 1889- I. Paton,
William Andrew, 1889— II. Zeff, Stephen A.
III. Demski, Joel S. IV. Dopuch, Nicholas.
HF5629.E68 657 78-27910
ISBN 0-87712-183-4
Contents

Foreword ix

Preface xi

Greer on Paton
Howard C. Greer 1

Observations on Internal Controls

Carl T. Devine 5

The Politicization of Accounting

David Solomons 25

The Feasibility and Desirability of Accounting Standards

Harold Bierman, Jr. 41

Accounting for Investments in Debt Securities

Maurice Moonitz 57

The Hard Core of Accounting

R. J. Chambers 73

Paton on the Effects of Changing Prices on Accounting,

1916-55
Stephen A. Zeff 91

Relationships among Income Measurements

Norton M. Bedford 139
Cost Accounting Principles for External Reporting:
A Conceptual Framework
Gordon Shillinglaw

Social Performance Accounting: An Unheeded Challenge

to the Accounting Profession
R. Lee Brummet

Bibliography
Contributors

STEPHEN A. ZEFF is Professor of Accounting at Rice University.

JOEL S. DEMSKI is Professor of Accounting at Stanford University.

NICHOLAS DOPUCH is Professor of Accounting at the University

of Chicago.

HOWARD C. GREER is a retired corporate executive and a former

member of the faculty of Ohio State University. He served as
president of the American Accounting Association in 1932.

CARL T. DEVINE is Professor Emeritus of Accounting at Florida

State University.

DAVID SOLOMONS is Arthur Young Professor of Accounting at

the Wharton School of the University of Pennsylvania.

HAROLD BIERMAN, JR. is the Nicholas H. Noyes Professor of

Business Administration at Cornell University.

MAURICE MOONITZ is Professor of Accounting at the University

of California (Berkeley).

R. J. CHAMBERS is Professor of Accounting at the University of

Sydney (Australia).
NORTON M. BEDFORD is Arthur Young Professor of Accounting
at the University of Illinois at Urbana-Champaign. •

GORDON SHILLINGLAW is Professor of Accounting at Columbia

University.

R. LEE BRUMMET is Willard J. Graham Professor of Business Ad¬

ministration at the University of North Carolina at Chapel Hill.
Foreword

Many individuals, organizations, and institutions have benefited

directly from William A. Paton's boundless energy, keen mind, and
unlimited dedication and enthusiasm for accounting. Foremost
among these are the University of Michigan's Graduate School of
Business Administration, its accounting faculty, and now the Paton
Center for Accounting Education and Research. But no less deserv¬
ing of mention are his numerous friends outside the University of
Michigan, former students, colleagues, and professional acquain¬
tances. Accordingly, we at the University are indeed gratified to see
him honored in this book of essays.
To the three editors and nine authors who have joined to make
this tribute possible, we express deep appreciation. Truly, William
A. Paton is a man of international renown, so respected and affec¬
tionately known throughout the academic and professional commu¬
nities for his perceptive and enduring contributions to accounting
theory and practice that all would do him honor.
With his colleagues, Professor Paton has left a legacy—and a
trust—of excellence in teaching, research, and professional service
to those of us now associated with the University of Michigan Ac¬
counting Faculty and the Paton Center. We acknowledge this legacy
as a challenge to make our service worthy of his example.

Donald H. Skadden
Chairman, Accounting Faculty and
first Director of the Paton Accounting Center

Floyd A. Bond
Dean, Graduate School of Business Administration

June 1, 1978
IX
Preface

Since the 1910s, William Andrew Paton has been a potent force in
the accounting literature. In the welter of disputation over practices,
rules, and conventions, Paton has consistently invoked economic
reasoning to counter the heavy hand of accounting tradition. He
has continually been concerned with the larger questions of valua¬
tion, the measurement of income, and the role of interest in ac¬
counting reckonings. To Paton, historical cost and accepted modes
of allocation were not hallowed shrines. Unlike many writers, he
has been interested not only in the technical content of accounting
but also in the likely impact of its measurements and disclosures on
the actions of management, investors, creditors, government, and
indeed the general public.
Above all, Paton has been an unflagging advocate of the causes
he believes to be just. Where Hatfield, the scholarly critic, might
tease and taunt his colleagues in the spirit of lively and even playful
debate, Paton has been forever serious in the pursuit of clearly
perceived goals. When he feared that accounting measures were
contributing to antibusiness sentiment following World War II, he
launched a vigorous campaign against accounting procedures that
gave rise to overstated profits and earning rates. When he believed
that utility regulatory commissions were following unsound poli¬
cies owing to their rigid adherence to "original cost," he set out in
hot pursuit of the regulators.
In a doctrinaire literature, Paton refused to join the crowd. He has
been an original thinker and an innovator, whose views are widely
quoted by others. It is evident from the writings of Hatfield, Can¬
ning, Sweeney, Gilman, Vatter, Davidson, Moonitz, Devine,
Chambers, Bierman, and others too numerous to mention, that

xi
xii Preface

Paton's arguments as well as the example of his incessant question¬

ing of accounting dogma have had a considerable impact on his
colleagues and successors in the literature.
Paton has advocated accounting reforms long before they finally
found their way into practice. One can recall his recommendation
as early as 1916 that debt discount be deducted from the face
amount of the long-term liability (and not shown as a deferred
charge among the assets), a practice that finally came to be accepted
in the United States in 1971. Also in 1971, the American accounting
profession officially endorsed the practice of accounting on a fair
value basis for the implicit interest on long-term contracts, a subject
on which Paton had written as long ago as 1916. The practice of
reducing gross revenue rather than charging "Bad Debt Expense"
when allowing for estimated uncollectibles was recommended by
Paton as early as 1922; even other textbook writers are today gradu¬
ally accepting this dictum. Paton also was a pioneer of new termi¬
nology. The terms "Allowance for Depreciation" and "Allowance
for Uncollectible Accounts" may be found in his textbooks as far
back as 1918, when "Reserves" were seen everywhere else. In 1948
a committee of the American Institute of [Certified Public] Accoun¬
tants recommended that the term "Reserve" no longer be used for
valuation accounts.
Paton may not have been the first American to write textbooks
that vigorously took issue with accepted accounting practice, but he
has been probably the most celebrated and influential author of this
genre of books. His succession of texts, beginning in 1916-18 with
the heretical Principles of Accounting (co-authored with Russell A.
Stevenson) and continuing with Accounting (1924), Essentials of Ac¬
counting (1938 and 1949), Advanced Accounting (1941), Asset Account¬
ing (1952), and Corporation Accounts and Statements (1955), plus sev¬
eral that were co-authored in later years, were hardly prosaic
primers on the rudiments of extant practice. They challenged the
intellect and obliged teacher and student alike to define and rede¬
fine the role of accounting in economic society. Paton was some¬
times criticized as a theorist who had little time for the "real
world." Yet the manuals which accompanied most of his textbooks
contained many problems and cases which were drawn from con¬
tracts, prospectuses, annual reports, court opinions, and newspaper
accounts, thus setting a very practical stage on which to discuss and
assess the implications of Paton's theories. He was among the first
authors of American accounting textbooks to make extensive use of
actual case materials.
 Preface xiii

Paton has won the widespread respect of his peers. He served as

president in 1922 of the American Association of University Instruc¬
tors in Accounting (forerunner of the American Accounting Asso¬
ciation). He was one of three accounting academics (the others were
A.C. Littleton and Roy B. Kester) who were named to the American
Institute's Committee on Accounting Procedure in 1938, when that
committee was first given authority to issue bulletins on accounting
principles on behalf of the Institute. In 1940 he became the first
academic to deliver the Dickinson Lecture in Accounting at the
Harvard Graduate School of Business Administration. In 1944 he
was one of the first three recipients of the Institute's Gold Medal for
distinguished service to the accounting profession, and in 1950 he
was among the first three individuals to be inducted into Ohio
State University's Accounting Hall of Fame. Paton was the first
academic to receive both of these accolades. In 1953 he was given
the Alpha Kappa Psi Foundation's Accounting Award.
Paton was a founder and first editor of The Accounting Review. He
was an author of the 1936 statement on accounting principles of the
American Accounting Association, and in 1940 he and A.C. Little¬
ton produced An Introduction to Corporate Accounting Standards,
which has doubtless been the most influential treatment of account¬
ing theory in the English language.
In 1947 Paton was one of approximately ten professors selected by
the University of Michigan to hold the new rank of University Pro¬
fessor. Twenty-nine years later, the same university opened the
William A. Paton Center for Accounting Education and Research
building, which was financed by contributions from his many
former students and other friends.
Now in his ninetieth year, Paton continues to write in the de¬
fense of free enterprise and sound accounting. We are pleased and
honored to be joined by eight of our colleagues in saluting this
extraordinary man. The essays in this volume have been written
exclusively for this Festschrift, and we are grateful to each of the
authors, to Howard C. Greer, and to the Division of Research at the
University of Michigan Graduate School of Business Administration
for making the publication possible. We thank Joseph M. Whelan
for his assistance in compiling the comprehensive bibliography.
Stephen A. Zeff

Joel S. Demski

Nicholas Dopuch
July 1978
Greer on Paton

In the January 1965 issue of the Accounting Review, Howard C.

Greer, a long-time friend of Paton's and himself a major contributor
to the accounting literature, wrote an admiring review of Paton on
Accounting, an anthology of Paton's articles and essays, edited by
Herbert F. Taggart. With the permission of the American Account¬
ing Association, we are pleased to reprint the review, as slightly
amended for this occasion by Greer.

Examining this monumental compendium of Bill Paton's voluminous

writings, I have been reminded of what happened, long years ago, to the
brothers Scott, two distinguished members of the faculty of my Alma
Mater, Northwestern University. John Scott was a renowned classical
languages scholar; his brother, Walter Dill Scott, was a noted teacher of
psychology, and later the university's long-time president.
As John Scott used to tell it: when the presidency became vacant, in
the early twenties, the brothers agreed that one of the Scotts should be
chosen for the job. It was decided that John would urge Walter's selec¬
tion, and Walter would promote John's candidacy. "But it turned out,"
mourned John, "that Walter didn't have any influence—and I didn't get
the job."
My experience has been similar. Years ago Bill Paton and I formed a
mutual admiration society, the terms of which were that each of us
would advertise the merits of the other's writings. I could suggest that
the wider audience attained by his works implies my greater success as a
promoter, though the relative merits of our literary efforts concededly
may have played some part in the result. Anyhow, I'll stick to our bar¬
gain, and put in one more plug for my old friend, fully realizing that no
living author needs it less!
Give heed, then, to this epic tome—the "Selected Writings" of the
sage of Ann Arbor—a collection of some 46 articles from his busy pen.
They are here republished in a handsome format, in chronological
order, with illuminating editorial notes by Paton's long-time friend and

1
2 Howard C. Greer

colleague, Herb Taggart (himself an author of distinction, whose dis¬

cerning and humorous comments add much to the book).
The Paton articles span the years from 1917 to 1963, and run the gamut
from an early-day exposition of the ' Theory of the Double Entry System
to such esoteric topics as "Non-Application of Incremental Cost Principle
to Postal Cost Accounting." Expectably, the majority of the material is
from the files of The Journal of Accountancy (to which he was a faithful
contributor), and The Accounting Review (of which he was founder and
first editor), but his interests ranged widely, and we find him enlighten¬
ing the readers of such diverse publications as the Journal of Political
Economy, the Public Utilities Fortnightly, and the report of the Michigan
Academy of Sciences.
The breadth of his audience-appeal is not surprising, for whatever else
may be said of him, he is certainly the most interesting writer who has
ever invaded the accounting field (with high rank in economic literature
as well). The sharpness of his perceptions, the pungency of his observa¬
tions, the clarity of his expositions combine to make him a fascinating
commentator on any subject he may elect to discuss. And he has always
had something valuable and thought-provoking to say; his lucid reason¬
ing is forever delightful, even when one fails to share his conclusions.
In fact, one of Paton's greatest services to his profession has been his
illumination of the many diverse facets of accounting theory and prac¬
tice. It is at once the delight and despair of accountants to realize that
most problems have several possible solutions, each useful and appropri¬
ate in its own context. Thus one may admire and applaud a reasoned
conclusion, even though one's own experience and outlook suggest a
different one. May a reviewer admit to differing with the author's con¬
clusion on eight of 12 major issues presented, and still acknowledge a
debt for a richer and fuller understanding of the subjects than would
have been possible without his astute and imaginative analysis?
And so, whatever your beliefs, you should know what Paton has to
say about recording appreciation in your accounts, figuring income
charges in terms of replacement-cost value-expirations, and "equalizing"
year-to-year income computations by varied accrual and write-off prac¬
tices. And if you marvel at his scorn for the helpful last-in first-out
principle of inventory valuation, you'll know that it extends likewise to
the hallowed cost-or-market-whichever-lower philosophy. And you
should identify him as the patron saint of the "one-step" net income
computation, even if you share this reviewer's regret at his championing
this dubious device.
Never mind, Bill—your admirers will forgive all in their gratitude for
your stalwart and undeviating espousal of private enterprise, your cogent
exposition of the indispensability of a free-price system, and your insis¬
tence on realism in measuring investment, capital, profits, and rates of
return. And that paper you presented at the International Congress in 1962
(at the tender age of 73)—"Accounting and Utilization of Resources"—as a
summation of economic realities and objectives, it's superb.
This book, however, is by no means limited to major issues and high-
level economic philosophies. Paton can be measured and profound; he can
also be topical, witty, and sardonic, as recent commentaries demonstrate.
 Greer on Paton 3

Always quick to identify and deride fads and foibles, he can still turn
a fast gun on exaggeration and sophistry. At an age when most of us are
content to doze in the sun, he grabs the old muzzle-loader off the wall,
rams in some shot, and blazes away at such targets as the "Deferred
Income Misnomer," the "Cash-Flow Illusion," the neglect of overhead
costs in price-figuring, and the like. Thus, if you're tempted to say some¬
thing silly in some accounting publication, take heed of the old red-head
skulking in the shrubbery—your temerity may fetch you a charge of
buck-shot in the seat of the pants!
So you ought to have a copy of this enduring monument to the Great
Educator, even if paying the $12.50 tab may deprive you of an evening
spent in a bar-room or at a girlie-show. If you are strong enough to heft a
volume of 713 pages, plus introduction and bibliographies, try your
muscle on this one.
And while so doing, let your enjoyment of Paton the author be en¬
riched by some related appreciation of Paton the man. If you have not
had the privilege of his acquaintance, you may still savor his warm and
engaging personality, from the photo on the dust jacket (he's saying
"Greer, you don't know what the hell you're talking about"), and from
Herb Taggart's delightful biographical note, which prefaces the volume.
You will be wholesomely reminded that obscure origin and straitened
circumstances are no bar to achievement, and that success in life is not
reserved for those who took their early nourishment from a silver spoon.
For Paton's eminence was the fruit of a career in the true Horatio Alger
tradition. And if you are like those who once voiced doubt that Nazareth
was a place out of which any good thing could come, you might reflect
on what could reasonably be expected to emerge from the remote and
meagerly endowed village of Calumet (pop. 1,959) where our hero began
life. Or on what might result from an upbringing on a central Michigan
farm, of the kind where whole families subsisted on cash incomes of $400
a year but no one was on relief.
Perhaps there's something special in the Michigan climate. The sleepy
village of Montague (pop. 2,356), near which I once spent happy sum¬
mers, eventually produced a Miss America, and a very charming one at
that. Consider this a corollary of the fact that an isolated Copper Country
ghost town inaugurated the career of the guy who to many of us will
always be Mr. Accounting, for his enrichment of our knowledge, inspira¬
tion of our spirit, and gladdening of our dullest days.

Howard C. Greer
Lafayette, California
Observations on Internal Controls
Carl T. Devine

Overview

This essay begins with a lament over the lack of professional

interest in the behavioral aspects of internal controls and reviews
briefly some related literature in this area.1 Accountants have not
constructed indexes, designed measuring scales, or emphasized the
psychological barriers that such controls are designed to induce.
Attention is then directed to some decision models that might be
appropriate for the initiator of a violation proposal and for those
who are approached. The initiator is faced with probabilities of
having his proposal rejected, and, if rejected, with further prob¬
abilities that rejection will or will not lead to disclosure. These
possibilities are in addition to the usual chances of being successful
in carrying out plans if the proposal is in fact accepted. Finally,
some sort of bargaining model for sharing the expected gains is
necessary unless the custodian's strategy is to steal and run and
thus short-circuit the usual collusion requirement. The discussion
then turns to the controller, who is faced with his own set of ma-
nipulatable variables, his own probability assessments, and his
firm's cost-benefit situation.

1. For an interesting historical survey of early internal-control procedures see T. A.

Lee, "The Historical Development of Internal Control from the Earliest Times to the
End of the Seventeenth Century," Journal of Accounting Research (Spring 1971): ISO-
57.

5
6 Carl T. Devine

While accountants have long assumed that the separation of func¬

tions and the necessity for collusion are important aspects of inter¬
nal control, they have devoted little direct effort to assessing the
relative effectiveness of barriers introduced by traditional internal
controls. For example, how much is added to the strength of the
deterrent by the introduction of a third party, e.g., an auditor, into
the cycle? Does it matter if the participants are blue-collar or white-
collar? Male or female? Rural or urban? Young or old? How effective
is physical separation? Social distance? Common reference and peer
groups? Is there more reluctance for bookkeepers to approach custo¬
dians than the reverse? What sorts of approaches are used to reduce
the initiator's probability of rejection and the anticipated penalties
associated with rejection?
The usual procedure, as most accountants know, is to select or
construct an internal-control questionnaire. The audit supervisor
often selects a standardized questionnaire and modifies it for the
peculiarities of his specific engagement. Precisely how are the origi¬
nal questions selected and how are they evaluated? How can any
questionnaire be constructed without some very specific organiza¬
tional and behavioral assumptions? Presumably auditors identify a
"representative" organization staffed with "normal" individuals
whose reactions and attitudes are sufficiently homogeneous to be
used as standards. Obviously there are some homogeneous features
even in the primitive human condition, and it seems reasonable to
assume that the degree of homogeneity may be modified by screen¬
ing and indoctrination.
The typical audit supervisor selects a standardized questionnaire
and modifies it to meet the peculiarities of his specific engagement.
What peculiarities of a particular situation lead to what kinds of
modifications? What subsequent observations serve as clues for
change? The carping here is not with the assumption of some de¬
gree of homogeneity among members of an organization—such an
assumption is inescapable. Nor is it implied that current profes¬
sional procedures are necessarily bad. On the contrary, conven¬
tional wisdom has been filtered and modified until the resulting
guidelines are applied with confidence and professional respect.
The criticism at this point is that no shareable process of judgment
formation has been disclosed and that the literature is almost en¬
tirely composed of dicta with little evidential support or systematic
discussion.
Our purpose in this essay is to suggest a framework for investi¬
gating the internal control area. Accountants clearly hypothesize
 Observations on Internal Controls 7

that sharing socially unacceptable goals is done with reluctance and

that the necessity for approaching others with an invitation to par¬
ticipate in forbidden activity introduces an important barrier. An
obvious professional task is to evaluate the relative strength of such
barriers and to manipulate the avenues that are available for in¬
creasing their effectiveness. The suggested framework may be fa¬
miliar to accountants, for it combines a traditional probabilistic
view of rejection and disclosure with a cost-benefit evaluation of
the consequences of success and failure. The conclusions are in
terms of expected values with both probabilities and evaluations
considered to be partially controllable by would-be embezzlers and
by organizational bureaucrats.

Some Things We Know About Thieves

In spite of important work by Sorensen and others on the sociol¬

ogy of the profession,2 it is ironical that we still know considerably
more about professional thieves than we know about the behavior
of accounting and custodial personnel.3
It is generally believed that the requirements of technical skill,
status, consensus, differential association and organization are
common to all professions including the activities of connivery and
thievery. The skills of successful thieves take the form of planning
and executing, providing for failure, and designing techniques to
evaluate diverse alternatives. Status follows from skill, financial
standing, connections, personal image, wide knowledge and gen¬
eral sophistication. The term thief among criminal elements is
honorific. Consensus results from ability to work together, recogni¬
tion of common enemies, adoption of a general esprit de corps, and
acceptance of a common well-defined value system.

2. For example, James E. Sorensen, "Professional and Bureaucratic Organization in

the Public Accounting Firm," The Accounting Review (July 1967): 553-65; and James
E. Sorensen and Thomas L. Sorensen, "Comparison of 1965 and 1970 Organizational
and Professional Profiles and Migration Plans of Large-Firm CPAs," in Behavioral
Experiments in Accounting, Thomas J. Bums, ed. (Columbus: The Ohio State Univer¬
sity, 1972), pp. 3-38.
3. The leading reference here is The Professional Thief, by Chic Conwell, annotated
and interpreted by E.H. Sutherland (Chicago: University of Chicago Press, 1937). For
a discussion of restricted role behavior to which only members of the confidence
racket have access see D.W. Maurer, The Big Con (Indianapolis: Bobbs-Merrill, 1940).
Numerous semi-popular books are available about stock manipulation, white-collar
thievery, swindling, etc.
8 Carl T. Devine

Differential association results from ability to define requirements

of membership and to contact members of legitimate society in
order to steal from them.4 Thieves have organization in that they
employ their own language, laws, history, traditions and customs.
They establish specialized machinery for disposing of property and
procuring defenses. It is good business for professional thieves and
con men to shine their shoes, be pleasant company, and become
engaging conversationalists.
Is the sociology of accounting—as a profession—as well defined?
We have scarcely started the necessary investigations and conse¬
quently we simply do not know.

Some Items About Embezzlers

A study of embezzlers by Cressey is clearly one of the more useful

studies in the behavioral foundations for accounting. Cressey
wonders why "some persons in positions of financial trust violate
it.''5 In the course of his inquiry Cressey emphasized the following
conditions and variables.

Trusted persons become trust violators when they conceive of themselves

as having a financial problem which is non-shareable, are aware that this
problem can be secretly resolved by the violation of the position of finan¬
cial trust, and are able to apply to their conduct in that situation verbal¬
izations which enable them to adjust their conception of themselves as
users of the entrusted funds or property.6

Cressey's attempt to identify the determinants of such attitudes is

not entirely satisfactory. Accountants, especially, may wonder
whether he overemphasizes secret solutions and non-shareability of

4. With the typical sociologist's contempt for business (and politics), Sutherland
points out that the intimacy is usually cold-blooded. "He [the thief] is like a sales¬
man who attempts to understand a prospective customer only as a means of breaking
down sales resistance." Ibid., p. 277. Training in salesmanship deals exhaustively
with different approaches and possible responses, and stresses the search for infor¬
mation under uncertainty. Even the broader social aspects of trust violation may be
included. Unfortunately we accountants still know very little about the actual ap¬
proaches used by parties in an internal control collusion so that they can better
"understand" one another.
5. Donald R. Cressey, Other People's Money (Glencoe: The Free Press, 1953), p. 30;
see also Cressey, "Embezzlers, The Trusted Thieves," Fortune (November 1957) and
"Why Do Trusted People Commit Fraud? A Social-Psychological Study of Defalca¬
tors," Journal of Accountancy (November 1951).
6. Ibid., p. 30 ff. Observe that if such problems are in fact unshareable, ordinary
internal controls would not be effective.
 Observations on Internal Controls 9

experiences as key variables. Cressey feels that the shareability

variable can be reduced to status, and, in something of an oversim¬
plification, he relates status directly to financial conditions.7 White-
collar workers may find disclosure of gambling losses to be non-
shareable but find disclosure of business losses entirely acceptable.
(Needless to say other groups may find gambling losses highly
shareable and appropriate for extensive conversation!) The tradition
of being noble and going down with the ship may set up stresses
and lead to violation as well as provide material for subsequent
rationalization.8
With respect to ability to carry out violations, Cressey found lack
of knowledge to be relatively unimportant. Television, motion pic¬
tures, radio, newspapers, gaudy paperbacks, often suggest feasible
methods and possible defenses. The violators in his sample were
already near the resources, could hardly fail to recognize available
opportunities, and had little need for collusion. Thus the search for
additional information for decision making under uncertainty was
not an important part of Cressey's investigation. His subjects were
almost entirely upper middle-class executives so that differences in
their utility functions and value structures were not points for in¬
tensive study. Most violators were influenced by their peer-group
morality, were "sorry" even before they were caught, and, when
convicted, were reluctant to accept the role of common criminal.

About Auditors and Motivators

In view of obvious attempts by the auditing profession to estab¬

lish an image of trustworthiness, sagacity, dependability, and fear¬
less honesty, little seems to have been done to research either the
positive or deterrent effects of an audit.9 Churchill and Cooper have

7. A more careful association of status with peer-group influences and financial

condition relative to group norms appears to be a more productive orientation for
further research. Extreme loners, for example, may have few people with whom to
share, may be less status conscious, and have weaker reference-group influences. He
apparently assumes that loners are not likely to be in critical positions of trust.
8. The existentialist movement, to its credit, has undermined the persuasiveness
of some once very effective rationalizations: It was only natural! After all I am a
human being! I am a product of my culture! The devil made me do it! Zeus did it to
me!
9. Raymond A. Bauer has been interested in the relation of belief to source credi¬
bility and has called attention to an unpublished dissertation by Murray Hilibrand
(1964), Source Credibility and the Persuasive Process. See Raymond A. Bauer, "Applica¬
tions of Behavioral Science," Applied Science and Technological Process, June 1967.
10 Carl T. Devine

certainly made the most important empirical contribution.10 Cer¬

tain graduate students and industrial workers served as subjects,
and accordingly the authors were reluctant to generalize widely.
They found that the audit occurrence apparently exerted some influ¬
ence even when not used in conjunction with an explicit reward-
penalty system.* 11 The direction of the influence was toward confor¬
mance with accepted accounting principles. Apparently most sub¬
jects assumed that the auditors would do what conventional wis¬
dom assumes they will do—probe behind the documents to the
associated events. Anticipation of future audits seemed to support
traditional behavioral beliefs, and the threat of an audit had some
effectiveness as a crude motivational club.
Consider now some more recent discussions of auditing and sys¬
tems work in the area of internal controls. Until recently the former
have been concerned primarily with statistical aspects of the prob¬
lem, and to a large extent the latter have simply neglected the matter
or have concentrated on broad control systems and computer-
oriented controls. An initial breakthrough toward acceptance of au¬
diting as a university endeavor came with the incorporation of mod¬
ern statistical theory. Matters have greatly improved and full accept¬
ability may now have arrived with an entire supplement of the Jour¬
nal of Accounting Research devoted to Studies in Statistical Methodology
in Auditing (1975).12 A part of this supplement is concerned with

Reprinted in the Harvard University Graduate School of Business Administration

Reprint Series. For more general research in this direction see Carl I. Hovland, A.A.
Lumsdaine, and F.D. Sheffield, Experiments on Mass Communication (Princeton:
Princeton University Press, 1949). For a useful summary and critique of Hovland and
his various associates see A.R. Cohen, Attitude Change and Social Influence (New
York: Basic Books, 1964).
10. N.C. Churchill and W.W. Cooper, Effects of Auditing Records: Individual Accom¬
plishment and Organization Objectives (Pittsburgh: ONR Research Memorandum No.
98, Camegie-Mellon Graduate School of Industrial Administration).
11. Further investigation of implicit reward-penalty perceptions might prove inter-
esting. Specific fear of the system? Generalized anxiety? Unusually compliant citi¬
zenry? Desire to join the establishment?
12. The pioneer work is Lawrence Vance, Scientific Method for Auditing (Berkeley:
University of California Press, 1947). The research was apparently done for his doctor¬
al dissertation at Minnesota in 1942. Writers in the intervening period were, with
few exceptions, concerned with statistical rather than behavioral applications; R.
Gene Brown, "Objective Internal Control Evaluation," Journal of Accountancy (No¬
vember 1962): 50-56; John A. Tracy, "Bayesian Statistical Methods in Auditing," The
 Observations on Internal Controls 11

internal controls and decision models that include cost-benefit

analysis for determining whether further investigation is war¬
ranted. Normative decision theory may not be the final answer to
all mankind's worries, but it is applicable to auditing and it prom¬
ises substantial help to this branch of our troubled profession.
Non-trivial decisions involve choices, choices involve values, and
expected values require predictions. Auditing at all levels is con¬
cerned with alternative procedures and with the extent of detailed
work to be done. These matters require predictions, value assess¬
ments, and cost-benefit computations.13
An encouraging development from this situation is the attraction
of competent decision theorists to the problems of internal control.
A further bonus may be in the form of meaningful research cover¬
ing the precise interactions of internal controls and the amount of
auditing to be done. The literature dealing with applications of
Bayesian and stochastic processes is a tentative start in this direc¬
tion, but mechanics alone are not going to solve these interaction
problems. Our profession encounters such situations on a daily
basis. Sooner or later someone must undertake the behavioral
studies necessary for implementing these decisions and communi¬
cating the process to new members and to the financial world.

Accounting Review (January 1969): 90-98; James E. Sorensen, “Bayesian Analysis in

Auditing," The Accounting Review (July 1969): 555-61; William R. Scott, A Bayesian
Approach to the Accounting Problems of Asset Valuation and Audit Size (Ph.D. Disser¬
tation, University of Chicago, 1972); John C. Corless, "Assessing Prior Distribu¬
tions for Applying Bayesian Statistics in Auditing," The Accounting Review (July
1972): 555-66; Robert S. Kaplan, several articles, especially "A Stochastic Model for
Auditing," Journal of Accounting Research (Spring 1973): 38-46; Seongjae Yu and
John Neter, “A Stochastic Model of Internal Control Systems," (Working Paper,
University of Illinois, December, 1973); Barry E. Cushing, "A Mathematical Ap¬
proach to the Analysis and Design of Internal Control Systems," The Accounting
Review (January 1974): 24-42; Robert H. Ashton, "An Experimental Study of Inter¬
nal Control Judgments," Journal of Accounting Research (Spring 1974): 143-57; Wil¬
liam L. Felix, Jr., "Evidence on Alternative Means of Assessing Prior Probability
Distributions for Audit Decision Makers," The Accounting Review (October 1976):
800-7.
13. In this context see William R. Kinney, Jr., "A Decision-Theory Approach to
the Sampling Problem in Auditing," Journal of Accounting Research (Spring 1975):
117-32, and "Decision Theory Aspects of Internal Control System Design Compli¬
ance and Substantive Tests," Journal of Accounting Research, Supplement 1975: 14-
29.
12 Carl T. Devine

Internal Controls—Employee View

Turn now to some attitudes and problems of individuals who are

restricted by the effective operation of internal controls. To an impor¬
tant extent their problem is one of seeking information under condi¬
tions of uncertainty.14 The approachor must be concerned with the
probability that his invitation will be rejected. Moreover, if he is
rejected, he must evaluate the chances of having his actions reported
to authorities. If accepted, he must assess the further probabilities of
apprehension after collusion. Such evaluations lead naturally to the
assignment of values to possible outcomes and is related to appropri¬
ate reward-penalty systems. The approachee need not be concerned
with rejection probabilities, but he does need to consider possibili¬
ties of being caught, and, if caught, of being punished. The point
here is that many of these probabilities and conditional penalties
may be modified by appropriate behaviors, preliminary screening,
cautious approaches, and various buddy-forming devices.
Observe that custodians have access to the desired resources and
are faced with problems different from those faced by clerical
workers. For one thing, it is usually their assignment to circumvent
the physical controls surrounding the resources. There is always
some probability of lack of success in this activity, but the chance
that clerical colluders will be drawn into any resulting penalty situa¬
tions may be considerably less. If the custodian is a drifter or a
loner, he may prefer to take his chances at circumventing the physi¬
cal controls without help, and, if successful, simply abscond. In
such cases he should assess the probabilities of his being able to
change identity and keep clear of the hounds from bonding firms
and policing agencies. The resulting expected values should then be
compared with mathematical expectations from alternative courses
of action.
There are obvious reasons why custodians may opt for collusion
rather than for the steal-and-run policy. Tactics become especially
important for the selection of alternative approaches, and consid¬
erable cost may be justified in seeking information about potential
approachees. Additional information may decrease substantially the
probability of rejection as well as the probability of being reported

14. Erving Goffman was among the early behavioral scientists to emphasize the
necessity of searching for information about those with whom one must interact, the
influence of the overall situation, the importance of physical regions, and the concept
of Performance Teams. The Presentation of Self in Everyday Life (Garden City: Double¬
day Anchor Books, 1959: 76ff).
 Observations on Internal Controls 13

if rejected. If rejected and not reported, the custodian may still

retain the option to steal and run, while rejection and disclosure
may end this opportunity. The steal-and-run option should be a
lower-valued alternative not only because of possible future police
action, but also because it virtually destroys the opportunity for
continued embezzlement in the same organization. Keeping the
option open for repeated violations without the necessity and
danger of approaching other potentially hostile colluders and learn¬
ing new physical safeguards may justify the cost of extensive search
for additional information.
While the penalties for apprehension may be enormous, the ab¬
sconder has one factor working for him. He does not have to share
the spoils with his partners in collusion. If sharing becomes neces¬
sary, the process may be approximated by an ordinary two-party or
three-party bargaining model with interesting complications.
Clearly the bargaining and the probability of rejection and disclo¬
sure may not be independent, i.e., such probabilities may be influ¬
enced by the bargaining for the division of spoils.
For the above reasons we might advance a preliminary hypothe¬
sis that collusive inquiries will be initiated more frequently by
clerical workers and directed toward custodians. We might specu¬
late further that professionally trained custodians, say treasury per¬
sonnel, will be more likely to reject and report invitations than
custodians with less professional indoctrination, e.g., storeroom
and toolcrib attendants. In addition the greater guilt sometimes
associated with stealing money rather than other resources may
tend to make cash controls more effective. Observe, however, that
while cash may be more responsibly guarded due to such an atti¬
tude, currency can be disposed of more easily and with less prob¬
ability of discovery.
To a potential initiator the population of possible approachees is
certainly not homogeneous. Long sustained image construction
may mean that auditors as a group are perceived as high-risk
approachees, and it is possible that the training they receive on
the job and in universities does increase the risk and make the
group more homogeneous. Employees in the treasury may also
view themselves as protectors of financial integrity and thus ap¬
pear as high-risk targets. Without specific knowledge one wonders
how systems designers actually design their control systems. What
sorts of actions do colluding clerks make to reduce probabilities of
detection by auditors? How often do internal auditors use their
superior-inferior position to suggest proposals to clerks or directly
14 Carl T. Devine

to custodians? How do external auditors influence various groups

and reinforce various barriers? What about invitations from
middle and lower managers to subordinate employees?
A study of the techniques used by initiators to reduce probabili¬
ties of rejection and disclosure should prove to be useful. In spite of
indoctrination and some measure of group discipline individual
members are not homogeneous, and any thoughtful initiator should
be able to reduce his risk by incurring some costs to identify those
at the more cooperative tail of the distribution—the weak links.
Once these members have been tentatively spotted, further costs to
investigate techniques and methods appropriate for each case may
be justified. The investigator may find well-developed buddy paths
to potentially cooperative members in target groups. But how are
such paths developed? Which directions do they run? What tech¬
niques are used to undermine organizational and professional loy¬
alty? Do approaches develop by degrees so that retreat before final
rejection remains possible? Does pretense of a joke lessen the prob¬
ability of disclosure? What surrounding circumstances permit
graceful exits?
Potential embezzlers should be aware of mechanical weak points
in the general internal-control system and should be able to assess
probabilities of success. Experienced clerical personnel are usually
aware of undetected errors and shortages that have not been dis¬
closed by previous audits and other control devices. Attitudes of
superiors to such failings are normally noted. Meanwhile, custodi¬
ans should have knowledge of existing physical controls and be
able to appraise the probable success of circumventing them.
There should be a pooling of knowledge about the possibility of
being ensnared by physical and internal devices after the collusive
effort is under way. Gatekeepers and others may not remain per¬
suaded, become victims of panic, or be swayed by pangs of con¬
science. Chance factors and unforeseen investigations are potential
dangers. Changes in personnel, leaks from unskilled collaborators,
undependable fences, and organization spies are further dangers. In
general these factors are common to all approaches and are present
regardless of the particular techniques used to circumvent specific
internal controls. There may be, however, considerable interdepen¬
dence, for not all colluders are equally skilled and existing physical
controls certainly do influence the effectiveness of the internal con¬
trols. How shall we measure the importance of this interaction?
Consider for a moment cases in which approaches are made from
clerical to custodial groups. Since clerical workers do not have direct
 Observations on Internal Controls 15

access to the resources, the decision to abscond is not usually an

original option or a possible retreat from rejection. It may therefore
be omitted from their decision networks. Otherwise the procedures
are likely to be similar. Different value systems and different abilities
to assess probabilities may be involved. Buddy-path formation may
have different characteristics and different techniques may be re¬
quired. In each case, the initiator needs information to help him
select potential collaborators, choose appropriate approaches, ap¬
praise probabilities, arrive at acceptable sharing formulas and assess
payoffs generally. Further, the need for multiperiod knowledge is
clear. Many situations require estimates of optimal amounts to be
embezzled in each violation, the optimal frequency of violation, op¬
timal timing of attempts, and possibilities for varying the methods to
be employed. Moreover, the wise embezzler should develop alter¬
nate and sequential plans in case original schemes turn out to be
unsuccessful. Some sort of leadership arrangements will need to be
forged along with rules for assigning tasks and responsibilities.
Means for incorporating new information as the process is imple¬
mented should be provided, and alternative defensive and offensive
measures must be established when outcomes diverge from plans.

Suggestions for Decision Analysis

It is not difficult to construct a highly simplified decision tree for

an initiator of an invitation to violate trust. The usual technique for
applying such models is to assign probabilities for each branch and
to assess conditional rewards and penalties for the resulting states.
Presumably the line with the highest mathematical expectation—
highest expected value—can be defined as the rational choice. The
decision tree for an initiator from recordkeeping is similar except for
the opportunity to abscond. A decision framework for the approachee
does not involve the need for an initial proposal with probabilities of
acceptance or rejection, but it should provide for the alternative of
rejection with possible rewards for reporting the initiator.
From a behavioral viewpoint the problem facing possible partici¬
pants may be viewed as a problem of information seeking and evalu¬
ating under uncertainty. Inquiry may be directed to the kinds of
behavior that should be employed to procure the desired informa¬
tion at reasonable cost. More generally the problem can be structured
as a balancing of cost and benefit assessments in a complicated inter¬
personal situation. At a more specific level research can be concen¬
trated on efforts to change the probabilities of acceptance, detection,
16 Carl T. Devine

disclosure, etc., and to increasing the conditional rewards. To be

useful, therefore, a decision framework or diagram must provide for
search at each traditional decision point. The options are thus not
simply two-way branches. The desirable (optimal) amount of search
must be decided in terms of modifications of probabilities and con¬
ditional payoff assessments all along the line. The simultaneous in¬
teraction is obvious, and the need for severe simplification is clear.
In short a cost-benefit search should be made at each decision
point in order to bring about acceptable probabilities and condi¬
tional rewards. For the initiator these search activities may be de¬
signed to bring about more favorable expectations in the following
ways.
1. Decrease the probability of rejection, P(R);
2. Decrease the probability of being turned in if rejected, P(T), and for
the case of interdependence, decrease P(R^T) = P(R) • P(TIR) = P(T) •
P(RIT);
3. Decrease the probability of being discovered by auditors or other
internal-control workers who are not in the collusive agreement, e.g.,
optimal timing, number of violations;
4. Decrease the probability of being caught by enforcement agencies,
i.e., selecting competent colluders, discreet fences, effective escape
methods;
5. Influence the reward-penalty system: decrease conditional penalties,
and increase conditional rewards, e.g., finding optimal amounts for
violations, finding ways to soften the penalties.

It is possible that the techniques of game theory are appropriate

for analyzing problems of internal control. Clearly the division of
rewards from successful collusion for both the approachor and the
approachee depends in part on relative bargaining strength. After
the fact, what one gains, the other loses so that the first appearance
is that of a zero-sum game. In a more general context the zero-sum
format does not seem to be appropriate, because if one colluder is
caught, the others also lose to varying degrees. Thus the game has
cooperative features along with the zero-sum aspects of the sharing
model, and apparently there are at least two games to be given
attention even for a single pair of participants.15

15. Those interested in related aspects of this problem might refer to C.B. McGuire
and Roy Radner (editors). Decision and Organization (Amsterdam: North Holland,
1972), especially Radner's chapter, “Normative Theories of Organization: An Intro¬
duction." For an early, non-technical discussion of bargaining models generally see
Lawrence E. Fouraker and Sidney Siegel, Bargaining Behavior (New York: McGraw-
Hill Book Company, 1963).
 Observations on Internal Controls 17

Some further influences on the bargaining parties may deserve

investigation. Does the initiator have a superior bargaining posi¬
tion because of his demonstrated daring, ingenuity or leadership
qualities? Perhaps his position is inferior due to his vulnerability to
rejection and disclosure. One might speculate that the custodian is
usually in a superior position since he enjoys the option of ab¬
sconding. Yet the clerical worker may be able to endure discovery
with less serious consequences by creating the impression, if
caught, of a mistake—a human error. It may be less difficult for the
bookkeeper to explain his covering entries than for the custodian to
explain the presence of organizational resources at his country es¬
tate. Bookkeepers, clerks, and auditors may be in an inferior posi¬
tion, for they usually do not have access to the resources and collu¬
sion is essential in order for them to embezzle at all.
Finally, one may wonder whether the positions of the controller
(as a representative of the organization) and the colluders (as a
coalition) form the basis for a zero-sum game. Even simple zero-
sum games run into conceptual problems from differences in sub¬
jective utilities among participants, but the present situation is con¬
siderably worse. There is no reason to believe that the gain or loss
to the colluders will be exactly offset by the loss or gain to the
organization or to the particular corporate officers involved. The
possibility of prison sentences, loss of prestige and employment are
peculiar to the coalition members, and, while the monetary loss
may be balanced by an equal monetary effect (except for fencing
discounts, etc.), only by chance would the corporation enjoy mea¬
surable gains from an effective control system exactly equal to these
intangible losses of a particular set of coalition members.

Controller's Viewpoint

The controller's decision process may be approached in a similar

manner. In order to perform his functions the controller must be
able to predict the behavior of those subject to his controls. A num¬
ber of decisions and activities are usually available, and it is as¬
sumed that he will select alternatives on some sort of expected-
value, cost-benefit basis. These activities may be summarized
briefly.

1. Designing social factors to decrease the probabilities that the inter¬

nal control personnel will get together and establish channels of rap¬
port—age, sex and religious differences, social distances, status
barriers;
18 Carl T. Devine

2. Increasing the probabilities of subsequent disclosure through clerical

devices—internal and external auditors, internal checks;
3. Supporting actions to increase probabilities that embezzlers will sub¬
sequently be apprehended—additional physical safeguards, coopera¬
tion with security forces;
4. Shifting the reward-penalty structure—decreasing the availability of
valuable resources, increasing penalty recommendations.

In order to improve these decisions and implement these activi¬

ties the controller may procure help from the literature of human
engineering, personnel psychology, organization theory, and from
the general literature of the behavioral sciences.
Human engineers, in somewhat of an oversimplification, attempt
to design the conditions of work and rearrange the relevant world
to fit the man. Human nature, while not completely invariant, is
assumed to be highly inflexible, and traditional engineering holds
that it is easier to change the conditions of work than to change the
attitudes and responses of mankind.16 While it is not necessary for
the controller to subscribe to an extreme view, he should realize
that management may be able to ease the task by considering the
overall work situation, seeing that opportunities for collusion and
embezzlement are seldom available, and adjusting penalties and
rewards to serve as effective deterrents. Physical separation of cleri¬
cal and custodial personnel is often employed, although few have
attempted to assess the strength of collusive barriers in terms of
physical separation, factory layouts, room dividers, and ordinary
spacial measures. Separation of function often leads to manager-
worker, white-collar, blue-collar, and related differentiation, and to
the development of status barriers and social segmentation. For
example, different functions ordinarily demand different educa¬
tional requirements and thus generate differences in language hab¬
its, social graces, recreational customs, etc. Empirical members of
our profession should be able to scale such influences and perhaps
map them into some workable concept of deterrence. Our present
knowledge is meager indeed!
The literature of personnel psychology may offer suggestions in
the areas of employee selection, training, motivation, indoctrination

16. Interestingly, in spite of technological overtones, the position of human engi¬

neers is essentially that of traditional liberalism and is humanistic rather than ideo¬
logical. Frank H. Knight states, "... liberalism takes the individual as given, and
views the social problem as one of right relations between given individuals," Free¬
dom and Reform (New York: Harper & Bros., 1947): 69.
 Observations on Internal Controls 19

and related activities. The simplified message here is that it is usu¬

ally easier to select and modify individuals to fit the job than to
redesign the environment to fit stereotyped personalities.17 Con¬
trollers are, of course, familiar with the general recommendations of
this group. They have attempted to find predictors from back¬
ground and attitudinal variables to help select and retain employees
who fit relatively inflexible job requirements. Hiring is followed by
various training and indoctrination programs. For higher-level
clerical work the indoctrination has already begun in the usual uni¬
versity courses. Unfortunately clerks, low-level bookkeepers and
store handlers receive much less training of this type so that this
pool of potential hirees tends to be less homogeneous. Education
for the specific role requirements of each job is ordinarily the re¬
sponsibility of the hiring organization. Employees are often encour¬
aged to join professional groups in order to hasten professional role
development and to make up suspected deficiencies in previous
training. Association with various service and religious organiza¬
tions may help form desirable attitudes, and the usual procedures
for raises, promotions, title and job enlargement, etc., may be use¬
ful reinforcement devices.
The literature of the somewhat new school of social (organiza¬
tional) psychologists may prove to be helpful. The emphasis of this
group is on rearranging the structures and dynamics of pertinent
groups to accommodate individuals who have some degree of flexi¬
bility but tend to retain many of their own values and personal role
characteristics. The speed and strength of goal internalization are
obviously major topics of interest to controllers. These variables
along with notions of role modification, identity formation, attach¬
ment to semantic symbols, reference group influence, personality
integration and related ideas from social psychology merit serious
consideration.18

17. The extreme is represented by the ideologists—the rigid sociology-of-

knowledge advocates. This view is the accepted position of totalitarians of both the
left and the right. More mellow advocates stress the selection process and tend to feel
comfortable with the reinforcement hypotheses of B.F. Skinner and his followers. For
his original position see The Behavior of Organisms (New York: Appleton-Century-
Crofts, 1938).
18. The literature in this area is too extensive to summarize here. Published
studies range from prisoner rehabilitation, soldier indoctrination, and waitress train¬
ing to the necessary goal displacement of scientists engaged permanently in govern¬
ment research. Inflexible role transference is illustrated by the inability of many
revolutionary leaders to adapt to the requirements of mundane administration and
the tendency of passivist resistors to continue to retard economic development long
20 Carl T. Devine

To summarize these considerations, we offer the general specula¬

tion that the probabilities of being rejected and reported are related
to the level of group indoctrination, to the homogeneity of group
membership, to perceptions of rewards and penalties, and to the
persuasive ability of initiators. Like potential embezzlers, auditors,
system builders and controllers utilize a modified cost-benefit
analysis with various degrees of control over many variables that
influence employee behavior. Certain variables are manipulated in
the context of internal controls to influence the actions of human
beings whose own goals often differ from those of the host orga¬
nization. In this situation, unlike the area of physical controls, the
appropriate wisdom and knowledge is likely to come from psy¬
chologists, sociologists and various combinations such as social
psychologists and organization theorists.

Some Possibilities for Research

In general terms the professional problem in this area is to assess

the deterrent effect of all sorts of physical and group arrangements
against all sorts of background variables and contrived modifica¬
tions. Continuing discussions by our lawmakers over such topics as
the effectiveness of capital punishment and harsh drug laws attest
to the difficulties of forecasting the deterrent effects of rule-making.
While the problem may be difficult to specify and impossible to
measure, it arises in our profession and in our society every day,

after their countries are free. The outstanding work for a general view is George C.
Homans, The Human Group (New York: Harcourt, Brace, 1950). Some useful exten¬
sions in organization theory may be found in James G. March and Herbert A. Simon,
Organizations (New York: John Wiley & Sons, Inc., 1958).
Many psychologists have been concerned with roles, reference groups and person¬
ality factors. George H. Mead emphasized personality development in terms of signs
that require adaptive responses from others and the importance of membership
groups. (Mind, Self and Society, University of Chicago Press, 1934). Robert K. Merton
made improvements in the concept of membership groups and earlier reference-
group theory. Merton was also interested in objectives, legitimate and institutional¬
ized means of attaining objectives. (Social Theory and Social Structure, Free Press,
1957). Leon Festinger has been a leader through his "urges" to compare beliefs, in
order to be consistent with group attitudes, and to reduce dissonance. He also gave
limited support to the hypothesis that smaller rewards may be more effective than
larger ones. See Conflict, Decision and Dissonance (Stanford University Press, 1964).
Earlier writings include Festinger and J.M. Carlsmith, "Cognitive Consequences of
Forces Compliance," Journal of Abnormal and Social Psychology (1959): 203-10.
 Observations on Internal Controls 21

and every day it must somehow be met. Implicit solutions may not
be optimal, but they obviously must be acceptable to some impor¬
tant segments of society.
In the specific field of internal controls there is an obvious lack of
knowledge about the reluctance (deterrence) engendered by the ne¬
cessity for collusion. The following summary is only suggestive.

1. Influences of pre-professional associations and groups—church mem¬

bership, service clubs, parental attitudes, university course work;
2. Images of groups involved in internal-control circles—including the
degrees of group homogeneity;
3. Interaction of physical deterrents and internal-control devices;
4. Influences of social distance and possible tradeoffs in various direc¬
tions and at various levels;
5. Types of approaches used with variations from group to group;
6. Data on the directions invitations for violation of trust tend to take;
7. Sharing models used and relative bargaining strengths of each group
under varying conditions;
8. Ways in which standardized questionnaires are constructed;
9. Mechanics for trade-offs of audit completeness and internal-control
sufficiency;
10. Interpretation and modification of standardized questionnaires to
meet particular situations.

Meaningful research in these areas is clearly difficult and may

require the help of competent research professionals. Yet accoun¬
tants are capable of carrying on some less demanding investigations
along with many specialized inquiries within the profession. One
less demanding technique is based on ordinary utility analysis with
appropriate subjects being asked to adjust probabilities or condi¬
tional values. With a given reward for approaching some represen¬
tative member of say the auditing group, the subject may be asked
what reward would make him indifferent to approaching some
typical member of another group in the internal-control circle. He is
in effect asked to equate various expected values. Differences may
then be used to construct measures of relative reluctance to ap¬
proach members of each group involved. Intermediate and further
steps may be designed to relate these differences to measures of
rejection probabilities, turn-in probabilities, and perhaps prob¬
abilities of apprehension. Further questions may be designed to
probe possible reasons for their judgments and to help isolate the
determinants of their opinions. If the latter step is effective, the
22 Carl T. Devine

profession may get direction for planning both its control situations
and its educational programs.
The obvious weakness in the above recommendation is that most
proposals for trust violation undoubtedly travel from weak link to
weak link rather than from average member to average member. It
may be useful therefore to question subjects about their attitudes
toward other members of their own internal-control groups. These
questions may be designed primarily to evaluate the degree of ho¬
mogeneity within groups. Thus a member of the bookkeeping
group may be asked to consider a conditional reward and to think
of the person in his own group whom he would be least (or most)
reluctant to approach. Next he may be asked to place a relative
figure on his reluctance to approach a group member selected at
random. The purpose, of course, is to get some feel for the homoge¬
neity of the group. Intergroup perceptions also require testing.
Probably the simplest approach is through a questionnaire de¬
signed to test differences in reluctance. A simple incremental ap¬
proach might present unidentified individuals to the subject group.
Subjects may then be asked to assess their own reluctance to deal
with these individuals in matters of trust violation. After evaluating
the responses the experimenter may offer additional information
about the individuals, e.g., that they are auditors, cashiers, store
clerks. Reevaluations may then be requested. Hopefully the evalua¬
tive differences will be influenced by the additional information.
One advantage of this variation is that all sorts of background vari¬
ables and attitudes may be appraised and relative values assigned.
Other characteristics of each individual may be relatively stable so
that differences in responses may be associated with the additional
role requirements of the position. Incidentally, possibilities for ask¬
ing the subjects to consider all sorts of sequential bits of information
about group stereotypes may prove to be useful. The ordering re¬
quirements imposed on the subjects for simple transformations of
this kind should not prove to be too difficult or impossible.19
Some subjects may perform erratically due to the moral burden of
involvement in even a hypothetical embezzlement. This burden
may be more than they can accept and still be effective subjects.
Such subjects may be excused, but perhaps they can be made into
effective participants by simply stipulating that they approach the

19. These suggestions seem to be in line with the vague Braybrooke-Lindblom

concept of "disjointed incrementalism." See A Strategy of Decision (New York: The
Free Press, 1963), esp. Chapter 5.
 Observations on Internal Controls 23

situation as if they were already irrevocably committed to the re¬

quired violation. A simple statement that the subject through no
fault of his own has been subjected to an intolerable stress and that
violation of trust is his only possible course of action may be effec¬
tive. This stipulation should decrease uneasiness about revealing
his individual character, and it should direct attention to the vari¬
able of concern—the relative reluctance and the relative strength of
psychological barriers.

December 1976
The Politicization of Accounting
David Solomons

There was once a time, not so many years ago, when accounting
could be thought of as an essentially nonpolitical subject. If it was
not as far removed from politics as mathematics or astronomy were,
it was at least no more political than psychology or surveying or
computer technology or statistics. Even in those areas of accounting
such as taxation, which might be thought to be most relevant to
questions of public policy, practitioners were generally content to
confine themselves to technical issues, without getting involved as
accountants in the discussion of tax policy.
Today, to judge from current discussions of the standard-setting
process, accounting can no longer be thought of as nonpolitical. The
numbers that accountants report have, or at least are widely
thought to have, a significant impact on economic behavior. Ac¬
counting rules therefore affect human behavior. Hence the process
by which they are made is said to be political. It is then only a short
step to the assertion that such rules are properly to be made in the
political arena, by counting heads and deciding accounting issues
by some voting mechanism.
There are several articulate spokesmen for this point of view. Dale
Gerboth writes that

a politicization of accounting rule-making [is] not only inevitable but

just. In a society committed to democratic legitimization of authority,
only politically responsive institutions have the right to command others
to obey their rules.1

1. Dale L. Gerboth, "Research, Intuition and Politics in Accounting Inquiry,"

Accounting Review (July 1973), p. 481.

25
26 David Solomons

And in another passage from the same paper, Gerboth says:

When a decision-making process depends for its success on public confi¬
dence, the critical issues are not technical; they are political .... In the
face of conflict between competing interests, rationality as well as pru¬
dence lies not in seeking final answers but, rather, in compromise—
essentially a political process.2

In the same vein, Charles Horngren writes:

The setting of accounting standards is as much a product of political
action as of flawless logic or empirical findings. Why? Because the setting
of standards is a social decision. Standards place restrictions on behav¬
ior; therefore, they must be accepted by the affected parties. Acceptance
may be forced or voluntary or some of both. In a democratic society,
getting acceptance is an exceedingly complicated process that requires
skillful marketing in a political arena.3

May and Sundem take a similar position:

... in practice as well as in theory, the social welfare impact of account¬
ing reports apparently is recognized. Therefore it is no surprise that the
Financial] Accounting] Standards] B[oard] is a political body and, con¬
sequently, that the process of selecting an acceptable accounting alterna¬
tive is a political process. If the social welfare impact of accounting policy
decisions were ignored, the basis for the existence of a regulatory body
would disappear. Therefore, the FASB must consider explicitly political
(i.e., social welfare) aspects as well as accounting theory and research in
its decisions.4

Other voices which call for an explicit recognition of the probable

economic and social impact of a new accounting standard are not
always easily distinguished from those who assert that political
considerations should determine what the standard should be.5
However, these two views should not be confused. I shall return to
this point later.
The Structure Committee of the Financial Accounting Foundation
grappled with the question of the political nature of the standard¬
setting task in their report on The Structure of Establishing Financial
Accounting Standards. On the nature of the standard-setting process,
they say:

2. Ibid., p. 479.
3. Charles T. Horngren, “The Marketing of Accounting Standards," Journal of
Accountancy (October 1973), p. 61.
4. Robert G. May and Gary L. Sundem, "Research for Accounting Policy: an
Overview," Accounting Review (October 1976), p. 750.
5. John Buckley, in "FASB Impact Analysis," Management Accounting (April 1976),
straddles this line most uncomfortably. His article has been thought to support
politically slanted standards, though he nowhere explicitly says that he does.
 The Politicization of Accounting 27

The process of setting accounting standards can be described as demo¬

cratic because like all rule-making bodies the Board's right to make rules
depends ultimately on the consent of the ruled. But because standard
setting requires some perspective it would not be appropriate to estab¬
lish a standard based solely on a canvass of the constituents. Similarly,
the process can be described as legislative because it must be delibera¬
tive and because all views must be heard. But the standard setters are
expected to represent the entire constituency as a whole and not be
representatives of a specific constituent group. The process can be de¬
scribed as political because there is an educational effort involved in
getting a new standard accepted. But it is not political in the sense that
an accommodation is required to get a statement issued.6

There is something here to please everyone. Yet the committee does

finally come out on the side of the angels.

We have used the word constituency to indicate that the FASB is ac¬
countable to everyone who has an interest. We are not suggesting that
the Board members are in place to represent them or that the standards
must necessarily be based on a numerical consensus.7

That accounting influences human behavior, if only because it

conveys information, is obvious enough, though research into the
workings of "the efficient market" has cast doubt on some of the
supposed results of accounting choices. There are, without ques¬
tion, political aspects of accounting. There are similarly political
aspects of physics, which result in enormous expenditures on re¬
search into nuclear energy and weaponry. Geology, in its concern
with the world's reserves of fossil fuels, obviously has political im¬
plications. Research into sickle cell anemia became a political ques¬
tion when the heavy incidence of this disease among black Ameri¬
cans came to light. There are very few areas of human knowledge
which are devoid of political significance. But that does not mean
that the processes by which knowledge is advanced or by which
new applications are found for old knowledge are themselves politi¬
cal processes, in the sense in which that term is usually understood.
Political motives for asking a question may be entirely appropriate.
A politically motivated answer may or may not be appropriate. It
obviously depends on the nature of the question.
It may be useful to look more carefully at the part which politics
should and should not play in accounting standard setting. The
future of the Financial Accounting Standards Board may depend on

6. The Structure of Establishing Financial Accounting Standards, (Financial Account¬

ing Foundation, April 1977), p. 19.
7. Ibid.
28 David Solomons

a better understanding of that issue. Indeed, the very credibility of

accounting itself may be at stake.

Accounting and National Goals

The most extreme expression, so far as I am aware, of the view

that political considerations should enter into the formulation of
accounting standards—not merely into the choice of accounting
questions to be studied, but also into the formulation of the stan¬
dards themselves—is to be found in a lecture given in New York in
November 1973 by Professor David F. Hawkins.8 Hawkins noted
that Congress and the executive branch of the federal government
were

becoming more and more aware of the behavioral aspects of corporate

reporting and its macro economic implications. Increasingly, I believe,
these policy makers will demand . . . that the decisions of those charged
with determining what constitutes approved corporate reporting stan¬
dards will lead to individual economic behavior that is consistent with
the nation's macro economic objectives,. . .This awareness on the part of
economic planners brings accounting standard setting into the realm of
political economics.9

Events since 1973 have not shown any diminution of this aware¬
ness. The question is whether this trend is to be regarded as a
threat to the integrity of accounting or as an opportunity for ac¬
countants, perhaps even an obligation, to cooperate with govern¬
ment in furthering its economic policy. Hawkins left us in no doubt
where he stqod in this matter.

The [Financial Accounting Standards] Board's objectives must be respon¬

sive to many more considerations than accounting theory or our notions
of economically useful data.10

And:

Corporate reporting standards should result in data that are useful for
economic decisions provided that the standard is consistent with the national
macro economic objectives and the economic programs designed to reach these
goals.11

8. "Financial Accounting, The Standards Board and Economic Development,"

one of the 1973-74 Emanuel Saxe Distinguished Lectures in Accounting (New York:
Bernard M. Baruch College, C.U.N.Y., April 1975).
9. Ibid., pp. 7-8.
10. Ibid., p. 17.
11. Ibid., pp. 9-10. Italics in the original.
 The Politicization of Accounting 29

As if that were not enough, he added that "because the Standards

Board has the power to influence economic behavior it has an obli¬
gation to support the government's economic plans."12
In that last-quoted passage, the word "because" is noteworthy,
implying as it does that the power to influence economic behavior
always carries with it an obligation to support the government's
plans. Even if the matter under discussion were, say, pricing policy
or wage policy or some aspect of environmental protection, the
assertion would be open to argument. In relation to accounting,
where the end product is a system of measurement, the posture
which Hawkins urges on the Standards Board could, I believe,
threaten the integrity of financial reporting and deprive it of what¬
ever credibility it now has.
There is no question as to the sensitiveness of some, indeed
most, of the issues which have been or are now on the agenda of
the FASB or its predecessors, and of course this stems from the fact
that standards dealing with these issues have influenced or will
influence behavior. One can only conclude that there is widespread
skepticism about the "efficient market" hypothesis. Financial men
are not at all indifferent to the accounting rules imposed on them by
the Standards Board, but it is not the purpose of this paper to
explore the nature of this concern.13 For our purposes it will be
enough to recognize that the FASB's constituents think it matters
whether leases are capitalized or not, whether foreign currency
transactions are accounted for by one method or another, whether
contingencies are provided for by charges against income or by
allowing retained earnings to accumulate. These questions do not
affect the amount of information that is disclosed but simply the
way in which these economic phenomena are reported; yet this fact
does not make them less sensitive. Perhaps investors really are
naive. Only on the basis of such an assumption (and on the as¬
sumption that no new information will be disclosed by a politically

12. Ibid., p. 11.

13. Yet one cannot ignore the troublesome paradox posed by the numerous em¬
pirical studies which have shown "that the capital market does distinguish between
[accounting] changes that appear to be reporting changes of no economic importance
and those that appear to have substantive economic implications.” (Nicholas J. Go-
nedes and Nicholas Dopuch, "Capital Market Equilibrium, Information Production
and Selecting Accounting Techniques: Theoretical Framework and Review of Empiri¬
cal Work," Studies in Financial Accounting Objectives: 1974, Supplement to vol. 12 of
the Journal of Accounting Research.) If the market can "see through" accounting
changes which result from changes in standards, why do they generate so much
heat?
30 David Solomons

motivated standard) is the impact of politics on accounting stan¬

dards worth discussing at all. ■

The Economic Impact of Accounting Standards

Few if any accounting standards are without some economic im¬

pact. The requirement that U.S. companies write off purchased
goodwill is said to give an advantage to foreign companies in bid¬
ding for American businesses because, not being subject to the
same accounting requirement, foreign firms can afford to offer a
higher price. FASB Statement No. 2, which requires research and
development to be expensed as incurred, has been said to consti¬
tute a threat to technological progress, especially for smaller compa¬
nies which may be planning to seek access to the capital market and
will therefore want to show good profits before doing so.14 Many
people believe that a ban on “full costing" by oil and gas companies
and a requirement that “field" or “successful effort" costing alone
be used will diminish investment to develop new sources of en¬
ergy, especially by smaller companies.15 FASB Statement No. 5 on
Accounting for Contingencies, by greatly restricting the circumstances
in which an estimated loss from a loss contingency can be accrued
by a charge to income, is said to have caused American insurance
companies to reinsure risks where they would previously have re¬
lied on self-insurance.
One of the most sensitive standards has been that dealing with
foreign currency translation (Statement No. 8). Under the so-called
temporal method mandated by the Standards Board, monetary as¬
sets and liabilities of a foreign subsidiary of a U.S. corporation have
to be translated, for consolidation purposes, at the rate of exchange
current at the balance sheet date. Assets which in accordance with
GAAP are carried at cost or cost less depreciation have to be trans¬
lated at the rate current at the time they were acquired. Exchange
gains and losses, realized and unrealized, have to be brought into
the income statement. For companies which formerly used a cur¬
rent-noncurrent classification, the important changes from their
former practice lie in the treatment of inventories and of long-term

14. This argument led Hawkins in 1973, when the treatment of research and
development was still on the FASB's agenda, to say, "l do not believe the Board can
eliminate the alternative capitalization.” Ibid., p. 14. Events proved him wrong.
15. See the FASB's Status Report No. 55 (October 14, 1977) on congressional sup¬
port for the retention of alternative treatments for oil and gas accounting.
 The Politicization of Accounting 31

debt. Inventories, as current assets, were formerly carried at the

current rate and are now carried at the historical rate. Long-term
debt, as a noncurrent liability, was formerly carried at the historical
rate; now, as a monetary item, it is carried at the current rate.
Moreover, unrealized translation gains, formerly kept out of the
income statement, now have to be brought in. The result has been
greatly to increase the volatility of the reported earnings of compa¬
nies with important foreign operations. Criticism of Statement No.
8 has focused on this increased volatility rather than on whether the
new rules result in a better or worse representation of financial
performance.
Numerous other politically sensitive accounting issues could be
cited; but none has received as much attention as accounting for
inflation, for none has such widespread potential repercussions
throughout the business world. Each method which has been pro¬
posed to replace or to modify our traditional methods would affect
different companies differently, making some look more prosperous
than they are under present methods, and others less prosperous.
For example, current purchasing power adjustments to historical
cost accounting (general price level accounting) tend to make utili¬
ties with heavy debt capital look better off; replacement cost ac¬
counting tends to make companies with a large investment in de¬
preciable assets—steel companies, for example—look relatively less
profitable. A system using exit values (e.g., continuously contem¬
porary accounting or COCOA) would make firms using assets
which are not readily salable look bad. Though the protracted argu¬
ments about the relative merits of these and other rival systems
have not generally overtly recognized the vested interests which
stand to gain or lose by the way the argument goes, the political
implications of inflation accounting have probably done as much to
hamper agreement on the direction of future moves as have the
technical problems involved.
In some of these instances, notably those concerning contin¬
gency reserves and foreign currency translations, critics of the
FASB are asserting that economic behavior, such as reinsurance or
hedging, which would not have been rational under the old ac¬
counting rules becomes rational under the new ones. Such an as¬
sertion is difficult to defend, because the new rules have not
changed the underlying cash flows or the risks attached to them.
Only if significance is attached exclusively to “the bottom line,"
rather than to the present value of the enterprise, can the change
in behavior be defended.
32 David Solomons

Measurement and Politics

The above examples will serve to illustrate some of the points of

contact between accounting and politics. Many more could be cited.
Indeed, because the need for standards arises mainly in areas where
there is controversy, it is highly probable that in every case some
will think the new treatment less favorable to them than the old,
and there is constantly a temptation for such people to rush off to
their legislative representatives to get the government to interfere.
That sort of initiative represents the gravest threat on the horizon to
the private control of standard setting.16
If we are looking for ways to achieve political ends by tinkering
with methods of measurement, plenty of scope can be found out¬
side the accounting field. Indeed, the danger has already been ob¬
served in other areas. For instance, the index of retail prices has a
powerful effect on wage settlements in many industries. There is
nothing absolute about a price index. The number obtained de¬
pends on the choice of base year, the items chosen for inclusion in
the market basket, and the weights attached to the items in con¬
structing the index. A statistician who agreed with Hawkins about
the responsibilities of those concerned with measurement could
easily construct an index which would damp down price changes—
then take credit for aiding in the fight against inflation.17
When is a worker “unemployed”? How a government defines
unemployment can have a powerful impact on monetary and fiscal
policy in a society which has definite views on what is and is not
an acceptable number of jobless citizens, for if 5 percent unemploy¬
ment is accepted as “full employment" for policy purposes, a re¬
corded level of 7 percent will be read as signaling the need for
monetary expansion or lower taxes. The way in which unemploy¬
ment is defined and measured therefore has a direct impact on
economic policy. Would anyone seriously argue that the statistical

16. The letter dated October 6, 1977, addressed to the Chairman-designate of the
FASB by Senator Proxmire and four Wisconsin congressmen, and reported in Status
Report No. 55, October 14, 1977, is a case in point.
17. There is nothing far-fetched about this. In The Final Days (Avon Books, 1977,
p. 177), Woodward and Bernstein state that "late in 1971, Nixon had summmoned
the White House personnel chief, Fred Malek, to his office to discuss a 'Jewish cabal'
in the Bureau of Labor Statistics. The 'cabal,' Nixon said, was tilting economic
figures to make his administration look bad." Another example came to my notice
when I was in Singapore in 1976. There the administration was accused of keeping
the price index down by changing the grade of rice included in the index.
 The Politicization of Accounting 33

measure of unemployment should be rigged (or changed from time

to time) to favor an expansionist policy or its opposite?
I have suggested elsewhere that the highway authorities could
lower the average speed at which vehicles are driven, and thus
reduce the accident rate, if they would have all speedometers con¬
sistently overstate speeds, so that drivers would think they were
driving faster than they actually were.18 Speedometers influence
behavior. Why not influence it in a beneficent direction?
This last example will serve to lay bare the profound threat to
accounting implicit in the propositions of Hawkins and others
which have been referred to above. If it ever became accepted that
accounting might be used to achieve ends other than purely mea¬
surement ends, then faith in it would be destroyed, just as faith in
speedometers would be destroyed once people realized that they
could be falsified to influence driving habits.19
The view that “because the Standards Board has the power to
influence economic behavior, it has an obligation to support the
government's economic plans" is, I believe, not only destructive of
accounting, it is also infeasible.20 Governments have a habit of
changing their plans from year to year, and even from month to
month. Are accounting standards to be changed with every change
in the political climate? One has only to recall President Nixon's
turnabout from a “no wage and price controls" stance to an espousal
of rigorous controls in 1971-72, or President Ford's switch from pro¬
posals for tax increases to “whip inflation now" in favor of tax cuts to
stimulate employment in 1974, to see how futile it is to talk about
supporting the government's economic plans, or how impossible it
would be for a Standards Board to keep up with the government.

The Importance of Neutrality

The mere fact that information has an effect on human behavior

does not mean that it should not seek to be neutral as between
different modes of behavior. Unless it is as neutral as the accoun¬
tant can make it, one can hardly see how it can be relied on to guide
behavior. As Chambers observes, “If the form of accounting is

18. David Solomons, Price Waterhouse lecture at Stanford University in 1972,

"Financial Accounting Standards: Regulation or Self-Regulation?"
19. Support for this view is to be found in Arthur R. Wyatt's article, "The Eco¬
nomic Impact of Financial Accounting Standards," Arthur Andersen Chronicle, Sep¬
tember 1977.
20. See note 13.
34 David Solomons

permitted to change with changes in policy, any attempt to scrutin¬

ize and to evaluate specific policies will be thwarted/'21
Neutrality in accounting implies representational accuracy. Curi¬
ously, this quality has been little discussed, though others related
to it have received more attention. The American Accounting Asso¬
ciation's 1977 Committee on Concepts and Standards for External
Financial Reports gets near the heart of the matter when they say:

Users of financial information prefer that it have a high degree of reli¬

ability. Reliability is that quality which permits users of data to depend
upon it with confidence as representative of what it purports to repre¬
sent. But reliable information is not necessarily useful. It could, for ex¬
ample, be reliable but unrelated to the use at hand. Several relatively
general terms are often used as synonyms for, or to cover parts of, the
concept of reliability. Thus, verifiability, objectivity, lack of bias, neu¬
trality, and accuracy all are related to reliability. Like relevance, reliabil¬
ity (above some minimal level) is a necessary but not a sufficient condi¬
tion for usefulness of data.22

If "like relevance, reliability (above some minimal level) is a neces¬

sary but not a sufficient condition for usefulness of data," the two
qualities together go far towards ensuring usefulness. Relevance
comprehends subsidiary characteristics of information that one
might list—such as timeliness. And the essential element in the
reliability of information (at least for our present purpose) is that it
shall as accurately as possible represent what it purports to repre¬
sent.23 This is what I mean by neutrality.
Neutrality, in the sense in which the term is used here, does not
imply that no one gets hurt. It is true, as the American Accounting
Association's 1977 Committee on the Social Consequences of Ac¬
counting Information says,

Every policy choice represents a trade-off among differing individual pref¬

erences, and possibly among alternative consequences, regardless of
whether the policy-makers see it that way or not. In this sense, accounting

21. Raymond J. Chambers, Accounting, Evaluation and Economic Behavior, (Engle¬

wood Cliffs, N.J.: Prentice-Hall, 1966), p. 326.
22. Statement on Accounting Theory and Theory Acceptance (Sarasota, Fla.: Commit¬
tee on Concepts and Standards for External Financial Reports, American Accounting
Association, 1977), p. 16.
23. This is close to Ijiri's statement that “in general, a system is said to be reliable
if it works the way it is supposed to." (The Foundations of Accounting Measurement,
Prentice-Hall, 1967, p. 137). But his more formal definition of reliability is couched
more in terms of the predictive value of information, an aspect of the matter with
which I am not here concerned.
 The Politicization of Accounting 35

policy choices can never be neutral. There is someone who is granted his
preference, and someone who is not.24

The same thing could be said of the draft, when draft numbers
were drawn by lot. Some people were chosen to serve while others
escaped. The system was still, by and large, neutral in the sense
that all males of draft age were equally likely to be selected.

Accounting as Financial Cartography

Information cannot be neutral—it cannot therefore be reliable—if

it is selected or presented for the purpose of producing some chosen
effect on human behavior. It is this quality of neutrality which
makes a map reliable; and the essential nature of accounting, I
believe, is cartographic. Accounting is financial map-making. The
better the map, the more completely it represents the complex
phenomena that are being mapped. We do not judge a map by the
behavioral effects it produces. The distribution of natural wealth or
rainfall shown on a map may lead to population shifts or changes in
industrial location which the government may like or dislike. That
should be no concern of the cartographer. We judge his map by
how well it represents the facts. People can then react to it as they
will.
Cartographers represent different facts in different ways and
match the scale of their maps to their purpose. Every map repre¬
sents a selection of a small portion of available data, for no map
could show physical, political, demographic, climatic, geological,
vegetational, and numerous other kinds of data and still be intelli¬
gible. The need to be selective in the data that one represents does
not normally (though it could) rob the map of its neutrality.
Like the geographic features that cartographers map, different
financial facts need to be represented in different ways, and differ¬
ent facts are needed for different purposes. It is perfectly proper for
measurements to be selected with particular political ends in mind
or to be adapted to a political end if users of the measurement know
what is being done. For example, the government is entitled, for
taxation purposes, to define taxable income in whatever way it
considers suitable. It would be quite another matter if the govern¬
ment required accountants to use this definition for all purposes to
which they might put an income number.

24. Committee Report, p. 24.

36 David Solomons

That maps, like financial statements, can mislead, even when

completely accurate, is undeniable. John K. Wright, writing on the
subjective element in maps, gave a good example:
Assume, for example, that the frontier province of Pomeria, which for¬
merly belonged to Sudia, was annexed by Nordia in the last war and that
a recent census has shown that half of the population of Pomeria are
Nordians and half Sudians. The Nordians are concentrated in the towns,
the Sudians form the bulk of the rural population. On a detailed map of
Pomeria in the Bulletin of the Sudian Geographical Society the areas
where less than 10 percent of the total population is Sudian are left white
and those where more than 90 percent is Sudian are shown in dark red,
with intermediate gradations of lighter red to bring out intermediate
percentages. Clearly most of the map would be red, giving an impression
of a preponderant Sudian population.25

Wright did not use the phrase “creative cartography," but he might
have applied it to many of the map features that he mentioned in
his paper.
Cartographers sometimes map the unmappable. They use a term
“isopleth" to mean “a line of assumed constant value." For ex¬
ample, a population isopleth is a line joining points having equal
population densities. Of course, a point cannot have a population
density. Yet we have no difficulty in accepting such a line as having
representational value.

Some Contrary Views

Recently a different view of accounting from mine has been ex¬
pressed, and it deserves comment here. Sometimes the difference in
the weight to be given to economic impact in standard setting is
merely a matter of emphasis. Sometimes it is more fundamental.
Sometimes neutrality is dismissed on other grounds.
Probably no one argues that those who formulate accounting
standards should do so with total unconcern for the economic con¬
sequences. Indeed, without some concern for such consequences,
the problem areas which call for standards could not be selected.
Presumably the economic consequences of not having a standard to
deal with some particular problem directed attention to that area in
the first place, just as early cartographers concerned themselves pri¬
marily with territory which had the greatest economic significance
to their contemporaries. To require the Standards Board to report
on the probable economic impact of a proposed standard when an

25. John K. Wright, "Map Makers are Human," Geographical Review, XXXII (Oc¬
tober 1942), 530.
 The Politicization of Accounting 37

exposure draft is issued26—if it can be done, for the impact will

often be unclear or ambiguous—is not at all objectionable, so long as
the standard is designed to bring about a better representation of the
facts of a situation, with whatever behavioral results flow from that, and
not to promote some preselected economic objective or mode of behavior.
Some of those who would play down the value of neutrality in
accounting standards do so because, they argue, the financial
phenomena which accountants must report are not independent of
the reporting methods which are selected. This view is expressed
by the American Accounting Association's (AAA) 1977 Committee
on the Social Consequences of Accounting Information, in the fol¬
lowing passage from its report:

The view that measurement merely involves representing or describing

stocks and flows is a static view. It assumes that the stocks and flows are
history, fixed forever, no matter how you measure them. But what about
tomorrow's stocks and flows? They are governed by the business deci¬
sions of enterprises—decisions which might change depending upon
how you choose to measure the stocks and flows. The traditional frame¬
work fails to take this interdependence of measurement and decisional
behavior into consideration.27

Where human beings are the subjects of measurement, it is true

that behavior and measurements are not independent of each other.
But this does not make neutrality a less desirable quality of mea¬
surement in such cases. If one substitutes speedometers for ac¬
counting and driving behavior for stocks and flows in the AAA
committee's statement above, one can see that as an argument
against neutrality it is quite unconvincing. There is nothing static
about the relation between the speed of a vehicle and the reading
on the speedometer, and there is unquestionably feedback. The
behavior of the driver is reflected on the dial, and what is on the
dial affects the behavior of the driver. Speedometers still should
register speed accurately and neutrally. The decision on how to
react to the reading must be left to the driver.
A different criterion for the selection of approved accounting
methods is put forward by Beaver and Dukes in a discussion of
interperiod tax allocation.

The method which produces earnings numbers having the highest asso¬
ciation with security prices is the most consistent with the information

26. As recommended by Prem Prakash and Alfred Rappaport in "The Feedback

Effects of Accounting," Business Week, January 12, 1976, p. 12.
27. Committee Report, p. 23.
38 David Solomons

that results in an efficient determination of security prices. Subject to

[certain] qualifications, it is the method which ought to be reported.-8

And having found that "deferral earnings are most consistent with
the information set used in setting security prices," they conclude:

If one accepts market efficiency, the results suggest that the Accounting]
P[olicy] B[oard] made the "correct" policy decision, in the sense that it
requires a method which is most consistent with the information im¬
pounded in an efficient determination of security prices.29

Beaver and Dukes themselves point out that any inferences to be

drawn from their evidence are "conditional upon the prediction
models used to test the accounting measures. . . . Any findings are
the joint result of prediction models and accounting methods, and
only appropriately specified joint statements are warranted."30 In
other words the identification of the accounting method which is
found to generate earnings numbers or cash flow numbers most
closely associated with security prices depends on the way that
"unexpected returns" are defined. The results of this analysis do
not point unambiguously, therefore, towards a particular account¬
ing method.
This could well explain why, left to themselves, companies do not
all choose the same accounting methods. They do not all use the
same prediction models, and therefore the accounting method that
contains the most information for one company is not the one that
contains the most information for another company. One moral
might be that we do not need accounting standards at all, but rather
that in an efficient market laissez-faire should prevail. A different
conclusion about the Beaver and Dukes study is reached by Go-
nedes and Dopuch when they say that "under the contemporary
institutional setting, capital market efficiency—taken by itself—
does not imply that the prices of firms' ownership shares can be
used in assessing the desirability of alternative information-
production decisions."31 In any case, whichever way the efficient
market points us, it does not point us towards politically motivated
accounting standards.

28. William H. Beaver and Roland E. Dukes, "Interperiod Tax Allocation, Earn¬
ings, Expectations and the Behavior of Security Prices," Accounting Review, April
1972, p. 321. They add, in a footnote, that "the criterion suggested above provides a
simplified method for preference ordering of alternative measurement methods."
29. Ibid., p. 331.
30. Ibid., p. 332.
31. "Capital Market Equilibrium p. 92.
 The Politicization of Accounting 39

Limitations of the Analogy with Cartography

There is a danger, with any analogy, of pushing it too far, and the
analogy between accounting and cartography is no exception. Most
maps represent external phenomena which have an independent
existence of their own. The accountant is on safe ground only when
he is doing the same thing as the cartographer—representing exter¬
nal phenomena, in his case cash flows, contractual rights, market
values, and the like. Of course, cartographers have sometimes
amused themselves by drawing maps of fictitious countries, like
Erewhon or Atlantis, an activity which, too, has had its counter¬
parts in accounting.
Whatever limitations representational accuracy may have in
pointing us towards right accounting answers, it will at least some¬
times enable us to detect a wrong answer. For instance, FASB State¬
ment No. 2, which requires all research and development expendi¬
tures to be expensed as incurred, is bad cartography, because to
represent the value of the continuing benefits of past research ex¬
penditures as zero will usually not accord with the facts of the
situation, however expedient the treatment may be. Off-balance-
sheet financing requires that certain unattractive features of the
landscape be left off the map, so again the map is defective. The
criterion for judging these rules is not the effect which they may or
may not have on business behavior. It is the accuracy with which
they reflect the facts of the situation.

Conclusion

It is not at all palatable for accountants to be forced to choose

between appearing indifferent to national objectives or endanger¬
ing the integrity of their measurement techniques. But if the future
well-being of our discipline is what matters, the right choice should
be easy to make. It is our job—as accountants—to make the best maps
zve can. Others, or accountants acting in some other capacity, have the
job of using those maps to steer the economy in the right direction. If the
distinction between these two tasks is lost sight of, we shall greatly
diminish our capacity to serve society, and in the long run every¬
body must lose.

October 1977
I
The Feasibility and Desirability
of Accounting Standards
Harold Bierman, Jr.

William A. Paton and A.C. Littleton start their classic monograph

on accounting standards with this statement:

The purpose of accounting is to furnish financial data concerning a busi¬

ness enterprise, compiled and presented to meet the needs of manage¬
ment, investors, and the public.

A consistent framework of standards is needed to serve as a basis for

judgment in constructing and interpreting financial statements.

Accounting standards should be systematic and coherent, impartial and

impersonal, and in harmony with observable, objective conditions.1

Later they explain their choice of terms.

The term "standards” is used advisedly. "Principles" would generally

suggest a universality and degree of permanence which cannot exist in a
human-service institution such as accounting.2

These introductory statements act as a counterweight to a series

of attacks that are currently being made on the suitability of estab¬
lishing standards. For example, Demski writes:

. . . generally speaking, we cannot rely on standards to provide a nor¬

mative theory of accounting. No set of standards exists that will always
rank alternatives in accordance with preferences and beliefs—no matter

Note: In calendar time it is over twenty years since I took my last course from
Professor Paton. In real time it is a great deal less.
1. W. A. Paton and A.C. Littleton, An Introduction to Corporate Accounting Standards
(Chicago: American Accounting Association, 1957), p. 1.
2. Ibid., p. 4.

41
42 Harold Bierman, Jr.

what these preferences and beliefs are, as long as they are consistent in
admitting to the expected utility characterization.3

The Demski position is not completely inconsistent with that of

Paton and Littleton if we recognize that Demski is attacking what
Paton and Littleton define to be principles. It is clear that Paton
and Littleton recognized the danger of indicating absolute prefer¬
ences that were all-encompassing. Certainly their explanations for
avoiding the word principles is not inconsistent with the spirit of
Demski.4
The primary object of this paper is not to define explicit detailed
standards for accounting, but rather to argue the feasibility and
desirability of setting standards. Let us consider two approaches to
setting standards.5
The most direct way to set standards is to take specific business
transactions requiring accounting entries and recommend the exact
manner in which these transactions will be recorded and presented.
The Accounting Research Bulletins and the Opinions of the Accounting
Principles Board of the American Institute of Certified Public Ac¬
countants—and now the "Financial Accounting Standards" of the
Financial Accounting Standards Board as well—have all been pre¬
pared and presented in this spirit. The overall structure of account¬
ing is not defined by these standards but rather is implied by the
specific recommendations.
The difficulty with such an approach is that frequently the stan¬
dards lack theoretical foundations and are aimed at being short-run
solutions to long-run problems, offered for political reasons or as a
compromise. One result is that many standards adopted one year
are revised later and then revised again. This is the situation with
the accounting standards that have been established for accounting-
tax differences, the investment tax credit, and the accounting for
leases. A second result is that solutions relating to different types of
transactions may be contradictory despite basic similarities in the

3. Joel Demski, "General Impossibility of Normative Standards," Accounting Re¬

view (Oct. 1973), p. 721. Elsewhere (p. 720) he states: "... no set of standards exist
that will single out the most preferred accounting alternative without specifically
incorporating the individual's beliefs and preferences."
4. See also R.J. Chambers, "The Possibility of a Normative Accounting Standard,"
Accounting Review (July 1976), pp. 646-52, and Joel Demski, "An Economic Analysis
of the Chambers' Normative Standard," ibid., pp. 653-56.
5. In place of standard, words like framework, foundation, guidelines, rules, or
conventions could be used. The use of the term standard is meant to imply some¬
what more authority and no flexibility in choosing the method of accounting.
 Accounting Standards 43

transactions. Finally, the solutions may well violate certain logical

foundations of accounting which have been generally accepted by
accounting theoreticians.
The last sentence above suggests a second approach to setting
standards: defining a general set of operating guidelines that in
turn can be used to evaluate the accounting for specific transac¬
tions. Paton and Littleton's monograph is consistent with this ap¬
proach. Consider two possible standards: First, Match revenues of a
period and the expenses of earning these revenues to compute the income
of a period. Second, Record transactions as if the firm is to continue
operations unless there are specific reasons for assuming termination of
operations. These two standards are well known to all accountants.
As an a priori judgment they seem to be a reasonable start toward
constructing a basis for recording accounting transactions. Now
consider Financial Accounting Standard (2) of the Financial Ac¬
counting Standards Board, which requires the expensing of all re¬
search and development expenditures. This is a conservative proce¬
dure in the sense that there is zero probability that research and
development assets will be overstated: there will be no assets.6
However, the choice of conservatism (not recording a risky asset as
an asset) in preference to matching of revenues and expenses (the
research and development costs give rise to a positive expected
value) should be noted. The violence done to matching by Financial
Accounting Standard (2) is a major departure from the logically
consistent financial accounting that some of us would like to see.
The primary justification offered for expensing research and de¬
velopment (and thus abandoning matching) is that it ensures com¬
parability among different firms. Unfortunately the comparability is
to a great extent an illusion. There is little comparability between
otherwise similar firms undertaking different amounts of research
and development. On an ex-ante basis there is no reason for con¬
cluding that all research and development is worthless and should
be expensed.
The Financial Accounting Standards Board started recommending
accounting practices for such things as research and development
without having established basic concepts. If matching is not a stan¬
dard but conservatism is, then one can recommend the expensing of

6. In a previous paper Professor Roland E. Dukes and the present author com¬
mented in detail on this statement. The arguments of that paper are not reported
here. H. Bierman, Jr. and Roland E. Dukes, "Accounting for Research and Develop¬
ment Costs/'Journal of Accountancy (April, 1975), pp. 48-55.
44 Harold Bierman, Jr.

research and development in the interest of advancing conserva¬

tism. The accounting for research and development is a good ex¬
ample of the dilemma facing accounting. If research and develop¬
ment is capitalized as an asset and systematically depreciated, we
can conclude in some situations, after the fact, that the asset was
overstated. If research and development is expensed, research as¬
sets will never be overstated. On the average, we can be certain that
assets of firms engaging in research and development will have
their assets understated. This certainty is based on the reasonable
hypothesis, backed by empirical evidence, that the expected value
of research and development is equal to or greater than its cost.
It has been argued that since some people can be harmed and
others helped by the same type of error, there is no one correct
method of accounting, and that the accountant should merely sup¬
ply the basic information, such as the amount expended for re¬
search and development. In that case the calculation of financial
position and income would be shifted to the user of the data. It is
reasonable to recognize and to make clear the difficulty of account¬
ing for certain types of transactions and the possibility that the
values presented are estimates of random variables. It is not reason¬
able to argue that the accounting profession should not define stan¬
dards that help the practicing accountant record these transactions.

Simple Counting Problems

If someone bets that one plus one equals three and you testify as
an expert witness that one plus one equals two, at least one person
will not be happy with your judgment. The question then becomes
whether one and one can be said to equal two without regard to the
uses of the information and the extent to which the parties using
the information will be benefited or harmed.
In general we would say that one plus one equals two and would
be willing to see our conclusion in print. In the above example,
however, we are aware that one person benefits if a different
answer is given. When we know the consequences of our response,
therefore, we may find reasons for believing a different answer to
be appropriate.
Most of us think, however, that an accountant should report the
sum to be two, not three, since it is useful for society to know that
one plus one always equals two. The standard here is simply that one
should follow a specific counting rule regardless of what the person
receiving the information might prefer for a specific purpose. We
 Accounting Standards 45

assume it helps society to know that one plus one equals two no
matter what accountant does the measuring, and no matter how the
result is going to be used.
An arbitrary conclusion (or hypothesis) exists that society is bet¬
ter off with an exact counting scheme than with situations where an
unconventional result is presented because it might be more useful
for one or more parties. The use of the counting procedure is exact
and well defined; and although it may ignore the preferences of
some individuals, it has long-run advantages that are considered
more important than any short-run gains from a more flexible
procedure.
According to the above example, we want a number system
where the rules of addition are not affected by the preferences and
utility functions of individuals, but one that still accomplishes cer¬
tain objectives. As we move from a number system to a measure¬
ment system should there not be some rules (or standards) that we
can agree are desirable and better than alternative rules, even
though some individuals may be harmed by the application of such
rules? Management might be happier if the accounting reports fol¬
lowing a given set of rules did not reveal a downturn in income,
but this is not a sufficient justification, in itself, for departing from
accepted standards for an alternative set of rules.
In setting economic policy, such choices are made all the time.
The same person can concede that interpersonal utility functions are
not comparable, but at the same time he can strongly support or
oppose a taxing law which implicitly assumes that the utility of one
method of taxation is higher than the utility of another method. We
can likewise conclude that one method of measurement, account¬
ing, is superior to other methods.
Let us proceed from abstract examples and consider the account¬
ing for readily marketable securities. Is it possible to agree that one
measurement scheme is more desirable than another, leaving out
tax considerations, regulatory policy, and any other legal peculiari¬
ties? Assume that the accountant is measuring financial position at
the end of an accounting period. The choice is between cost, mar¬
ket, some mixture of the two (such as the lower of cost or market),
or a direct estimate of the cash flows of the asset.
Having studied economic decision-making, we know that a sunk
cost is not relevant for any decision. The cost of marketable securi¬
ties is a sunk cost and thus not relevant. While we might use cost as
an estimator of value in some situations, where value is not readily
available, this action is not appropriate here since the market value
46 Harold Bierman, Jr.

is easily ascertained. One could argue in favor of using a risk-

adjusted present value measure; but it is not necessary to estimate
value by this technique, since the market value offers an objective
measure of the market's evaluation of the future benefits adjusted
both for time and risk. There is uncertainty, but the market price
reflects the relevant risk considerations.
A priori there is a logical reason for using market value as the
basis of measurement, assuming that these market values can be
readily and cheaply obtained. The user of information based on
market values of marketable securities will have better information
for making decisions than one using cost or some mixture of cost
and market. If a rational decision-maker is going to use the infor¬
mation to make a normal investment decision, and if presenting
and obtaining additional information has a cost—if he has to make
an outlay to obtain it—the decision-maker will want a market value
measure if he behaves in a rational manner. (Remember that there
are no taxes.) We do not have to know the decision-maker's utility
function to reach this conclusion; the only requirements are ration¬
ality and a conventional decision. This does not mean that all users
of the financial information will be happy with the use of market
values.
A third party may benefit if the decision-maker has faulty infor¬
mation, preferring cost as the basis of reporting: for example, a
current stockholder might be able to unload shares of stock to un¬
wary purchasers if cost is used as the basis of reporting. In other
instances, when the regulation of a firm might be based on the use
of cost, the use of values which neglected regulatory considerations
might cause confusion for investors.
For such reasons one or more users of the information might prefer
the presentation that is more faulty from the point of view of conven¬
tional decision-making. Should these considerations affect the over¬
all reporting procedure? A conclusion on how to present readily mar¬
ketable securities, without taking a survey of preferences among the
individuals using the reports, is possible only if one is willing to
make assumptions about the objectives of accounting.
We can assume that the normal investor wants information about
the market value of the securities held by an organization and the
tax basis of the securities. Someone might also want to know the
cost basis of the securities and the date of acquisition in order to
evaluate the quality of managerial decisions. But we should be able
to agree that cost information alone is not sufficient. We should be
 Accounting Standards 47

able to establish a reasonable standard for presenting information

about marketable securities that includes market value measures.
One can assume that the objective is to supply information for
conventional decision-making. If the use of values results in unde¬
sirable consequences for a regulated firm such as a bank, then per¬
haps the regulations should be changed. Furthermore accounting
cannot be use to safeguard the interests of all investors, since the
interests of some groups of investors will be directly opposed. For
example, if the use of cost were to enable an investor to sell at a
higher price, then an unsuspecting buyer is paying more than he
would otherwise pay.
If the value information is readily available, even though the cost
information is presented, it can be argued that the stock price of the
firm will be set by an efficient market and the form of the account¬
ing information is not important. This argument assumes that the
market value information is readily available to enough investors to
cause the market price to adjust in response to the value informa¬
tion. But in some situations the harm is done if only one buyer is
taken in by the misleading accounting. For example, when one firm
is acquiring a second firm which is privately owned, there is no
market price. Rather the price is negotiated. If one bidder is faked
out, this may be sufficient to set a price that is out of line with the
facts of the situation.
An efficient market, by definition, readily digests and reflects
publicly available information; but if the information is not readily
available, the market is not likely to reflect it accurately. Suppose
that a company has information about the value of land it possesses
but that this information is not published. The market can only
guess the value of the land. The company might indicate that it
owns two million acres of timberland, but such a description is not
a sufficient basis for evaluating the resources of the firm.

Measurement of Value

We have used the term value but have not defined it exactly. The
word is somewhat ambiguous since it can have several meanings.
First, we should recognize that the value of an asset is not necessar¬
ily measured by its cost, though the cost and value may happen to
be equal. Three different measures of value are worth considering:
the value in exchange, that is, either buying or selling the asset; the
value in liquidation; and the value in use.
48 Harold Bierman, Jr.

These three measures do not exhaust all the possibilities. For

example, here we will consider value only from the point of view of
the individual firm, but one could also consider it from the point of
view of society or the economy.
The value in exchange, at the time of the acquisition, is equal to
the cost of the asset. With the passage of time the value in exchange
will not necessarily equal the replacement cost. If the asset is cur¬
rently held, the owner might be more interested in what he could
realize from its sale than what he would have to pay to replace it.
The exchange value, if one is acquiring an asset, may not be identi¬
cal to the exchange value if one is selling it, assuming that we are
dealing with imperfect markets and there are implicit selling costs
and discrimination by sellers and buyers.
Now let us consider the difference between the exchange value
and the liquidation value. Liquidation value implies that the owner
is forced to dispose of the asset within a given time. This pressure
may require him to accept less than the exchange value in order to
complete the sale within the time limit.
Finally, we must consider the value in use. Once a firm acquires an
asset, the asset will have a value specifically associated with the firm
that acquired it. For example, when a firm builds a plant, the plant
may have one value to that firm, because of the firm's special skills,
but another value if it were owned by a different firm lacking these
skills. The amount that the asset could be sold for is less important
than the value resulting from its prospective use by the firm.
Unfortunately there remain many unresolved problems in mea¬
suring values in use for assets and liabilities. Recognizing that un¬
certainty is present, rather than one value in use, we may have a
probability distribution of outcomes. The actual value in use will
depend on which state of nature takes place in the future. The
present equivalent value, or risk-adjusted present value, depends
on how the market measures and weights the different possible
outcomes. This implies adjustments for time value, probabilities,
and risk preferences. Also, value in use is not independent of the
cost of replacing the asset since replacement cost sets a maximum
limit on the asset's value.

Probability Distributions

Most accounting measures deal with economic resources to which

some degree of utility attaches. In accounting, as in the physical or
social sciences, the true or exactly correct measurement is seldom at¬
tainable. It may in fact be impossible to define. There is a likelihood
 Accounting Standards 49

that a measurement will coincide with the true state of nature, but it
is more correct to speak of a probability distribution about the mea¬
sure. A person can measure a table and say it is 6 feet long, but a
more exact measure, perhaps with a different tool, may indicate 6.1
feet, or 6.15, or 6.158, and so on. We might find that the measure
depends on temperature. Any of these measures, or an average of
several measures, could be interpreted as a mean of a distribution,
with the unknown true measure falling within a given interval. The
"true” measure exists, but we do not know its value.
The word true is being used here in the same sense that Raiffa
and Schlaiffer use it in the following passage.7
. . . we shall be concerned with the logical analysis of choice among
courses of action when (a) the consequences of any course of action will
depend upon the "state of the world," (b) the true state is as yet un¬
known, but (c) it is possible at a cost to obtain additional information
about the state.

We can rarely observe the true value; more generally we will have
to estimate it. For example, J. Johnston makes this point in estab¬
lishing the foundation of econometric solution to the problem of
estimation:8 "It may be that a variable Z is exactly related in a linear
fashion to X by the relation Z = a + (3 X, but errors of measurement
obscure the true value Z, and instead of Z we observe Y — Z + U
where U denotes the measurement term."
The true measure in accounting may turn out to be a long de¬
scription. For example, a firm might face a gamble when two possi¬
ble outcomes occur at one period from now, as in the figure:

Time
0 1

‘t'TOOO

The figure gives us all the necessary basic information, but anyone
faced with a large number of such choices might find some summary
measures useful.
7. Howard Raiffa and Robert Schlaiffer, Applied Statistical Decision Theory (Boston:
Division of Research, Graduate School of Business Administration, Harvard Univer¬
sity, 1961), p. 3.
8. J. Johnston, Econometric Methods (New York: McGraw-Hill, 1960), pp. 6-7.
50 Harold Bierman, Jr.

One summary measure is to compute the expected monetary

value. This is $600. But the cash flows occur at time V and must be
adjusted for timing. If the $1,000 cash flow is adjusted for time
value, assuming that the discount rate is 10 percent, we obtain $909
for a present value and $545 for expected present value. It is still
necessary to subtract an amount for the market's risk aversion. Rea¬
sonable persons might disagree on the amount of cash flows and
their probabilities, the time value factor that is used, the deduction
for risk aversion, and the sequence and method of calculation. But
there should not be any disagreement on the desirability of devel¬
oping the basic elements of the calculation, a summary measure of
the risk-adjusted present value, as well as a full description of the
nature of the asset.
It was suggested above that one could move the estimate of value
away, on the average, from the unknown true value. This can be
illustrated by considering the following biased measures for the
example:
1. The use of $1,000, the maximum amount
2. The use of $0, the minimum amount
3. The use of $600, the expected value at time 1, without any time value
adjustment
4. The use of $545, the expected present value, without any adjustment
for risk preferences

In any particular situation one of the above may be better than a

more sophisticated calculation. But in general all of the above mea¬
sures have severe faults that introduce a bias away from a likely
market value measure.
The accountant uses the unit of dollars to measure financial posi¬
tion at a given moment and income for a given period. Except in an
artificially simple situation no one can measure either of these items
exactly, but we can ordinarily tell when we are moving closer to the
true measure or away from it. Assume that the true measure of total
assets is A and that we are attempting to estimate A. We can mea¬
sure a component of A in several ways and be reasonably certain
that one of these ways will lead to a less biased measure of A than
the others would yield. For example, we can count cash and con¬
sider that sum to be a component of A, or we can take the sum of
cash and subtract $1 million in the interest of conservatism. The
second procedure introduces a known bias and as a rule moves the
estimate of total assets away from a true estimate of A. Thus, al¬
though one cannot expect to find the true measures of income and
 Accounting Standards 51

financial position—these terms are even difficult to define—we can

move closer to the true measures instead of allowing known biases
to exist.
The misconception held by some, that accountants should be able
to present the one true measure, has hindered progress in the re¬
porting of financial information. If accounting were limited to pre¬
senting the amount of cash in the bank, a true measurement might
be attainable. But when the scope of accounting is broadened from
the limited objective of measuring cash, the possibility of finding
true measures or measures that are always correct disappears be¬
yond reach. The goal should be to present useful financial informa¬
tion arrived at in a fair or reasonable manner.
Instead of fair, we could say objective—not in the conventional
sense of the practicing accountant, but meaning relative freedom
from the biases of specific individuals. It is reasonable to attempt to
remove excessive biases of individuals and to reach common agree¬
ment, but such agreement should be what a reasonable person
would conclude from the evidence. One should not expect that all
reasonable persons would arrive at the same estimate, but rather
that the estimates of a group of qualified, unbiased experts would
not be too widely dispersed. Practicing accountants have pre¬
empted the term objective for a more narrow meaning than that
suggested here.
Today the accountant presents information, and the knowledge¬
able reader knows that the reports are prepared in accordance with
given rules. These rules restrict the differences between one ac¬
countant's results and those that another accountant might obtain;
the dispersion of possible outcomes arising from hiring different
accountants is thus relatively small, although by no means close to
zero.
One must realize that measures of income or financial position are
actually probability distributions. At best the accountant is present¬
ing estimates and the true values (forever unknown) may be differ¬
ent. The physicist is willing to estimate the distances to stars in
thousands of light years. There is no reason why accountants cannot
apply the same sort of reasoned judgment to financial affairs.

More than One Method

The accounting profession has tended to search for one method of

accounting that will fill all needs. Historically we have had the
choices between cost-based accounting, value accounting, and price
52 Harold Bierman, Jr.

level accounting (price index applied to some items, common dol¬

lar, or some variation of replacement cost).
We should note that there are objective value measures for some
assets, such as marketable securities, and where they are available
and reliable these value measures should be used. For other assets,
reliable value measures may be much more difficult, perhaps even
impossible, to obtain. The value of specific assets may be so woven
into the entire economic fabric of the corporation that if value were
used to record the assets the accountant would have to value the
firm. (He would then need to justify the value measure obtained,
compared to the market value in total of the company's securities
outstanding.)
In the presence of imperfect markets, uncertainty and long time
horizons, and interrelationships of assets, the accountant may have
to use original cost adjusted in some systematic fashion for wear
and tear, that is, some depreciation procedure. But while we may
be able to define reasonably the type of information which would
be best for a rational decision-maker, it may not be feasible to
supply that information.
The recommendation that cost may be a safe haven from a storm
of uncertainty and complex relationships does not preclude a system¬
atic adjustment for price level changes if the circumstances sur¬
rounding the decision require that type of adjustment. The accoun¬
tant might even supply value estimators if reasonable bases for
making the estimates are available.
For example, consider a stand of timber owned by a corporation.
The cost of the seedlings thirty years earlier is obviously one bit of
available information. But is that cost information more useful than
a professional estimate of the current value, after the trees have
grown for thirty years? A prospective purchaser of the firm would
want an estimate of today's value of the timber and the land. A
buyer of common stock deserves the same information.
Instead of timberland we could shift to a coal mine or an oil well.
With such properties the costs of discovery or development are not
likely to indicate the value of the resource. An investor in the secu¬
rities of a corporation possessing such assets recorded at cost is
handicapped in appraising the financial health and prospects of the
corporation.
With respect to an oil company, we might not all agree on the
best set of information, but could we agree that as a prospective
investor we would be interested in these items?
 Accounting Standards 53

1. The proven recoverable oil reserves and the types of oil

2. The location of the oil reserves

3. The costs of recovery and of getting the oil to market

4. The current price of the oil and the expected future prices
5. The rate at which the oil can be pumped

6. Any political or institutional factors, such as tax status, that might

affect value

If this information were combined into one value measure we

would no doubt differ on such estimates as future prices, time
discount factors, risk adjustments, and the like. But we would
easily agree that some measures for the listed items are needed by a
rational investor. It will be more difficult to agree on the exact value
of a summary measure of risk-adjusted present value.
Thus we should try to concur on the basic elements of informa¬
tion which a reader of a financial statement certainly needs, and
then recognize that there is room for disagreement about the refine¬
ments of presentation.
We may still be willing to accept some generalizations as stan¬
dards. The following statements might help to establish such stan¬
dards:

1. Revenues should be recognized when earned.

2. When reliable value measures are available, they should be used.
3. When cash flows are to be received or disbursed in the future, the
time value of money should be taken into consideration using the
best estimators of interest rates.
4. Depreciation expense accounting should be a systematic method of
allocating the cost of an asset to expense over the life of the asset,
taking into consideration value decreases when justification exists for
estimating such decreases.

The above statements are not presented as absolutely correct or

all-inclusive standards, but rather as candidates for consideration.
Others could be suggested. The establishment of a set of standard
guidelines is useful, however, as a basis for accounting procedures
for specific transactions.
Are there any standards affecting accounting that we can accept
universally? Let us consider the time value of money. Most accoun¬
tants would agree, one hopes, that in the normal commercial world,
with its opportunities to lend and borrow funds, a dollar received
today is worth more than the dollar received one year from today.
If the basic concept of time value of money is accepted, then
54 Harold Bierman, Jr.

accounting transactions should take note of interest costs. For ex¬

ample, a building that will cost $10 million if payment is made at
time 0 may cost $12 million if payment is made two years later,
when the building is ready for occupancy. What is the cost of the
building? Does the cost depend on whether the payment is $10
million at time 0 or $12 million at time 2? Would it help to know
that the firm can borrow or lend at an interest rate of 10 percent?
Would you want to know whether the building was financed with
debt or common stock?
However one answers the above questions, it is clearly incorrect
to record the building at a cost of $10 million if the payment hap¬
pens to be made at time 0 and $12 million if the payment happens
to be made at time 2. This situation illustrates the ever-present and
unresolved problem of interest during construction. Some firms rec¬
ognize interest and some do not, a practice suggesting that the
presence or absence of interest is a function of the specific corpora¬
tion for whom the accounting is being done. Accounting should
acknowledge the universality of interest (time value of money) and
then decide how it should be treated in situations where there are
interest costs.
General standards can never be expected to solve all accounting
problems. There will always be specific situations that require
analysis peculiar to the circumstances. For example, one of the ma¬
jor deficiencies of current reporting practices is the failure to con¬
solidate the financial affairs of a parent and its captive finance com¬
pany. The captive finance company carries the receivables of the
parent (financed with a large percentage of debt), but this separate
entity is not consolidated since it is a "different type of business."
A standard is needed to define more exactly those corporations
which are not to be consolidated.

Conclusions

We sometimes tend to think of accounting as entailing unique

difficulties of measurement. Its problems are common to other dis¬
ciplines, an example being the relatively easy problem of measuring
the maximum life span of a person. One can give a facile answer
based on recorded history, but consider the following statement by
William Feller:9

9. William Feller, An Introduction to Probability Theory and Its Applications (Lon¬

don: John Wiley & Sons, Inc., 1950), pp. 7-8.
 Accounting Standards 55

It is impossible to measure the life span of an atom or a person with¬

out some error, but for theoretical purposes it is expedient to imagine
that these quantities are exact numbers. ... Is there a maximal age be¬
yond which life is impossible, or is any age conceivable?. . .Moreover, if
we were seriously to discard the possibility of living 1000 years, we
should have to accept the existence of a maximum age, and the assump¬
tion that it should be possible to live x years and impossible to live x
years and two seconds is as unappealing as the idea of unlimited life.

In the same spirit that Feller is willing to accept the probability

that a person might live to be a thousand years old, we should be
willing to estimate the financial position and income of a business
firm.
Henry Morgeneau has written that measurement cannot be “ex¬
empt from analysis; for if it were, every measurement would re¬
quire to be simply accepted as a protocol of truth, and one should
never ask which of two conflicting measurements is correct, or
preferable."10
The accounting profession cannot consider the basic questions of
measurement to be exempt from analysis. Present accounting prac¬
tices of measurement are not “protocols of truth." We must under¬
stand the nature and problems of measurement if the practice of
accounting is to realize its full potential in reporting financial infor¬
mation. Not all accountants are entirely in agreement, and account¬
ing conventions are constantly changing. The accountant is charged
with reporting financial affairs in an uncertain world. Inevitably
there is controversy about the methods of accounting and about the
actual measures presented. This controversy is welcome. One's pro¬
fession is much more interesting when it is dynamic, ever chang¬
ing, and challenging than when it is stagnant.
The relevant question facing accountants is whether a set of intu¬
itively appealing, logically consistent standards can be established
on an a priori basis. Or is it necessary to determine the utility of the
alternatives and choose the alternative with the highest measurable
utility? If we insist on the latter course we must conclude that the
setting of standards is not feasible. If on the other hand we recog¬
nize that we must choose between different measurements despite
the incompleteness of our information, and that the solutions we
reach will not necessarily be the best in all respects, we can then
make these choices with good conscience.

10. Henry Morgeneau, “Philosophical Problems Concerning the Meaning of Mea¬

surement on Physics," in C.W. Churchman and P. Ratoosh (eds.). Measurement,
Definitions, and Theories (New York: John Wiley & Sons, Inc., 1959).
56 Harold Bierman, Jr.

Accounting is not a science, subject to the same exact rules as

counting. On the other hand, accounting should riot be without a
well-defined structure.
Accounting is the art of measuring and communicating financial
information about economic events in accordance with reasonably
well-defined conventions that are intuitively appealing. Once we
agree that the tasks of measuring the financial position and income
of a period with perfect accuracy are impossible, we can proceed
with improving the state of the art.

June 1976
Accounting for Investments in Debt
Securities
Maurice Moonitz

This study of accounting for investments in debt securities treats

a part of the broader topic of accounting for intercorporate invest¬
ments and is a companion to a recent study of intercorporate invest¬
ment in shares of capital stock.1 The discussion here is conceived as
an aid to decision-makers in the field of accounting. Accordingly it
starts from the conventional practices of the day, subjects them to
close scrutiny, considers the alternative use of the consistent use of
market values, and makes a few recommendations.

Nature of an Investment

Investments in equities (capital stock and rights to acquire capital

stock) and in debt instruments clearly qualify as assets of the inves¬
tor. They are economic resources that are sought after, or valued,
because they embody the right to or expectation of future cash
receipts; those rights and expectations are protected in varying de¬
grees by contractual and statutory provisions, enforceable if neces¬
sary in a court of law. In brief, investments possess the usual de¬
sired attributes of assets.
Investments differ, however, in one important respect from other
items classed as assets. Always in form and frequently in substance
they represent an interest in someone else's operations. Their value

1. Reed K. Storey and Maurice Moonitz, Market Value Methods for Intercorporate
Investments in Stock. Accounting Research Monograph No. 2 (New York: American
Institute of CPAs, 1976).

57
58 Maurice Moonitz

depends in large measure on the extent to which the issuers of the

securities manage their operations. The owner-investor can do little
except, given a well-organized securities market, decide to hold, to
buy (invest), or to sell (disinvest). An obvious exception to this
statement is the parent-subsidiary relationship: in such cases the
securities that represent the legal tie between parent and subsidiary
do not appear in the consolidated financial statements that are pre¬
pared to reflect the economic substance of the activities conducted
within the controlled group.
The management of an economic entity has direct control over its
assets other than investments. The costs and benefits of managing
those other assets flow directly to the entity, not indirectly as is the
case with investments generally. The skills needed to manage the
two groups of assets are somewhat different. To manage a factory
requires skills in designing and operating factory facilities, for ex¬
ample, or procuring raw materials, recruiting and training a labor
force, and marketing the product. To manage a portfolio of securi¬
ties requires some skill in assessing the quality of business opera¬
tions and estimating the effect of securities-market forces on par¬
ticular investments. In large measure a skilled factory manager
looks inward, a skilled portfolio manager outward.
These considerations suggest that financial reporting could reflect
this basic difference among the assets. For example, a statement of
financial position could group the investments together under a
single caption separate and apart from the assets devoted to the
other activities of the entity. An income statement could report the
results of investment activities (for example, dividends, interest,
capital gains and losses) in one place or section, distinct and apart
from the results of other activities; and a statement of the flow of
funds could be handled similarly.

Difference between Equity and Debt Securities

Equity securities are open ended. Ownership of these entitles
investors to participate in the results accruing to the class of secu¬
rity held, according to the priorities and formulas contained in the
contracts with the issuers. If results are poor, investors in equity
securities stand to gain little, if anything, and may lose part or even
all of their investment. If results are good, they stand to gain much.
This characteristic is evident, for example, in the prevalent rule that
dividends on such securities do not inure to the benefit of investors
until they are earned and declared by the directors of the issuer. No
financial return is guaranteed to investors until earnings emerge
 Investments in Debt Securities 59

and dividends are declared. Other evidence of open-endedness is

the absence, in most cases, of a maturity date. There is no guaran¬
tee that investors will ever get their money back. In such a situation
the equity and the market price of an investment are more impor¬
tant than its historical cost in assessing its significance.
Debt securities are typically closed, rather than open ended. By
contract, ownership entitles an investor in debt securities to a
stipulated return, regardless of the results accruing to the issuer
(debtor) as an entity. As long as the issuers have the wherewithal to
pay, investors in debt securities are entitled to their interest and
principal. But they have no claim to anything more. Debt securities
have an absolute priority over equity securities. Interest accrues on
the debt as a function of time, whether earned by the issuer or not.
Debt securities also ordinarily have a fixed maturity date, at which
time the remaining principal (or other amount unpaid under the
debt contract) must be paid by the issuer to the investors. As a
result of these attributes, the equity of debt securities in the opera¬
tions of the debtor has no special or distinctive meaning. Cost and
market are the figures of importance in assessing the significance of
this type of investment.
The distinction drawn above presumes the existence of two sepa¬
rate classes of securities: equity and debt. In the world of finance
the two classes do exist and are of considerable importance. But in
addition some securities are designed to combine features of both:
for example, convertible debentures and income bonds. There are
also some debt instruments, such as low-grade bonds, issued by
companies that no longer have the unchallenged capacity to pay
according to contract terms. In that case the low-grade bond be¬
haves in many respects like an equity investment.
The present study does not consider securities that combine debt
and equity characteristics, as those are described above, the motive
being to focus attention on the important cases in which issuers of
debt securities do in fact live up to their commitments. By this
device we can isolate the effects of market forces on the value of the
debt instrument and analyze the consequences for accounting.
In the case of a debt security the terms of the contract assume
great importance. The contract spells out the amounts of money to
be paid by the issuer to the investors and the dates on which those
amounts will move. At the date the investment is made the market
rate of interest serves as a discount factor to equate the promises
just described with the market price. After that date the market
price fluctuates in response to two factors.
60 Maurice Moonitz

The first is the degree of doubt that emerges regarding the ability
of issuers to keep their promises. As that doubt increases, the mar¬
ket value of the investment will ordinarily fall. Since this factor
derives from forces confined to the issuer, and not to the market
generally, we will not analyze it here but merely note its existence.
Our attention will instead be devoted to the case in which a change
in market price, if any, is attributable to the second factor, the
market rate of interest.
When the market rate of interest drops, the discounted (present)
value of the fixed set of promises contained in a debt instrument
increases. It takes more money today to bring forth the fixed prom¬
ises of the debt instrument. When the market rate increases, the
discounted (present) value declines. It takes less money today to
bring forth the same stream of cash receipts.
If investors hold debt securities to maturity, they will recover
their initial dollar investment (principal), regardless of changes in
the market rate of interest, because at maturity any debt instru¬
ment is worth its face value. If, for example, the market rate of
interest has been increasing, the market value of the debt security
will decline at first but then rise again to face value as the matu¬
rity date approaches.
Many investors do not, of course, hold debt securities until matu¬
rity but sell them instead to other investors. If so, the first investors
stand to gain or lose from changes in the market rate of interest,
since that rate in turn helps determine the market price of the secu¬
rity. Their proceeds from the sale will be more, or less, than the cost
of their investments.
Here is another way of looking at the difference between holding
a debt security until maturity and selling it prior to that time. As
long as investors hold debt securities, they will receive the amounts
promised in the contract and will therefore earn the market rate
expressed or implied (the effective rate) in the price paid at date of
investment, regardless of fluctuations in market price. This may be
termed the contractual gain. The contractual gain may always be
realized (unless insolvency or bankruptcy of the issuer occurs)
merely by holding the investment to maturity.
At the same time, if the market price does in fact rise or fall,
investors can realize an additional gain or loss by selling the invest¬
ments. This we may term a holding gain (loss), also frequently re¬
ferred to as a capital gain (loss). The market price of a debt security
may then be described as the algebraic sum of its cost (amortized cost
 Investments in Debt Securities 61

if premium or discount is present) and the holding gain (loss). The

holding gain (loss) becomes explicit (realized) to the seller if inves¬
tors do in fact sell. It is implicit (unrealized) if they do in fact hold.
But the opportunity to realize the holding gain is lost if the invest¬
ment is held to maturity.
Two related issues then emerge in accounting for these invest¬
ments: when to recognize and report the realized (explicit) holding
gains (losses); and when to recognize and report the unrealized
(implicit) holding gains (losses).

Analysis of Change in Market Rate of Interest

In order to pinpoint the assertions made in the preceding section

and to provide a vehicle for further discussion, we present a de¬
tailed analysis of the behavior of an investment in a debt instru¬
ment during a rise in interest rates. Specifically we examine the
behavior of a $1,000 bond with a ten-year maturity, issued with a 4
percent coupon in a 4 percent market, the bond being therefore
initially issued at par or face amount. In this way we avoid the
complications of the amortization of premium or discount and in¬
stead assume effectively that we are dealing with direct placement.
We chose to avoid the presence initially of a premium or a discount
because they are not essential elements of any debt contract. Pre¬
mium and discount emerge as technical market devices to equate a
coupon (contractual) rate with a market or effective rate that is lower
(premium) or higher (discount) than the coupon rate. Premium or
discount can always be avoided by stating the contractual rate at the
market rate, as is done in the case of direct placements.
During the ten years of the bond's life the market rate of interest
is at 4 percent for two periods; then it rises to 5 percent for two
more periods and finally levels off at 6 percent. As a result of the
behavior of the market rate of interest, the market value of the 4
percent bond drops to a low of $916 at the end of the fifth period
then rises slowly to $1,000 at maturity, the end of the tenth period.
All of this is summarized in the first three columns of Table 1.
The money flows are there displayed for two cases. In case A, the
investor holds the 4 percent bond to maturity. In case B, the inves¬
tor switches from the 4 percent bond to a 6 percent issue at the end
of the fifth period by selling the 4 percent bond and investing the
proceeds in a 6 percent issue. The details are shown in the last two
columns of Table 1.
62 Maurice Moonitz

Table 1
BEHAVIOR OF INVESTMENT IN A $1,000 BOND DURING A RISE
IN INTEREST RATES

Market Market Price Case A Case B

Rate of of $1,000,* Hold 4% Sell at f,, and
Date Particulars
Interest 10-year, Bond to Reinvest in
(Percentage) 4% Bond* Maturity 6% Bond

to 4 $1,000 Invest $1,000 in 4% bond $(1,000) $(1,000)

fi 4 1,000 First coupon (4% of $1,000) 40 40
t2 4 1,000 Second coupon 40 40
t3 5 942 Third coupon 40 40
t* 5 949 Fourth coupon 40 40
f5 6 916 Fifth coupon 40 40
Sell 4% bond 916
Invest in 6% bond (916)
^6 6 931 Sixth coupon (4% bond) 40
First coupon (6% bond = 6%
x $916) 55
t- 6 947 Seventh coupon (4% bond) 40
Second coupon (6% bond) 55
^8 6 963 Eighth coupon (4% bond) 40
Third coupon (6% bond) 55
to 6 981 Ninth coupon (4% bond) 40
Fourth coupon (6% bond) 55
^10 6 1,000 Tenth coupon (4% bond) 40
Fifth coupon (6% bond) 55
Redeem bonds—Face amount 1,000 916
Gain for ten-year period $ 400 $ 391

^Market price is calculated net of the coupon falling due on the respective dates. The formula for
market price is:

1
1
(1 + i)n
$1,000 x ---+ $40 x
(1 + 0"

where i is the market rate of interest at the valuation date, and n is the number of periods from
the valuation date to maturity. The first term in the bond formula above is the market value of
the principal, the single sum of $1,000 due at maturity. The second term is the market value of
the coupons, which constitute an ordinary annuity of $40 per period for n periods.
 Investments in Debt Securities 63

Case A—The Bond Is Held to Maturity

Under conventional accounting for commercial and industrial

companies, the investment in the $1,000, 4 percent, ten-year bond is
carried at $1,000 throughout the life of the bond. Each year one
coupon of $40 matures and is collected. The same amount is reported
as interest earned. The financial statements therefore report a steady
4 percent rate of return on this particular investment. This rate is
squarely in line with the rate contracted for when the investor lent
$1,000 to the issuer against a set of promises to pay $40 per year for
ten years, plus $1,000 in a lump sum at the end of the tenth year.
As far as the formal financial statements are concerned, conven¬
tional accounting conceals the drop, and subsequent rise, in the
market price of this bond. Where material, this drop would no
doubt be disclosed either by parenthetical notation in the balance
sheet or by other notes to the financial statements. The failure to
recognize the implicit holding loss (and subsequent gain) is there¬
fore not too serious as far as the asset side of the balance sheet is
concerned. Readers can substitute the disclosed market price for the
historical cost figure, if they so wish. But what the statements do
not disclose is the fact that investments of the type represented by
this bond are earning 5 percent for part of the ten-year period and 6
percent for another part. This type of information cannot be gleaned
from formal financial statements, parenthetical disclosures, or
notes. It is information external to the financial reports. Thus the
fact is not disclosed that the investor committed funds to a 4 percent
contract, when the funds could have been earning 5 percent and 6
percent on the same kind of investment for a large part of the
ten-year period.
If the type of conventional accounting just described is modified
to incorporate the changing market price of this bond in the formal
accounts, a loss of $58 will be reported at the end of the third
period, when the market rate of interest rises from 4 percent to 5
percent. (This loss of $58 will be offset in large part by interest
income of $40 from the third coupon.) The financial statements for
the fourth and fifth years will then report a rate of return of 5
percent from this investment. The 5 percent is represented by the
sum of the $40 coupon plus appreciation in market price, divided
by the market price of the bond at the beginning of the period.
Thus for the fourth period, total income from the investment equals
the $40 coupon, plus a $7 increase in market. And $47 is 5 percent
of the market price of $942 at the beginning of that period.
64 Maurice Moonitz

Another loss of $33 is reported at the beginning of the sixth year,

reflecting the rise in the market rate of interest from 5 percent to 6
percent. The financial reports will reflect a 6 percent rate of return
for the last five periods, a rate calculated by taking the sum of the
coupon ($40) and the appreciation in market price during each pe¬
riod and dividing that sum by the market price at the beginning of
the same period. Accordingly the incorporation of market prices in
the formal accounts not only has the advantage of reporting the
current realizable value of the investment. It also shows the loss
from holding a 4 percent (contract rate) investment in a 5 percent
and 6 percent market.
Incidentally, at maturity the $1,000 invested in the bond is freed,
regardless of the method of accounting pursued. At that time inves¬
tors can shift their money to 6 percent debt and enjoy a $60 return
each year, or they can elect to get out of debt securities and use the
money in some other way.

Case B—The Holder Chooses to Reinvest

In this case we assume a swap or switchover at the end of the
fifth period, ignoring transaction costs. At that time the holding
loss of $84 becomes explicit (realized) and is recognized under con¬
ventional accounting procedures. The $1,000, 4 percent bond is sold
for its market price of $916. The proceeds of $916 are immediately
reinvested in a 6 percent coupon bond. The drop in price resulting
from holding a 4 percent coupon bond in a 5 percent and 6 percent
market becomes manifest and cannot be ignored.
As in case A, conventional accounting would report a 4 percent
return ($40 coupon each year on an asset carried at $1,000) during the
first five years, even though the market rate of interest rose to 5
percent at the end of the third year and the market price of the bond
dropped accordingly. Conventional accounting then reports the loss
of $84, an investment in a new bond of $916, and a return of 6 percent
for the last five years of the ten-year period. The return of 6 percent is
represented by the coupon of $55 on the new bond. For the last five
years, the financial statements are squarely in accord with the reali¬
ties of the market because the market rate of interest remains un¬
changed. At maturity, with the end of the tenth year, the investors
receive the last coupon of $55 and principal of $916. They can then
reinvest the proceeds of $916 in another 6 percent debt security,
thereby continuing to receive $55 per year on the coupons, or they
can decide to invest in some other direction altogether.
 Investments in Debt Securities 65

If conventional accounting is modified to incorporate the changes

in market price, the financial statements will report a $58 holding
loss at the end of the third period, a $7 holding gain for the fourth
period, and a $33 holding loss at the end of the fifth period. These
three amounts net out to $84, the loss recognized at the time of the
switchover by conventional accounting. In each of the ten years the
financial statements will report a rate of return on the investment
equal to the market rate in effect during that period. The rate of
return for any period in which the market rate of interest does not
change is calculated by taking the algebraic sum of the coupon for
that period plus, or minus, the change in market price during the
period, and dividing that sum by the market price of the bond at
the beginning of that period.

Comparison of the Two Cases

We are now in a position to make some comparisons and raise a
question or two about the exchange of a 4 percent coupon bond for
a 6 percent coupon bond in a 6 percent market. Table 1 shows that
the net gain for the entire ten-year period from investing in a 4
percent coupon bond and holding it to maturity is $400. The net
gain for the same period from switching in midstream is $391, or $9
less than if the investor had stood pat.
The shortfall of $9 in net cash inflow in case B compared with case
A results from the earlier receipt of cash in the last five years for
case B. Case B yields cash from coupons of $55 each year of the last
five; case A yields $40. If the yearly excess of $15 were invested
promptly at 6 percent, it would amount to $84 at the end of the
tenth year:

$15 x 5.6371 = $84.

(5.6371 is the amount to which an ordinary annuity of one per

period will grow in five periods at 6 percent, compounded once per
period.)
The resultant amount of $84 is $9 more than the deposits of $75. It
is also exactly the amount of the loss that was recognized at the time
of the sale of the old 4 percent bond at the end of the fifth period
and the immediate reinvestment of the proceeds in the new 6 per¬
cent bond.
The important conclusion from these technical considerations is
that the swap created no income for the ten years taken as a whole.
It merely rearranged the cash movements, moving the coupon
66 Maurice Moonitz

payments forward in time. This rearrangement, however, occurs at

the cost of a reduced terminal value—in the case before us, a reduc¬
tion of $84. The fundamental question is this: where is the gain that
we hear so much about from bond swaps or switchovers?
The only way in which the reinvestment in case B can be made to
seem more profitable is to focus exclusively on the cash receipts
from coupons and to ignore the changes in market price, even the
one realized explicitly in the sale of the 4 percent bond at the end of
the fifth period. We are fully aware that a rearrangement of cash
movements may be of advantage to certain investors. They are the
best judges of the pattern of receipts and outlays that they want.
But that pattern involves working capital management, not the cal¬
culation of periodic profits or losses from operations. In ordinary
accounting terms we once again confront a confusion in many dis¬
cussions between the kinds of data that are to be included in an
income statement, broadly conceived, and the kinds to be included
in a statement of source and application of funds.
The preceding analysis also points up a type of locked-in effect for
investments in debt securities. If investors elect to stay with debt
securities of the type already held, and if interest rates rise, they can
shift to a security carrying the current market rate of interest, but
only at the cost of taking a loss on principal. In formal language, they
can only exchange one set of fixed promises—their original holdings
as in case A above—for another set of fixed promises with the equiva¬
lent (discounted) value. This is illustrated in detail in case B.
If they want to stay in debt securities, investors should hold on to
their investments until maturity. At that time they get $1,000 and
can reinvest in a 6 percent bond, as in our example above, and
enjoy a $60 per annum return from that point on. Referring again to
case B, they could take the extra $15 from each coupon during the
last five years, reinvest it at 6 percent, the current market rate, and
get back the $84 lost at the date of the switchover. But this is
equivalent to considering just $40 of each coupon during the last
five years as earned, the result investors get if they stand pat on the
original 4 percent coupon investment.
Investors have another option. They can elect to get out of debt
securities altogether. For example, assume that in case B investors
expect equity securities to yield more than the 6 percent rate pre¬
vailing in the market for debt. They will then be better off to sell the
4 percent bond and reinvest the proceeds in a nondebt security.
A further point implied by the preceding analysis is that the
historical-cost basis of carrying an investment in debt securities
 Investments in Debt Securities 67

leaves the way open for income management by investors. If market

rates of interest drop, the market value of the holdings will in¬
crease; but the increase will not be recognized until and unless the
investors choose to sell. If market rates rise, the holdings drop in
market value; but again, the change will not be recognized until the
owners choose to sell. A thoroughgoing market-price basis will
make it impossible for investors to manage reported income in this
manner. Furthermore it opens the way to the use of realizable profit
or loss as the preferred method of measuring the results of invest¬
ment activity.
One other important observation is in order. In a sense the loss
from a rise in interest rates, implicit or explicit, is timeless. It does
not accrue, as we ordinarily use that term. It results from a change
in discount factor, from 4 percent in our example to 5 percent, and
then to 6 percent. In ordinary accounting terms, it clearly qualifies
as an extraordinary item, even as a prior period adjustment in re¬
tained earnings. This kind of loss—or gain, if interest rates drop—is
different from the gain we showed above, for example, in the mar¬
ket price of the 4 percent bond during the last five years. The gain
in that case came about because the future cash receipts, especially
the large principal payment at the end of the tenth year, moved one
period closer, but with no change in the market rate of interest used as
a discount factor. We should have no difficulty in agreeing upon the
nature of a rise in market price from this cause, even though we
may disagree on the precise time at which it should be recognized
in the formal financial statements. Market appreciation and declina¬
tion resulting from changing interest rates, however, are of a differ¬
ent type. In the accounts they should not be combined or classed
with the results of operations directly attributable to the passage of
time.

Bond Swaps of Public Employees' Retirement Systems

Highlighting some of the points made above are the problems

encountered by certain public employees' retirement systems
(PERS) or pension funds. Two types of events related to our discus¬
sion occur in the management and operation of these funds.
1. A PERS sells its low-coupon, tax-exempt municipal and state
bonds and invests the proceeds in higher-coupon corporate bonds.
For the ordinary investor the yield from these corporate bonds is
fully taxable under existing provisions of the U.S. Internal Revenue
68 Maurice Moonitz

Code. But the PERS itself is exempt from U.S. income tax. As a
result, the swap is clearly advantageous. It is only-possible because
the rules governing these retirement systems permit them to invest
in high-grade corporate bonds.
This type of swap should create no special accounting problems.
For example, a PERS invests $1,000 in a 4 percent tax-exempt mu¬
nicipal bond, principal amount, $1,000. It sells this bond for $1,000
and invests the proceeds in a 7 percent corporate bond, paying par
for the new bond. The swap itself gives rise to no gain or loss on
the books of the PERS. Its income rises from $40 to $70 per year,
without the loss of principal we encountered in case B. Such arbi¬
trage is possible in this instance because the markets for the two
classes of bonds are distinct—the market for tax exempts being
characterized by relatively low coupon rates, and the one for taxable
corporates being characterized by relatively high coupon rates. A
PERS is in a position to move between these two markets without
affecting its tax-exempt status. It is therefore in a position to regis¬
ter a real gain by making a substantive change in its investment
pattern.
2. A PERS sells its low-coupon bonds and invests the proceeds in
higher-coupon bonds of the same class (for example, high-grade cor¬
porate bonds) in order to increase its cash flow. In this case the
PERS may object to recording the loss on the sale of the old bond as
realized all in one year. Instead the PERS proposes to defer the loss
and amortize it over the remaining life of the new issue. But to
defer the loss contradicts the known facts of this case. The loss is
the cumulative result of the changes in interest rates that have oc¬
curred since the initial investment in the bonds. It belongs to the
past. The latest date for it to be reflected in income or retained
earnings is the date of the swap.
As we can see, this is precisely the kind of swap analyzed at some
length in case B. The analysis there is applicable here. In brief, the
so-called gain from increased cash flow is offset by the loss on the
sale of the old bond. On balance, for the life of the two issues taken
together, no gain or loss emerges.
As long as a PERS is required by law or regulation to follow
generally accepted accounting principles it should account for its
operations as though it were a profit-seeking investor. As a policy
matter, a special exemption for such entities might be acceptable,
but not the pretense that their bond transactions are somehow dif¬
ferent from other investors' bond transactions, and therefore that
the accounting principle applied to them is different from the one
 Investments in Debt Securities 69

applied to investors generally. Different principles for different pur¬

poses lead to confusion and chaos. The path should rather be to¬
ward narrowing the area of differences in accounting.

Determination of the Market Rate of Interest

The previous discussion assumes that we know or can readily

determine the market rate of interest applicable to an investment in
a debt security. As a practical matter, market rates vary for different
kinds of securities, so that a constellation of rates seems to exist
rather than a single rate applicable to a broad class. The selection of
the rate to be used as a discount factor then becomes of some
importance if the debt security is not listed, and therefore no
quoted market value is available. In that case, the market rate to be
used as a discount factor has to be estimated.
The Accounting Principles Board (APB) issued Opinion No. 21,
“Interest on Receivables and Payables" in August 1971. That Opin¬
ion is not directly concerned with bonds or the other forms of long¬
term debt that we have been discussing. Instead it is concerned
primarily with so-called non-interest-bearing debt instruments and
those carrying rates of interest on their face so low as to be entirely
unrealistic. In those cases the APB requires the calculation of a
realistic present value of the receivable or payable by the use of an
interest rate as a discount factor. The discussion in Opinion No. 21
of the ways to select an appropriate rate of interest is what should
be helpful in resolving the problems with which we are concerned.

Realized Gains, Unrealized Gains, and Market Values

The central point of this section is that attempts to combine prof¬

its measured by realization with profits measured by changes in
market values in a single calculation of net income will at best
confuse and at worst mislead.2
If historical cost and its twin, realization by sale, are to be the
basis of accounting for investments in debt securities, then changes
in market price should not affect the calculation of net income or the
totals shown in the balance sheet for assets, liabilities, and owners'
equity. Market values of securities held at balance-sheet date can be

2. The discussion in this section draws heavily upon Chapters 4 and 5 of Storey
and Moonitz, Market Value Methods.
70 Maurice Moonitz

disclosed parenthetically or by note, but they should not otherwise

enter into the financial statements.
If market values are to be the basis of accounting, then realized
gains (the difference between acquisition cost and proceeds from
sale) should not be included as an element of net income. Cost basis
of securities still held can be disclosed parenthetically or by note in
the balance sheet; the difference between acquisition cost and sales
proceeds can be shown parenthetically or by note in the statement
of source and application of funds, but it too should not enter into
the financial statements reporting the results of operations.
To convey the reasons underlying the conclusions just stated, we
need some symbols. Capital letters represent items that could ap¬
pear in financial statements—balance sheets, income statements,
statements of retained earnings, and funds statements. Lower-case
letters identify the particular securities involved:

M is the market value of securities held.

C is the cost of securities acquired, held, or sold.
P are the proceeds (cash received) from the sale of securities.
G is the gain or loss under the market-value method, exclusive of inter¬
est income on the debt instruments.
R is the gain or loss under the cost method, exclusive of interest in¬
come.
b designates securities held at the beginning of the period.
e designates securities held at the end of the period.
a designates securities acquired during the period.
s designates securities sold during the period.

G is an interesting element. It consists of changes in the market

values of four mutually exclusive classes of securities. We therefore
create four subdivisions of G:

Gw denotes the change, from the beginning of the period to its end, in
the market value of securities held throughout the period.
Gx denotes the change, from acquisition to sale, in the market value of
securities bought and sold within the period.
Gy denotes the change during the period in the market value of securi¬
ties held at the beginning of the period but sold within the period.
Gz denotes the change in the market value of securities bought within
the period and still held at its end.

We now set forth the differences between the cost and market-
value methods:
 Investments in Debt Securities 71

Market Value Cost

Mb + Ca + C — Ps = Me Balance sheet Cb + Ca — Cs = Ce
G = Gw + Gx + Gy + Gz Income statement R = Ps - Cs
Ps — Ca = F Funds statement Ps — Ca = F

Note: F is the net inflow or outflow of funds from investments in debt securities.

This tabulation indicates that the funds statement is the only one
that contains identical data under both methods. The neutrality of
the funds statement toward alternative valuation formulas is one of
its greatest strengths and should be built on. The proper place to
show what was received and what was spent is the funds state¬
ment, not the income statement. The income statement is the prop¬
er place to show the results of operations, measured by accrual
accounting as accurately and with as much sophistication as is at¬
tainable at any particular moment.
The formulas for the income statement indicate that G and R are
incompatible concepts and measures. Gx, the gain or loss on securi¬
ties bought and sold in this period, is an element in R; but no other
part of gain or loss under the market-value method is found in the
gain or loss under the cost method.
What then of the proposals to report realized and unrealized
gains and losses in the financial statements? To complete the analy¬
sis, we need additional symbols:

U denotes the unrealized gain or loss in our holdings of securities.

Ub denotes the unrealized gain or loss at the beginning of the period,
Ub = Mb - Cb.
Ue denotes the unrealized gain or loss at the end of the period,
Ue = Me - Ce.

A basic arithmetic relationship now emerges: Ub + G — R = Ue. In

terms of a credit-balance account for unrealized gain, this relation¬
ship describes how that account changes—the balance unrealized at
the beginning of the period, plus the increase in the market value of
all securities held at any time during this period, less the amount of
gain realized on those securities sold in this period equals the bal¬
ance unrealized at the end of the period. A simple rearrangement of
the symbols reveals a useful shortcut: G = (Ue — Ub) + R.
This form of the arithmetic relationship indicates that we do not
have to calculate the change in market value during the period for
each of the four elements of G, namely, Gw, Gx, Gy, and Gz. Instead
we merely have to value the portfolio at the beginning and end of
the period, take the difference (Ue — Ub), and add the realized gain
72 Maurice Moonitz

from sales during the period in order to get the total amount of G.
This is a useful work-sheet technique, but hardly a.'solid basis for
reporting these elements in the financial statements. The central
point is that G and R are incompatible and noncomparable magni¬
tudes for conventional accounting periods of a year or less.
If we include realized gains and the change in unrealized gains in
the same income statement —that is, if we add together R and (Ue —
Ub)—we get G, the gain under a pure market-value basis. Why then
bother to separate G into two incompatible elements? In that case
both the balance sheet and the income statement are, in principle, on
a pure market-value basis. The market-value method should there¬
fore be used consistently throughout the financial reporting system.
If we exclude unrealized gains from the calculation of income and
from owners' equity, we have a spurious type of market-value
method. The income statement is on a historical-cost basis, and so
is the balance sheet; but the balance sheet looks as though it is on
the market-value basis. A more straightforward approach is to stay
with the historical-cost basis in the formal financial statements, and
to disclose market value of the securities parenthetically or by note.
Inclusion of both realized and unrealized elements, so labeled, in
the same set of financial statements is sometimes defended as a
transitional device, a means of smoothing the shift from accounting
on a historical-cost basis to the use of a market-value (or other
current value) basis. The motive is commendable; but the tech¬
nique, as we have seen, is logically faulty and risks confusing or
misleading the reader of the financial statements. It would be more
straightforward first to calculate net income on a market-value basis
and transfer it to retained earnings, after which management as a
precautionary or conservative step could earmark a portion of re¬
tained earnings as restricted.
This procedure makes a clear separation between the measurement
problem, reflected principally in an income statement calculated on
the basis of the most up-to-date information available, and the policy
decision as to the proper disposition of the reported net income. To
the fullest extent possible, we should resist the incorporation of man¬
agement's policies and biases into the steps we take to measure
financial position and results of operations. To be most useful, ac¬
counting measurements need to be neutral toward the competing
and sometimes conflicting interests of the various groups that have a
legitimate interest in the affairs of the accounting entity.

September 1976
The Hard Core of Accounting
R.J. Chambers

Through the work of William Andrew Paton there runs a strand

of confident assurance of the essential serviceability of accounting.
Its function is indispensable in the conduct of economic affairs; it
needs neither exaggeration, nor bombast, nor apologetics. In that
confidence Paton has striven as a painstaking craftsman so that
accounting would come to serve the financial community as widely
and as well as human intellect can make it. He has always been
aware of blemishes in the art and in its theoretical foundations; his
characteristic reaction has been to seek out the principle at issue
and to explore afresh the source or cause of an apparent difficulty or
dilemma. It seems beyond doubt that, if his successors had fol¬
lowed up more of his many suggestions, the present state of ac¬
counting and accounting theory might have been happier than it
appears to be.
The last two decades have witnessed a growing uneasiness about
the possibility of developing firm and well-founded accounting
principles. Rules and principles which once were asserted confi¬
dently are now hedged about with a plethora of provisos, qualifica¬
tions, and exceptions. There is perfervid talk about problems,
issues, and questions but no corresponding drive to solve, resolve,
or answer them.
There are signs of a belief that accounting problems cannot be
solved by intellectual effort but must be tackled by the procedures of
quasi-political compromise. But the belief is defeatist—and otiose.
For, if we must make compromises, we should need to know what is
being sacrificed for what; we need some notions of what is ideal.
We are told frequently that accounting is a man-made device, and
that its doctrines and rules can never be as precise as those of the

73
74 R. J. Chambers

sciences. But this is mere excuse, and false in substance. The body
of knowledge popularly described as "science" is als'o man made. It
is under constant examination and reconstruction. As for its preci¬
sion, it is precise only at the level of principles; at the level of
technological application it is no more precise or exact than human
skill and judgment can make it.
We are told that there are no fundamental truths which underlie
and which could give firm shape to the practice of accounting. But
this is only quibbling. Artisans who have no knowledge of funda¬
mental truths have pursued their work with great skill, simply by
observing carefully the outcome of their work and accepting the
discipline that the characteristics of their materials impose.
We are told that there is no singular concept of income, that no
single method of general purpose accounting is "best," and that
rules and practices must remain matters of judgment. This is mis¬
chievous. Judgment is an element of all technical arts. But if we
have no firm idea of what is serviceable in a given setting, we have
no way of determining whether a particular judgment is good or
bad, and hence reliable or unreliable.
Such misconceptions as these threaten to make accounting like
clay in the hands of an artless potter; or, to change the simile, like a
clump of tumbleweed, blown about by every shift in the economic
and political breeze. This may not have mattered when accounts
were largely a matter of private concern. But now, to a much greater
degree than fifty years ago, financial statements and the informa¬
tion they contain are used in a multiplicity of settings in which
conflicts of interest arise. These conflicts turn on the disposal of
money and money's worth. It is the concern of accounting with
money and money's worth (or, generally, with dated money
equivalents) which prompts the title of this paper—"The Hard Core
of Accounting." For if accounting deals with flows and holdings of
money and money's worth, its rules and processes must be disci¬
plined by what can legitimately be said from time to time about
those flows and holdings, and by the use which can properly be
made of what is said.

The Variety of Financial Calculations

The forms and uses of financial calculations, statements, and re¬

ports are numerous and varied. Expositions of accounting com¬
monly deal at length with their individual contents or components,
treating each class of statement to a considerable degree in isolation
 The Hard Core of Accounting 75

from the others. We shall deal with all statements here as systemati¬
cally related or relatable, for that is the way they are used in the
reasoned conduct of financial affairs. For reasons which will be¬
come apparent we will adopt a temporal classification.
There are past factual calculations, describing the financial conse¬
quences of actions and events up to specified past dates. For pres¬
ent purposes, "past dates" will refer to all dates other than those
defining the most recent period and balancing date. An income
account will be taken to represent the aggregate change in the
amount of net assets during the period ended on any such date. The
sources, varieties, or subclasses of revenues and charges against
revenues may be specified. But in essence the account is an aggre¬
gative account, explaining the change in the net wealth or net assets
of a firm. A balance sheet will be taken to represent the financial
position of the firm at a stated date. That position is indicated by
subaggregates of assets and equities of different kinds. Because
different kinds of assets and equities subject the firm to different
risks or obligations and put it in a position to gain from different
kinds of operation or events, judgments of the past and of the
future turn on the composition of assets and equities from time to
time. Successive balance sheets represent the shifts in the composi¬
tion of assets and equities during the periods they enclose.
There are contemporary factual calculations. For present purposes
"contemporary factual" will relate to statements covering the imme¬
diately past period and the latest balancing date. To take a fine point,
these statements are past factual, for all factual positions and results
are discovered after the event. But since the most recent statements
have a more direct bearing on judgments about future actions than
more remote statements, it is useful to classify them separately.
The components or contents of all retrospective financial state¬
ments, past factual and contemporary factual, are determinate in
principle. The actions have been taken; the events have occurred;
their financial consequences within a past period up to a stated date
are or have been observable. Those consequences are indepen¬
dently testable.
There are exploratory calculations regarding the possible outcomes
of alternative future courses of action. The most widely ac¬
knowledged form of exploratory calculation is the calculation of pres¬
ent values or discounted cash flows. A present value is an aggrega¬
tive figure, in the same general fashion that an income figure is
aggregative. It is the discounted expected consequence of the use in a
specific way of a complex of assets. But just as an income statement
76 R. J. Chambers

represents only one aspect of the consequences of past actions and

events, so does a present value calculation represent only one aspect
of contemplated actions and events. Other aspects are represented
in exploratory calculations of periodical incomes and financial posi¬
tions. These are no less necessary than present value calculations,
for any choice from among alternative courses of action will turn, in
part, on the actual timing of net receipts (for that bears on dividend
paying and other similar prospects) and on the expected future
compositions of assets and equities. There will, of course, be ade¬
quate exploratory calculations in respect of each alternative course
of action available at the time of any choice of actions.
Into these exploratory calculations will enter the purchase prices
of new assets, the expected selling prices of present assets, and the
replacement prices of substitute assets if replacement is one of the
possible future actions. Commonly there are more ways than one of
doing a given technical job, each entailing potentially different out¬
lays and inflows. The only purchase price or replacement price
which becomes of consequence to the firm is that for the chosen
course of action; and it becomes of consequence only at the time of
choice. Only for that purpose (and consequently for budgets em¬
bracing that choice) are replacement prices pertinent.
There are anticipatory calculations in respect of the outcome of a
selected course of future action. These are embodied in budget
statements of operations, income, and resultant financial position,
representing the changes in the amount of net assets and in the
composition of assets and equities that are expected to flow from a
selected complex of actions. In principle a budgetary calculation will
be one of the sets of exploratory calculations mentioned above,
though in practice it may be more elaborate. The function of budget
statements is to give direction and guidance to the activities of the
firm, and to provide a point of reference for any revision of opera¬
tions or expectations in the light of experiences during the course of
the budget period.
There are also auxiliary calculations in respect of particular facets
of the operations of business firms, calculations or product costs,
process costs, sectional costs, and so on. These may be retrospec¬
tive, exploratory, or anticipatory.

The Linkages of Past, Present, and Future

The past, which is the subject of retrospective calculations, is

fixed, unalterable. Whatever can be known about the facts of the
 The Hard Core of Accounting 77

past can be known at the end of each successive period. Certainly it

cannot be known whether some action taken in the past, but not
completely consummated at any present date, will turn out in the
future as was intended or expected. But the progress made up to
any present date is discoverable; and when it is discovered one can
then decide whether to continue with any such action or to aban¬
don it, in the light of past progress and the currently expected
outcome. (The originally expected outcome is no longer relevant, for
the circumstances of the firm are no longer what they were when
the action was taken.)
Exploratory and anticipatory calculations are distinctively differ¬
ent from retrospective calculations. The firm does not face a known
or determinate "state of nature." Neither the firm's future environ¬
ment nor its own operations in that environment can be considered
fixed and determinate. The future is "to be made." Every action of
the firm and every interaction of it with firms and other persons
will influence its results and position in ways unknown at the time
of calculation. All calculations relating to the future are imaginative
constructions. Under these unstructured conditions, there is noth¬
ing firm which would provide a basis for assigning probabilities to
possible outcomes.
The base point of these imaginative constructions of the future is a
known present: for, first, no calculation about the future can be made
without knowing the means available for the operations contem¬
plated; and, second, no future position can be preferred to a present
position without knowing that present position. All other compo¬
nents of exploratory and anticipatory calculations are expectations or
guesswork. They may be founded, in part, on known past trends and
currently known facts; but in principle they are hypothetical or heu¬
ristic. Of course, in selecting these hypothetical components, the
imaginations of managers and their aides are not foot-loose and
fancy-free. In the most general terms, they are constrained by the
financial position at the time, and they are constrained by the wish
or preference to be better off or in some better position at a future
time.
Further, although temporally different magnitudes must be de¬
rived by different processes, the objects or properties of which they
are magnitudes must be conceptually similar. "Income," for ex¬
ample, must refer to the same kind of thing in the term "attained
income" as it does in "expected income." The income we strive for
is the income we expect to spend, or use otherwise, if we attain it.
If "expected income" and "attained income" did not refer to a
78 R. J. Chambers

common and particular notion of income, there would be no sense

in comparing one with the other; no sense in sayihg, for example,
that a firm's income was better than expected. The same applies to
all other notions which have a past manifestation and are the sub¬
ject of a future expectation. There are thus not different concepts of
past and future magnitudes; there are simply differently dated
quantifications of the same concept. This terminological linkage is
essential to a coherent discussion of the coherent progression of
thoughtfully managed enterprises of any kind.
These generalizations may be illustrated by reference to a specific
type of problem situation. Suppose a firm's debt to equity ratio has
been rising (past factual knowledge). Suppose the ratio has now
reached a point where, under an existing bond indenture or other
contract, the firm has little reserve borrowing power (present factual
knowledge). And suppose it has in contemplation a future course of
action requiring more cash than is expected to be available in the
ordinary course of business:

1. The firm may make a new issue of shares, simultaneously reducing

its debt to equity ratio and providing the required cash. Whether it will
decide to do this will depend on its past rates of return and dividend
record, its expected future returns from all sources, and its present
knowledge of the state of the security investment market.
2. Or the firm may vary its present asset holdings, disposing of some
to acquire the requisite cash. To consider this will entail a set of calcula¬
tions of the (future) exploratory type, based on the present position and
knowledge of the past, for each of the assets or combinations of assets it
seems feasible to realize, and for the contemplated action and any auxili¬
ary actions made possible or necessary by the sale of present assets. The
present values, and the consequential results and positions, of all feasible
variants would be compared.
3. Or the firm may vary its present policies to make available the
requisite cash. It may reduce credit given, or the inventory levels it
holds, or the dividends it plans to pay.

In appraising these possibilities the firm will make use of past and
present knowledge and make judgments about the future reactions
to or consequences of the variations considered.
In any such exercise, the products of calculations relating to the
past, present, and future are all engaged, each in its own way.
There are many such "actions in contemplation": the retirement of
debt and the settlement of other obligations; expansion or reduction
in the scale of business; variations in wage and salary levels and in
labor usage; claims for governmental aid or bounties; reactions to
changes in exchange rates in respect of foreign purchases; sales of
 The Hard Core of Accounting 79

investments; and so on. In every case use will be made of some

combination of past and present information and future expecta¬
tions. Since the processes of accommodating the firm to the numer¬
ous and varied shifts of circumstance that it may encounter are
well-nigh continuous, its accounting system must stand ready, con¬
tinuously, to supply information pertinent to all possible problem
situations. That information, we shall hold, will necessarily repre¬
sent stocks and flows of money and money's worth.

Money and Money's Worth

Every shift in the circumstances of a firm and every possible

response has a money-quantifiable or financial aspect. Each, of
course, has other aspects; and decisions must be made in the light
of all aspects of the actions in contemplation. Each aspect must
therefore be assessed severally before a judgment is made or a pre¬
ferred course of action determined—“all things considered."
Physical, commercial, technical, economic, and financial aspects
are separately conceivable, each in its own terms or its own appro¬
priate “language." But it is not possible to discuss or describe any
one of them in the language of another. Physical characteristics of
assets are not described in monetary terms or financial aspects in
physical terms. There is no determinate and invariant relationship
between technical or physical features and the financial features of
assets. Given the price of a five-horsepower motor there is no way
of deriving from it the price of one-horsepower of “service poten¬
tial"; given the price of a five-passenger automobile, there is no
way of deriving from it the price of "one-passenger service," or
"ten-passenger service." However, the traditional calculation of de¬
preciation charges and depreciated asset values, being based on
expected output or service life or service potential (all of them non-
financial notions), presumes that there is such a relationship. This
long-standing solecism has persisted in financial accounting, in lieu
of the direct discovery, by reference to market prices, of the peri¬
odical financial magnitudes of depreciation and asset values. It has
yet to be recognized that there is no such thing as a physico-finan-
cial magnitude, or a physico-financial income statement and bal¬
ance sheet.
To return to money. Money flows in, money flows out, and
money balances are held from time to time. The survival of every
firm depends on its continued ability to command money receipts
and to pay money out for goods and services, wages and salaries,
80 R. J. Chambers

rates and taxes, interest and dividends, and the retirement of debt.
The importance of stocks and flows of money is. undeniable. The
traditional financial statements, however, include accruals and val¬
uation adjustments which are not the amounts or consequences of
money flows. The emergence of the so-called funds statement or
cash flow statement is token recognition of the interest of users of
financial statements in money amounts and money flows. The
whole of the recent emphasis on the use of accounts for the predic¬
tion of cash flows evidences at least a ventral apprehension of the
pertinence of money amounts and flows. Present value or "dis¬
counted cash flow" calculations seem also to command support on
the ground that the arbitrary "cost allocations" made in traditional
accounting are avoided. We shall argue, however, that these are
partial and unintegrated responses to a basic deficiency of financial
accounting.
The liquidation or conversion to cash of nonmonetary assets, in¬
ventories in particular, is the principal way in which firms acquire
the means of meeting day-to-day obligations and adapting their
operations to their circumstances. But cash holdings are precaution¬
ary, not profit making. Cash which is surplus to shortly expected
requirements is invested in other forms; and at any time cash hold¬
ings are but a small proportion of the total asset holdings of a firm.
The firm's "cash position" thus represents only a small part of its
aggregate financial position. Changes in the money's worth (market
resale prices) of noncash assets also affect the firm's financial posi¬
tion and results; and their magnitudes are just as relevant to inter¬
nal and external judgments about the firm as are movements in cash
holdings.
The traditional accounting process does not accrue these changes
as they occur. And because the funds statement is, in principle, a
derivative of the income account and balance sheet, it can do no
better. Taking the three statements together, their users remain un¬
informed about such accrued changes in the money's worth of non¬
cash assets as are disregarded, for whatever reason, in the account¬
ing process. But all vendible assets are potentially convertible to
cash; they represent accessible cash to the extent of their resale
prices at any time. They are "as good as cash" to that extent, for all
financial calculations. The amounts by which they are represented,
even when they do not represent money's worth or money equiva¬
lents, are included in calculations of asset-backing and debt cover¬
age and rates of return; that is, these assets are treated as if they
 The Hard Core of Accounting 81

were as good as money. If the income account and balance sheet

were so designed that they represented the consequences of all
shifts in money and the money's worth of other assets and in
money obligations, the case for an auxiliary funds statement would
collapse.
A further argument against the funds statement shows how tenu¬
ous is the belief in the usefulness of financial statements as indica¬
tors of prospective cash flows to investors and others. There is no
necessary relationship between aggregate cash flows and any other
financial aggregate. The management of the composition of assets is
opportunistic. Funds which are surplus to the ordinary trading
functions of a firm will be laid out in the most promising form of
expansion, renovation, or other investment. There is no way of
knowing in advance what opportunities will open up or close, what
changes in governmental, commercial, or fiscal policies will influ¬
ence the availability of and the demand for cash, or even what
pressures on the part of competition and factor suppliers will affect
the inflows and outflows from ordinary trading functions. If, as we
have said, the future is yet to be made, there is no way in which
any factual financial information published at a given date can yield
outsiders a reliable indication of prospective cash flows. But it can
be expected that dated financial information will indicate the finan¬
cial potential of a firm at that date, the amount of money and
money's worth which it commands at that date. This aggregate is
intelligible and usable; and reference to it may enable outside users
of financial statements to estimate at least the impact of external
events as they occur.

Present Value Calculations

Present value calculations, or discounted cash flow calculations,

are typically internal to the firm. Under our earlier classification
they are exploratory, not anticipatory. Their products do not enter
into budgets, although their component (undiscounted) figures for
a selected course of action may do so, and thus may yield an indica¬
tion of expected cash inflows and outflows. But the products of
present value calculations serve a different function. They are used
in the course of choosing, by people who are aware of (or can be
aware of) the hypothetical or imaginative elements of their compo¬
nents, and can therefore fairly judge between alternative prospec¬
tive projects.
82 R. J. Chambers

By intention, present value calculations exclude the traditional

provisions, such as depreciation charges, on the ground that these
do not represent real outflows of money. Money outflows are
deemed to occur at the date of purchase of the durables necessary
to any project; and inflows from disposal of these assets are
deemed to occur at the terminal date. In effect the outflow is
treated as a sunk cost and the assets as having no recurrent peri¬
odical financial significance.
But we have contended that money inflows and outflows are no
different in effect from changes in the money equivalents of non¬
monetary assets. The object of trading is an increment in wealth,
not merely increments in cash. It follows that the test of alternative
proposals is the discounted changes in wealth or money's worth
from each proposal.
The use of present value calculations for commercial projects is
derived from their use in purely financial settings. And those set¬
tings provide additional justification for the view we have just ex¬
pressed. First, in all compounding and discounting calculations it is
assumed that the accumulated sum at the end of one period is the
sum invested in the next. It does not have to be received in cash; it
is simply accrued, since it is "as good as" a receipt in cash. Second,
the typical finance textbook represents the rate of return on a secu¬
rity investment by a formula of the style, R = (D + AP)IP, where R
is the rate of return, D is the dividend paid in a period, P is the
price paid for it or its price at the beginning of a period, and AP is
the change during the period in the price of the security. Here also
the change in the money equivalent of the security enters into the
calculation, for it is "as good as" cash. If an exploratory calculation
of present value is made for a succession of future periods, the
return for each period would properly include the increment or
decrement in the "cash value" or money equivalent of the invest¬
ment in each period; otherwise the result will not represent the
effect of the temporal pattern of net flows.
The formal consequences of making present value calculations in
the traditional way and in the manner suggested by the preceding
argument may be illustrated by a simple example.
Suppose that one of the projects being considered by a firm at a
certain time entails the purchase of a machine at the end of year 0,
its use for five years, and its disposal at the end of the fifth year.
Suppose that all other expected receipts and payments occurred at
the ends of the years. Take the discount rate as 10 percent per
annum. The traditional calculation is as follows:
 The Hard Core of Accounting 83

Present Value—Cash Flows Only (traditional calculation)

Year 0 1 2 3 4 5
Revenues $ $400 $600 $600 $600 $600
Purchase (sale)
of machine -1,000 200
Operating expenses -200 -300 -300 -300 -300
Net cash inflow -1,000 200 300 300 300 500
Present values
of inflows $-1,000 $182 $248 $225 $205 $310
Net present value
of project $ 170

This calculation implies an immediate decline in wealth of $1,000

at the outset and a "rediscovery" of $200 on sale of the machine. But
if that decline were not immediate—that is to say, if the firm ex¬
pected residual money equivalents of the machine at the end of
each year—the traditional calculation would not represent the flows
of wealth in a temporally correct manner.
Suppose then that the expected money equivalents (resale prices)
of the machine at the ends of the years were those indicated in the
following revised or all-inclusive calculation, and that in all other
respects the data were the same. The calculation would be as follows:

Present Value—Money Equivalent (M.E.) Flows (inclusive calculation)

Year 0 1 2 3 4 5
Closing M.E. of machine $ $500 $400 $300 $200 $200
Opening M.E. of machine 1,000 500 400 300 200
Net M.E. inflow, (a)-(b) -1,000 500 -100 -100 -100 -

Revenues 400 600 600 600 600

Operating expenses -200 -300 -300 -300 -300
Total M.E.inflow, (c) + (d)-(e) -1,000 700 200 200 200 300
Present values of inflows $-1,000 $636 $165 $150 $137 $186
Net present value
of project $274

This inclusive calculation represents that, for example, at the end

of the first year the firm expects to hold an asset which is "as good
as cash" to the extent of $500. An equivalent amount becomes an
outlay, or outflow, in the second year, as in line (b). The closing
cash value, resale price, or money equivalent, of the asset in each
period is just as properly an offset to the outlays of the period as the
$200 cash receipt on sale of'the machine is an offset at the end of the
project.
84 R. J. Chambers

The textbook treatment of present value calculations makes a

point of the different timing or dating of expected receipts; early
receipts are discounted less heavily than later receipts. But it disre¬
gards the timing of expected diminutions in the money's worth of
holdings, whereas these are no less significant than the timing of
receipts in choosing between projects.
Standing by itself, a present value of a project has no more sig¬
nificance than as a calculation of a hypothetical outcome of hypo¬
thetical events. Both of the above calculations are identical in this
respect. Present values do not represent the market values or market
prices of projects, except in the case of perfect foreknowledge of the
outcomes of projects, and this is not possible for purchases and
sales of goods and services. But given that the decision-maker, or
the calculator in a decision-making setting, knows the hypotheses
on which the calculations are made, the present values of alterna¬
tive projects may legitimately be compared.
The above calculations are made in respect of a given project
which is expected to result in an aggregate net inflow of $600 over
the five years. But we may put the two sets of figures in a decision¬
making setting by taking them as inclusive calculations for two
distinct projects, T and R respectively, expected to yield the same
net money equivalent inflows. In this case the T-machine is as¬
sumed to be a nonvendible durable over the five-year period,
whereas the R-machine is vendible throughout the five-year period
(or the R project is some other project making no use of a machine
but otherwise having the same inflows and outflows and vendible
asset balances as the R example).
Now, clearly, R is more advantageous than T; for under T there is
no way of easing the adaptation of the firm to a change in technol¬
ogy or a fall in demand for the machine's products, whereas under
R the money equivalent of the machine could, at any appropriate
time, be redeployed in such circumstances. The advantage shows
up in the net present values (taking them as ''inclusive calculations"
for different projects), namely PV(T) = $170 andPV(R) = $274. If, on
the other hand, the present values of the two projects had been
calculated in the traditional manner (that is, for cash flows only),
both projects would appear to have the same present value; there
would be no quantified difference between them, although there is
a commonsense and practical difference. The decision-maker would
have no quantified basis for discriminating between them. That the
"inclusive calculation" provides such an explicit, quantified basis
 The Hard Core of Accounting 85

for choosing between the projects is the first advantage it has over
the traditional calculation.
A second advantage lies in the possibility of integrating the calcu¬
lations and their results with other calculations. The periodical net
cash inflows used in traditional present value calculations are not
relatable to the gross gain of the firm from any chosen project in¬
volving a vendible durable. Under the inclusive calculation, how¬
ever, the periodical net money equivalent inflows are estimates of
the periodical gains from the project. The component figures could
be entered in the exploratory financial statements mentioned ear¬
lier, and in the budget statements for any chosen project; in the
latter case they would also serve as the basis for subsequent reviews
of the progress of the project. The present value calculation and the
budgetary calculation would relate to the same elements and conse¬
quences of the project. The result would be considerable economies
of cerebration on the part of managers who make use of the prod¬
ucts of the several kinds of calculation; for instead of having men¬
tally to relate discounted cash flows, periodical expected net cash
flows, and periodical expected (all-inclusive) gains from projects,
they would be dealing with increments which were all of one
kind—all-inclusive increments in money-quantified wealth.
One additional excursion, into the quantification of profit, will
provide some further indications of the possibility of integrating
the varied kinds of calculations we have considered.

Profit

It has long been held that income is the resultant of matching

some quantum of charges, determined largely by judgments, with
the amount of periodical gross revenue. The balance sheet is, on
this view, a statement of the balances remaining after clearing those
charges and revenues from the accounts. The income account and
the balance sheet are formally articulated. But the rules which influ¬
ence the judgments have been found to yield results which, in some
settings and in some senses, are unrealistic. There have conse¬
quently been proposals to use different quantification rules for the
two statements, and hence to abandon their articulation. Indeed,
there have been proposals to abandon the aggregation of balance
sheet items, prompted by recognition of the fallacy of adding quan¬
tities of different kinds. These proposals for disarticulation and dis¬
aggregation can only lead to chaos. They seem designed to preserve
86 R. J. Chambers

the traditional rules rather than to force their rationalization or

modification in the interest of systematic and realistic accounting.
Let assets be defined as severable goods, and let them be repre¬
sented by their money amounts or money equivalents. Let equities
be defined as the money-quantified equitable rights in the assets.
Let liabilities be defined as equities ranking for settlement ahead of
ownership equities. The residual amount of ownership equities may
be described as net assets. Let income be defined as the periodical
increment in net assets. The definitional linking of these central
ideas—their articulation, if you will—is more than a formality. A
system which expresses the stated relations enables the effects of
actual or expected changes in the magnitude of one category to be
traced through to its effects on other categories. If judgments are
used in deriving some income account magnitudes, or if the records
are incomplete in any respect, deficiencies may be found by cross¬
checking with the independently discovered magnitudes of assets.
Income will not simply be the outcome of whatever judgments and
calculations are made; for it must agree with and be the equivalent of
the ascertained increment in net assets. Income will not simply be
"what income is calculated to be"; it will correspond with a real,
observable increment in the money amount of net assets.
The integral relationship between assets, equities, and income
was adumbrated in Paton's Accounting Theory (1922). It is also cen¬
tral to continuously contemporary accounting (CoCoA). But some
have found its implications in the latter case difficult to accept in
relation to durable assets and work in progress. Under CoCoA,
durables would be written down periodically to their money
equivalents, nonvendible durables to zero at the outset. Work in
progress which is not vendible in its balance date state would also
have a zero value. There would be corresponding charges, in some
cases heavy charges, in the income account or against some other
residual equity account. Critics of these consequences of the rules
of CoCoA contend that this is "hard doctrine" yielding "unrealis¬
tic" figures. But is it?
Consider again the hypothetical figures in the previous section on
present value calculations. An asset was to be acquired for $1,000,
the whole of which, under traditional present value calculations, is
implicitly deemed to be sunk, gone, no longer accessible. Whether
or not the asset, when acquired, turns out to be vendible or non¬
vendible, the common accounting treatment of it would be to
charge off a fraction of the amount, based on judgments of service
life and depreciation pattern. There is an obvious inconsistency in
 The Hard Core of Accounting 87

the modes of calculation. Certainly a present value calculation is

exploratory, and an account of events after their occurrence is, in
principle, past factual. But both concern the same set of events.
There can be no reason for choosing according to one financial
criterion (cash flows) and calculating the consequences of the choice
according to another (cash flows less periodical valuation adjust¬
ments); for the object of choosing is to procure a desired financial
consequence, and the function of accounting is to report on an
achieved financial consequence. Retrospective diagnosis is futile
unless "consequence" has reference to the same thing that it refers
to in exploratory calculations. Those who endorse the traditional
form of present value calculation, therefore, cannot legitimately or
consistently object to the rapid writing down of some durables
which CoCoA entails.
Nor can they object to the "dropping out" of a valuation of non¬
vendible work in progress and its restoration when it acquires vend¬
ible status, as under CoCoA. For that is exactly what is done in the
traditional present value calculation. The cash outlay is treated as
instantaneous dropout; the cash inflow expected from sale of the
asset appears when the asset is expected to be liquidated. Subject
only to the difference that CoCoA deals with money equivalents, its
treatment of work in progress follows the same pattern as the tradi¬
tional present value calculation.
But there is more substantial ground for integral calculation than
mere consistency of information-processing. Income is expected to
be a genuine increment in assets which is divisible, at least by those
who expect dividends. Income before tax is expected to represent a
genuine increment against which taxes may be levied. Income is
taken by parties to wage and price negotiations as a genuine incre¬
ment in assets, indicative of what may be divided up differently in
the future from what was done in the past. Retained income is con¬
sidered as an increment to the borrowing base of a firm and as an
increment providing for growth or the redeployment of assets. The
matter is put in a nutshell in the letter of transmittal by the Chairman
of the Securities and Exchange Commission of the staff report on the
financial collapse of the Penn Central Company (1972):

It is essential that the end result of applying accounting principles be a

realistic reflection of the true situation of the company on which a report
is prepared. Here, there was no adequate presentation of the fundamen¬
tal reality that reported income was not of a character to make a contribu¬
tion to the pressing debt maturities or liquidity needs of Penn Central,
nor was it of the sort that might reasonably be expected to be evidence of
88 R. J. Chambers

continuing earning power. Income is clearly expected to be "real/' repre¬

sented by money's worth.

The only way of being assured that this is its nature is by making
its calculation integral with a balance sheet representing assets and
equities in terms of money and money's worth.

Towards Integral Calculation

Our object has been to indicate the specific purposes to which

specific financial calculations and magnitudes are relevant, and to
suggest that the whole range of past, present, and future calcula¬
tions should follow a common systematic pattern. The integrating
element in this pattern is the use of dated money equivalents (in¬
cluding money amounts). The dominant importance of money
equivalents lies in their pertinence to the processes of exchange
which are the means of securing the survival, adaptation, and
growth of firms.
In the general case, the products of past factual, contemporary
factual, and exploratory calculations converge on the occasion of
evaluating and choosing; and anticipatory calculations represent the
expected consequences of choices made. If these temporally sequen¬
tial calculations embody a homogeneous class of appropriately dated
magnitudes, analysis of the past and preparation for the future may
be reasonably and coherently done. But only confusion and ineffi¬
ciency can be expected if any temporally identified calculation or
statement embraces components of a different temporal class. For if
users of the statements are aware of anachronisms, besides trying to
absorb what the statements say, they must make their own mental
reservations and adjustments as their deliberations proceed—a pro¬
digious task considering the variety of substantive and temporally
different quantities related to judgment. Or if users are ignorant of
the anachronisms or logical incompatibilities of the statements, they
will simply make inapt appraisals and judgments.
Statements of the past say nothing about the future. Statements of
expectation say nothing about the past. In particular the use of pres¬
ent values (even of the "inclusive" variety) in any dated statement of
financial position is inapt, for it represents a multiperiod expectation
reduced to a point valuation. Though it is one means of discriminat¬
ing between possible courses of action, in no way does it represent a
factual element of a dated financial position. The use of replacement
prices in any past or contemporary statement of financial position is
also inapt. They may be elements in any exploratory or anticipatory
 The Hard Core of Accounting 89

calculation, but that is their sole use. The long-standing and growing
tolerance for these magnitudes in statements of financial position is
symptomatic of a confused and anachronistic view of the functions of
different temporal classes of information.
Ideally accounting would deal systematically and consistently
with stocks and flows of money and money's worth—in representa¬
tions of financial states and results, and in all exploratory calcula¬
tions and budgetary statements. To calculate asset values, financial
positions, incomes, discounted cash flow valuations, and other fi¬
nancial magnitudes according to different rules is to esteem the
chaotic more highly than the systematic, the divergent more highly
than the convergent, the vagrant more highly than the disciplined,
the unintegrated or disintegrated more highly than the articulated—
in thought and in action.
There is a hard core of accounting. The money equivalents of assets
and equities are material in all judgments of financial consequences
and prospects. No system other than accounting deals with them.
But if the idea of a hard core is rejected, and every magnitude in
financial statements or calculations is determined uniquely accord¬
ing to its own item-specific rule, accounting will not be the monitor
of status and progress in the aggregate that we expect it to be.
The study and teaching of accounting processes, present value
calculations, and budgeting as separate techniques is perhaps in¬
evitable, if expertise and expert knowledge are to be developed. But
unless they are also seen as integrated, necessarily related elements
in the whole continuous process of administration and decision¬
making, each of them may be pressed severally into a shape and
style inconsistent with its specific function and inconsistent with
the related functions of them all. The result will abort all attempts to
conduct business affairs with economy and clarity of thought.
The range of matters which have commanded Paton's attention,
and the analytical skill which he has directed to their exploration and
elucidation, evidence his devotion to a belief in the disciplinary
power of reliable knowledge in the conduct of business affairs. In
tandem has been a belief in the versatility and essentiality of the
accounting framework, suitably refined, as a means of systematizing
the indications of financial progress and prospects on which efficient
administration and informed investment depend. One could hope
that one's own exercises might appear to be in like direction, with
like respect for intellectual tidiness and practical pertinence.

January 1977
Paton on the Effects of Changing
Prices on Accounting
1916-55
Stephen A. Zeff

From 1916, when he first contributed to the accounting literature,

to 1955, when he published the last major textbook of which he was
sole author, William A. Paton wrote frequently and with feeling on
the vexing subject of the impact on accounting of changing prices.
His writings have been widely quoted by both academics and prac¬
titioners, and one may confidently conclude that Paton was among
the most influential American writers on accounting in this period.
Unfortunately the accounting literature in the English language
contains very few historical studies showing how the thought of
major contributors has evolved. For this reason, it may not be gen¬
erally noticed when important normative writers alter their theo¬
retical constructs or, in the light of particular economic and political
conditions, introduce variables which affect their conclusions and
policy recommendations. That Paton has done so in his writings on
the impact of changing prices on accounting is my belief, and in
this essay I propose to examine his contributions over a period of
four decades in order to test this hypothesis.
More than most accounting writers, Paton has been acutely sensi¬
tive to shifting currents in the socioeconomic context in which ac¬
counting is performed. Over the period of this study, Paton seemed
to accommodate his vision of accounting to his changing percep¬
tions of the economic, legal, and political forces in the accounting
environment.
It is no surprise that in the first significant period of his writings.

Note: I am grateful to Harold Bierman, Jr., William W. Cooper, Maurice Moonitz,

William A. Paton, Jr., Edward Stamp, Herbert F. Taggart, and Roman L. Weil for their
useful comments on an earlier draft, but the responsibility for errors is solely mine.

91
92 Stephen A. Zeff

1917-18, Paton emerged as an idealist. Accounting writers are in¬

clined to cite his 1918 textbook. Principles of Accounting, which he
wrote with Russell A. Stevenson, as his most uncompromising chal¬
lenge to accounting orthodoxy. And it was. But beginning in the
1920s, and continuing to the end of the 1930s, Paton gradually re¬
treated from his enthusiastic and optimistic advocacy of 1917-18, as
preoccupations over the post-World War I inflation, companies'
questionable revaluation practices during the 1920s and early 1930s,
the economic stagnation of the Depression, and the growing influ¬
ence of the law in business operation collectively persuaded Paton
to become more circumspect and constrained in his avowal of ac¬
counting reform. By the early 1940s, his penchant for innovation
again became evident, and he began to employ his effective debat¬
ing style to gain support for his measures among the accounting
and legal authorities to whom deviation from precedent and con¬
vention was anathema. Following World War II, the accelerating
pace of inflation, coupled with the persistent attacks on business
and the free enterprise system, rekindled Paton's fire and enthusi¬
asm of 1917-18, this time to persuade accountants, the Securities
and Exchange Commission, Congress, and federal wage tribunals
that corporate profits were being vastly overstated.
Paton was profoundly influenced by Sweeney's "stabilized ac¬
counting," and he frequently cited Sweeney's opus Stabilized Ac¬
counting in articles and speeches.1 While he never preferred "stabil¬
ized accounting" to the booking of replacement costs, he regularly
recommended that reckonings on a "common-dollar" basis be pro¬
vided in supplementary disclosures. His enthusiasm for including
unrealized appreciation in net income faded entirely in the 1920s
and 1930s, and during these same two decades he showed growing
reluctance even to reflect replacement costs formally in the accounts
and financial statements. By the early 1940s, he had devised several
techniques—the upward quasi-reorganization, the "compromise
procedure" for supplementing plant costs and depreciation on a
replacement cost basis without altering reported income, and the
revaluation of assets upon the reacquisition of a corporation's own
shares at a cost different from their book value—by which current
costs or values might gain admission to the accounts and financial
statements. Following the war, his advocacy of these and other
reforms (including "common-dollar" accounting) assumed an inten¬
sity not before seen in his writings.

1. Henry W. Sweeney, Stabilized Accounting (New York: Harper & Bros., 1936).
 Paton on the Effects of Changing Prices 93

This historical essay is organized on chronological lines, and at

various stages the evolution of Paton's thought is explicitly exam¬
ined. It is not my objective to argue with Paton, but instead to
endeavor to understand and explain the development of his writ¬
ings on this controversial subject. A more ambitious study would
have included extensive references to the contemporaneous writ¬
ings of other authors, both in the United States and in other coun¬
tries. A complete examination of Paton's contribution to the litera¬
ture would require the development of these links with the works
of other writers, but the reader will appreciate that even my more
limited undertaking is a large assignment. It is to be hoped that
future studies will supply the missing dimension and that we might
finally have a definitive study of Paton's impact on the literature of
changing prices and accounting.

Principles of Accounting (1916)

The years 1916-18 were, following an uncertain start, the years of

Unbridled Advocacy, since they produced Paton's most enthusiastic
avowal of current-cost accounting. Idealism governed. Paton was a
graduate student in economics at the University of Michigan when
he and Russell A. Stevenson collaborated on a textbook entitled
Principles of Accounting. The book appeared in three versions, pub¬
lished in 1916, 1917, and 1918. The 1917 and 1918 versions are not
designated as second and third editions, although the titles are
identical. Of the three versions, the one published in 1918 is by far
the best known—indeed, few citations to its predecessors may be
found in the literature. The successive versions grew steadily from
222 pages to 373 and finally 685 pages, reflecting, among other
things, the authors' expansiveness and increasing conviction of the
rightness of their controversial conclusions.
The 1916 version of Principles of Accounting, containing Paton's
first published utterances on accounting measurement, bespeaks a
degree of irresolution, as if the authors had not fully grasped their
conception of accounting. While they endorse the principle of book¬
ing appreciation in the accounts, they hedge on its specific applica¬
tion. Paton and Stevenson emphasize that "accounting is concerned
with the value representation of things," not with physical facts.2

2. William A. Paton and Russell A. Stevenson, Principles of Accounting (Ann Ar¬

bor: [no publisher], 1916), pp. 14, 101.
94 Stephen A. Zeff

They add,
The accounts should always show the present value of the property
being used by an enterprise in producing its product, if accounting
statistics are to furnish the entrepreneur with the information which
shall enable him to make rational use of the economic resources at his
disposal.3

The authors' underlying philosophy is taken from Wesley C. Mitch¬

ell's Business Cycles:4
Prices render possible the rational direction of economic activity by ac¬
counting, for accounting is based upon the principle of representing all
the heterogeneous commodities, services, and rights with which a busi¬
ness enterprise is concerned in terms of money price.

Their general principle is stated as follows: "The accounts, if they

are to be scientifically accurate, should record all [value] changes,
and immediately. That is, the accounts should be as sensitive as
possible to all price and value changes."5 It is nowhere stated, how¬
ever, that revenues, or even net income, might in principle be cred¬
ited for appreciation. Paton and Stevenson leave the inference that
credits in respect of appreciation should be made to Surplus (a term
which, in those days, implied an amalgam composed of the excess
of total stockholders' equity over legal capital). When land apprecia¬
tion is discussed, the authors worry over "a tendency to credit these
increases to current revenue in lean years, which is illegitimate."
"Increases in land value," they conclude, "should not be used [to]
inflate current income, but should be credited to accumulated sur¬
plus." An equivocal qualification is added: "If, however, increases
in land value are conservative estimates, and are not used to juggle
the current Income sheet, there is no valid reason for keeping such
items out of the accounts."6 In the best light, the authors' discus¬
sion of the impact on net income of recorded appreciation is
sketchy and incomplete.
To what extent should appreciation be recorded at all—as a prac¬
tical matter? For merchandise, raw materials, etc., which move
quickly through enterprise, it is not "expedient" to keep a continu-

3. Ibid., p. 101. Although the authors do not define "present value," it is under¬
stood to mean any soundly based contemporary price or valuation.
4. Wesley C. Mitchell, Business Cycles, as quoted in Paton and Stevenson, Princi¬
ples of Accounting (1916), pp. 31-32.
5. Paton and Stevenson, Principles of Accounting (1916), p. 103.
6. Ibid., p. 105. Out of the accounts, or out of the Income sheet? The passage is
unclear.
 Paton on the Effects of Changing Prices 95

ing accounting record of changing prices. Only when "market

prices fall or rise considerably or a long interval is involved,"
should market prices be entered in the accounts. A lower-of-cost-
or-market solution is rejected as "unreasonable." For securities,
"the reasonable course is to use cost prices unless market prices are
widely divergent and for a long period, in which case securities
should be valued at the market."
The authors' preoccupation with railroad property (numerous ex¬
amples are chosen from this industry) evidently prompts a judg¬
ment that appreciation on land is the principal concern. Permanent
changes in the value of buildings and machinery are asserted to be
decreases more often than not, and any actual appreciation in such
properties is said to be ordinarily offset by depreciation. Only land
is left, and even here the authors are troubled by the booking of
"unwarranted estimates."
The authors' 1916 position, therefore, is to recommend that the
accounts continually reflect price and value changes; as a practical
matter, however, valuations of land are open to suspicion, any ap¬
preciation in buildings and land is "normally more than offset by
depreciation," and short-term assets should be kept at cost unless
the price changes are pronounced over a long period.

"The Theory of Accounts" (1917)

By 1917 Paton, writing alone, achieves a clearer integration of his

theory and policy recommendations. In his doctoral dissertation,
"The Theory of Accounts," Paton unqualifiedly endorses the book¬
ing of appreciation as part of "net revenue" (a term he often used
between 1917 and 1924 in place of net income).7 He continues to
espouse the accounting recognition of appreciation, but with the
same exceptions for merchandise, raw materials, and securities, in
virtually the identical language, as in the 1916 edition of Principles
of Accounting. But the credit in respect of currently accrued appre¬
ciation is to be made to "net revenue."8 He adds that "if an item of
appreciation covers several accounting periods before it is recog¬
nized in the accounts [it] should be credited, not to net revenue,

7. William A. Paton, "The Theory of Accounts," unpublished doctoral disserta¬

tion, University of Michigan, 1917. The preface is dated May 1, 1917, and the manu¬
script extends to 144 pages, double-spaced, including bibliography.
8. Ibid., pp. 119-20, 126-27.
96 Stephen A. Zeff

but to surplus.” The argument that a "net revenue” which includes

appreciation might obscure the results of "actual operation” is met
by the recommendation of a two-tier income statement, which
would contain the calculation of "net revenue from operation, so-
called, and total net revenue as well.”9 Paton shrewdly observes
that conventional net income already includes a speculative price
change, as, for example, when products purchased in a falling mar¬
ket are sold in a rising market. But his proposal would include
unrealized effects as well.
In the dissertation, Paton repeats the idealistic conceptual state¬
ments about appreciation from the 1916 version of Principles, but he
no longer harbors doubts about the practical implementation. In
characteristic Paton debating style, he confidently dismisses the
standard objections to booking appreciation and including appre¬
ciation in net income. As to the argument that appreciation is too
subjective an estimate to appear in the accounts, Paton glibly re¬
plies, "As a matter of fact the appreciation of fixed property can be
more accurately estimated than the depreciation.” To those (such as
Paton and Stevenson, in the 1916 version of Principles) who might
be concerned about possible manipulation by management of ap¬
preciation estimates, the Paton of 1917 responds, "It is no more
likely that illegitimate use will be made of appreciation than of
depreciation."10 In dealing with "that long-exploded notion that
profits must be available in liquid assets in order to be considered
as profits," Paton contends that stockholders can sell some shares
and thus realize on the enhanced value of their equity interests
whether "net revenue” is available in liquid form or not. Moreover,
a dividend could be paid in respect of appreciation on fixed assets
by borrowing "on short-term notes or otherwise." He concludes
with the firm opinion that "all decreases in the values of assets
constitute expense, and all increases represent revenue.” "The net
revenue figure for the accounting period," he adds, "is not accu¬
rately discovered until all these changes have been taken into con¬
sideration."11 A stronger endorsement of current-value accounting
could hardly be imagined.
The object of accounting is succinctly expressed: "From the stand¬
point of either the private enterprise or of the industrial community
the accounts should answer the question: what economic resources

9. Ibid., pp. 126-27.

10. Ibid., p. 125.
11. Ibid., pp. 128-29.
 Paton on the Effects of Changing Prices 97

are being devoted to this or that particular end?"12 Accounting is

seen as an essential element in the free enterprise system, so as
to make effective the direction of economic activity by the system
of market prices." Paton suggests that the severity of business
cycles may be diminished by an accounting that reflects current
price trends.13 Perhaps the factor that led Paton to alter his posi¬
tion on the inclusion of appreciation in "net revenue" was a con¬
cern for the temporal impact on creditors and stockholders of "net
revenue" measures that ignore appreciation. He notes that "the
personnel of the owners is usually changing from day to day due
to the ease with which securities can be transferred from one per¬
son to another."

Thus [he continues] if an error in stating net revenue is made in one

period it is probable that the rights of some of the individuals whose
equities have been misstated cannot be restored by the correction of the
error in a later period, because these rights are then in other hands. . . .
The income bondholder, for example, has a right to income only when
net revenue exists. ... In the modern enterprise the constant shifting of
the investors and the variety of property-rights as regards the aspects of
ownership render imperative the accurate determination of [net
revenue].14

The exigencies of a fluid market for equity interests, therefore,

argue for current valuation in the accounts notwithstanding the
absence of distributable funds corresponding to increments in net
income. According to Paton of 1917, the prompt reflection of appre¬
ciation in the prices of equity shares was of paramount importance.
In the dissertation, Paton's meaning of the term "appreciation"
is not always clear. At one point, it is described as "present
value." A footnote explains that present value "may be based
upon liquidating value, cost of reproduction, or going value to the
specific enterprise. The cost of reproduction basis is the most sig¬
nificant—certainly in competitive enterprise."15 A few pages later
Paton also seems to include in appreciation the changes in the
general level of prices, although this suggestion may have been
developed mainly for the benefit of the Interstate Commerce Com¬
mission, whose property accounting rules were frequently the ob¬
ject of Patonian disdain.

12. Ibid., p. 118.

13. Ibid., pp. 6-8.
14. Ibid., pp. 14-16.
15. Ibid., p. 117.
98 Stephen A. Zeff

Principles of Accounting (1917)

In their second version of Principles of Accounting, Paton and Ste¬

venson include a seven-page section on "Appreciation and Depre¬
ciation," which is taken almost verbatim from Paton's dissertation.16
Appreciation is to be credited to "net revenue," and all of the stan¬
dard arguments (save one) against the booking of appreciation and
the inclusion of appreciation in net income are rejected with the
same aplomb as in the dissertation.17 Two arguments in the disserta¬
tion, however, are omitted from the book. Nowhere in the book is
Paton's argument against the view "that profits must be available in
liquid assets in order to be considered as profits." His suggestions
that short-term borrowings can be used to pay dividends and that
stockholders can realize on the appreciation by selling off some of
their stockholdings are not to be found in the book.
Also omitted from Principles is Paton's closing judgment in the
dissertation that "all decreases in the values of assets constitute
expense, and all increases represent revenue."18 It is possible that
these omitted items were too controversial for Stevenson's taste. But
in other respects, the 1917 edition of Principles is entirely faithful to
the contents of Paton's dissertation.

Principles of Accounting (1918)

In the third version of Principles, the substance of the authors'

position on appreciation is unchanged from 1917, although their
discussion is more expansive and confidently expressed than be¬
fore.19 That Paton was evidently the principal author of the well-
known Chapter XX, "The Basis for Revaluation," is confirmed by
his article published in the same year, "The Significance and Treat¬
ment of Appreciation in the Accounts," which closely corresponds
to the chapter in Principles.20

16. William A. Paton and Russell A. Stevenson, Principles of Accounting (Ann

Arbor: George Wahr, 1917).
17. Ibid., pp. 213-19.
18. Paton, “Theory of Accounts," pp. 128-29.
19. William Andrew Paton and Russell Alger Stevenson, Principles of Accounting
(New York: Macmillan Co., 1918).
20. William A. Paton, "The Significance and Treatment of Appreciation in the
Accounts, in G.H. Coons, ed., Twentieth Annual Report of the Michigan Academy of
Science, 1918, pp. 35-49. Reprinted in Herbert F. Taggart, ed., Paton on Accounting
(Ann Arbor: Bureau of Business Research, University of Michigan, 1964), pp. 21-35;
and in Stephen A. Zeff, ed.. Asset Appreciation, Business Income and Price-Level Ac¬
counting: 1918-1935 (New York: Arno Press, 1976), second article.
 Paton on the Effects of Changing Prices 99

For the first time, Paton and Stevenson distinguish clearly be¬
tween the booking of appreciation based on changes in replacement
cost and the "illegitimate practice of forecasting profits in the ac¬
counts."21 They write:

To use selling prices in taking inventories—in other words to capitalize

the services of the firm before those services are performed—is to antici¬
pate profits. To recognize changing capital costs and equity changes due
to the appreciation of working or fixed assets is an entirely different
thing.22

The authors argue that an efficient management must be cogni¬

zant of replacement cost—that is, "the present significances of all
assets as nearly as these facts may be ascertained"—in order to
achieve "a wise utilization of available resources." But is it neces¬
sary that these figures appear in the actual accounts? Yes, reply
Paton and Stevenson, "since it is generally admitted that it is an
important function of the financial accounts to furnish information
which will assist the management in making rational decisions re¬
garding the employment of the investor's capital." In 1940, Paton
retreated from this view, contending that management could be
supplied with this information without disturbing the accounts.
But in 1917, Paton and Stevenson claimed that "it is the function of
the accounts to show economic facts."23
The authors cite Middleditch's path-breaking 1918 article when
discussing the possibility of explicitly showing "changes in the sig¬
nificance of the money unit in the accounts."24 They believe that
such changes should not be displayed in the accounts, supporting
their conclusion as follows:

It has been argued thus far that costs of replacement should form the
basis for the revaluation of assets for accounting purposes; and actual
sacrifice and cost of replacement are not likely to coincide. The change in
the value of the dollar reflects general price changes. But it is not values
in general but specific values which the accounts should show.

In later years, Paton continued to avow that specific price changes,

not the effects of general price trends, should be used if any modifi¬
cations were to be made in the accounts or in the body of the
financial statements.

21. Paton and Stevenson, Principles of Accounting (1918), p. 242.

22. Ibid., p. 464.
23. Ibid., pp. 456-57, 459 n.
24. Ibid., p. 461. Livingston Middleditch, Jr., "Should Accounts Reflect the Chang¬
ing Value of the Dollar?" Journal of Accountancy (Feb. 1918), pp. 114-20.
100 Stephen A. Zeff

Most of the objections raised in the 1917 dissertation against the

booking of appreciation and its inclusion in net income are again
unqualifiedly rejected in this version of Principles. Even the argu¬
ment that “profits must be available in liquid assets in order to be
considered as profits/' on which the 1917 version of Principles was
silent, is rehearsed here, and the authors dismiss it with the same
disdain displayed by Paton in his dissertation.25
Paton and Stevenson mention a new objection to the recording of
appreciation: “that such a practice savors of non-conservatism. . . .
A clear distinction should be made between conservatism and down¬
right concealment." They argue vigorously against this objection:

Ignoring the appreciation of unsold assets which results in an under¬

statement of assets and a corresponding misstatement of equities—is
simply another method of building up secret reserves; and it is essen¬
tially as misleading a practice as the charging of capital outlays to
expense. . . .
To insist that inventories in certain cases must be taken at a figure far
below the actual values in order to prevent a general overstatement of
assets is from the accountant an admission of incompetence.

Paton and Stevenson were willing to devise modes of presenta¬

tion in order to meet part-way the defenders of historical cost. Ap¬
preciation “need not obscure cost figures" and “might [even] be
kept out of the operating accounts." Supplementary statements
might be used to maintain the record of appreciation. But the au¬
thors persisted in the belief that “accounts and statements which
take into account all value changes would surely seem to be of more
practical use to all parties concerned than records prepared on any
other basis."26
Paton and Stevenson continue to defend the practice of crediting
currently accrued appreciation to net income, although they confess
at one point that "in general it is desirable to use the Surplus
account to reflect all speculative changes in the equities."27
Finally worth noting: the recommendation in the 1916 and 1917
versions of Principles that appreciation not be recorded on most
short-term assets, except when significant and over long periods,
appears not to be repeated in the 1918 edition.
This third version of Principles was, as Hatfield later suggested, a
bold argument for the "consistent treatment of fluctuations in

25. Paton and Stevenson, Principles of Accounting (1918), pp. 243, 467.
26. Ibid., pp. 467, 469.
27. Ibid., pp. 108, 238-42.
 Paton on the Effects of Changing Prices 101

value. ~8 In all of Paton's writings, it was his most inspired defense

of current-cost accounting.

"Depreciation, Appreciation and Productive Capacity" (1920)

In a 1920 article entitled "Depreciation, Appreciation and Produc¬

tive Capacity," Paton abruptly revises his stance on current-cost
accounting.29 No longer does he unreservedly favor crediting ap¬
preciation to net income. Indeed, his entire position on the booking
of appreciation is less than clear, for he fails to reply to criticisms
which he had calmly brushed aside in 1918. The tenor of this article
is in stark contrast to his earlier writings. While his 1918 book with
Stevenson brims with self-confidence and idealism, the 1920 article
is tainted with doubt and pragmatic skepticism.
The excuse for the article is a reply to Bauer and Rastall,30 chiefly
to argue that replacement-cost depreciation should not be charged
to operations unless the gross appreciation in the depreciable assets
is also accorded accounting recognition. Paton's evident retreat
from the view that appreciation credits should be made to net in¬
come seems to be explained, at least in part, by the persistent infla¬
tion of 1914-20, during which the consumer price index doubled.
Paton writes.

Is the proprietary credit, which should be made when a fixed asset is

charged with the increment necessary to bring the book value up to the
effective current cost, in any sense an index of income? So far as its
expressing improved economic condition is concerned this depends . . .
upon whether the price change in any particular instance involved is
more or less acute than the general movement. In the average case it
would no doubt be true that such appreciation would represent pretty
largely not income but the application of the new measuring unit to the
proprietary accounts.31

His discussion of the general price-level implications for accounting

is more fully developed than in 1918. He again refers to Middleditch
and cites a recent article by Scovill, and while he concludes that "it

28. Henry Rand Hatfield, "Book Reviews," journal of Accountancy (Nov. 1925), p.
390.
29. W.A. Paton, "Depreciation, Appreciation and Productive Capacity," Journal of
Accountancy (July 1920), pp. 1-11.
30. John Bauer, "Renewal Costs and Business Profits in Relation to Rising Prices,"
Journal of Accountancy (Dec. 1919), pp. 413-19; Ernest A. Rastall, "Depreciation Re¬
serves and Rising Prices," Journal of Accountancy (Feb. 1920), pp. 123-26.
31. Paton, "Depreciation," p. 10.
102 Stephen A. Zeff

is perhaps not unreasonable to argue that the accountant should

prepare supplementary statements at the end of each period de¬
signed to show—by making proper allowances for the change in the
value of money—the true comparative economic status of the enter¬
prise/' in the end he reaffirms his conclusion of 1918 that account¬
ing should have regard to changes in specific prices, not general
prices.32
But the principal point of interest in this article is Paton's less
ardent support, expressed in the final two pages, of the accounting
recognition of appreciation in any form. When Paton is sure of his
ground, he does not fail to reply energetically to opposing argu¬
ments. Here he leaves them unanswered. In a remarkable modifica¬
tion of his 1918 stand, perhaps induced by the cumulative effect of
an apparently unrelenting inflation coupled with the postwar eco¬
nomic uncertainty of 1919-20, Paton gives credence to arguments
which he had enthusiastically rejected two years earlier:

At any rate [an appreciation] credit in a practical case would not mea¬
sure an amount which could conveniently or safely be turned over to the
stockholders. . . .

In view ... of the conjectural character of asset values at best and the
consequent importance of conservatism [sic], the difficulties in the way
of determining effective replacement costs in the case of complex assets,
the constant fluctuation of such costs and the fact that having once made
an investment the management is often thereby committed to a policy for
a considerable period regardless of the movement of prices, probably
most accountants would feel that original cost is the best basis upon
which to value fixed assets.33

Does Paton include himself among "most accountants"? Is he

abandoning entirely the accounting recognition of appreciation, or
is he limiting this reply to Bauer and Rastall to the points raised in
their articles? However one might interpret this article, something
has taken the wind out of Paton's sails.
A comparison of the 1918 and 1920 positions vividly demonstrates
how much Paton was affected by the swirl of economic events. For
all of his life, he has been a firm believer in the market system; it
was, in effect, the text from which he read.
Evidently Paton's article had an impact. More than forty years
later, Sweeney wrote,

32. Ibid., p. 4.
33. Ibid., pp. 10-11.
 Paton on the Effects of Changing Prices 103

That excellent article, far in advance of its time, entitled "Depreciation,

Appreciation and Productive Capacity" . . . was the first exposition I had
seen of how accounting figures, representing depreciation, could be re¬
vised to show the changing purchasing power of money. It stimulated
and encouraged me to try to develop all the great possibilities that this
new subject in the United States seemed to offer.34

Accounting Theory (1922)

Paton's celebrated Accounting Theory—With Special Reference to

the Corporate Enterprise, published in 1922, reaffirms his position on
fixed assets as set out in the 1920 article and discloses his qualified
approval of showing appreciation on current assets.35 Accounting
Theory has erroneously been said to be the publication of Paton's
doctoral dissertation.36 In fact. Accounting Theory is four times the
length of the dissertation, and while Chapters I through VII and X
through XI of Accounting Theory largely correspond to material in
the dissertation, the remaining chapters (including four which, Pa¬
ton reported, were adapted in part from previously published arti¬
cles) either did not appear in the dissertation or were lightly treated
there.
Of particular importance is Paton's major change in posture on
appreciation between the 1917 dissertation and Accounting Theory,
published five years later. Where he concluded in the dissertation
that it would be inexpedient to record appreciation in most in¬
stances on current assets, in Accounting Theory (as, apparently, in
the 1918 version of Principles of Accounting) he actually espouses the
normal booking of appreciation on certain kinds of current assets,
and he would even credit such accruals to income.37 Where, in the
dissertation, he would credit the currently accrued appreciation of

34. Henry W. Sweeney, "Forty Years After: Or Stabilized Accounting Revisited,"

in the reissue of his Stabilized Accounting (New York: Holt, Rinehart and Winston,
Inc., 1964), p. xx.
35. William Andrew Paton, Accounting Theory—With Special Reference to the Corpo¬
rate Enterprise (New York: Ronald Press, 1922). Reprinted in 1962 by A.S.P. Account¬
ing Studies Press, Ltd., Chicago, and in 1973 by Scholars Book Co., Houston.
36. This writer is among those who have erred. In the publisher's preface to the
1962 reprint of Accounting Theory, signed by this writer and three others, it is said
that Accounting Theory "was [Paton's] doctoral dissertation at the University of
Michigan." In many respects. Accounting Theory was based on the dissertation, but it
went beyond the dissertation and departed from it in significant points (chiefly in
the treatment of appreciation).
37. Paton, Accounting Theory, pp. 466-68.
104 Stephen A. Zeff

fixed assets, notably land, to "net revenue," in Accounting Theory,

following the inconclusive discussion in the 1920'article, Paton re¬
views the pros and cons, and appears to conclude that the apprecia¬
tion on fixed assets is best not even entered in the accounts. If such
appreciation is recorded, however, he is exceedingly reluctant to
classify it as income. On one point he is decisive: the portion of
recorded appreciation which coincides with the general movement
of prices should be credited to capital.38
In Accounting Theory, not all current assets would qualify for the
booking of appreciation. "The consistent valuation of standard ma¬
terials and marketable securities on the basis of replacement cost,"
he counsels, "is a thoroughly sound procedure. . . . Goods in pro¬
cess and finished stock furnish a more dubious case." Moreover:
"With respect to plant and equipment and other classes of fixed
assets the propriety of cost-of-replacement valuations is still more
questionable ..." As in the dissertation, where he recommended
that the appreciation credit on fixed assets be classified as "net
revenue" from other than operations, in Accounting Theory he de¬
nominates the appreciation credit on eligible current assets as
"non-operating income." Paton's mild support for crediting income
with the currently accrued appreciation of current assets evaporated
by the 1930s, and he has never again argued for the inclusion in
income of appreciation accruals. In Accounting Theory he continues
to argue, as did Paton and Stevenson in 1918, that recorded appre¬
ciation should not be based on the selling prices of merchandise,
because its inclusion in income "may mean the [improper] recogni¬
tion of a fraction of the normal income margin in advance of sale."39
Paton's principal reason for rejecting the inclusion in income of
appreciation of fixed assets may be traced to his desire, first given
expression in the 1920 article, to assist management in maintaining
its physical capital. To achieve this end, depreciation should be
based on appreciation, but the appreciation accrual itself should not
be credited to income. "To revalue assets and base depreciation
upon replacement costs," he writes, "would be an entirely futile
policy unless the appreciation credit were excluded from the income
account—at least as far as disposable income is concerned."40 Be¬
ginning with his 1920 article, therefore, Paton left the ranks of those
who would endeavor to approximate economic income in the

38. Ibid., pp. 441-42.

39. Ibid., pp. 466-68.
40. Ibid., p. 441.
 Paton on the Effects of Changing Prices 105

income statement, and became a member of what I have called the

"disposable income school.”41
Paton s discussion of the accounting effects of general price
movements is more extensive in Accounting Theory than in his pre¬
vious writings. In somewhat more than seven pages, he concludes
that accountants should be interested primarily in "specific price
and value changes,” not "general price movements”; yet he grudg¬
ingly concedes:

It would perhaps not be unreasonable to urge that the accountant should

be held responsible for the preparation of supplementary statements at
the end of each period designed to show—by making proper allowance
for the change in the value of money—the true relation between the
current statement and the one immediately preceding (or a series of
earlier statements).42

This suggestion was also contained in the 1920 article, expressed in

the same cautious language.43
He also proposes (in more positive terms) that the accountant
prepare supplementary statements which reflect individual price
movements—in the event, one supposes, that appreciation is not
deemed proper for inclusion in the formal financial statements.
Accounting Theory is characterized by indecision and contradic¬
tion, as Paton was evidently passing through a period of profound
reassessment of the theories which he had propounded so confi¬
dently in 1917-18.

Accounting (1924)

Paton's first textbook. Accounting, after the publication in 1918 of

the third version of Principles of Accounting invites comparisons
between the two.44 The differences are instantly noticeable. No¬
where in Accounting does one find the idealistic formulations about
the valuation of assets in relation to income that are evident in
Principles. In this respect. Accounting is the more pragmatic and
down-to-earth of the two volumes.
After persuasively arguing that appreciation of readily marketable
securities has a strong claim as a credit to income, and offering the
generalization that "wherever appreciation consists in a bona fide

41. See Stephen A. Zeff, “Replacement Cost: Member of the Family, Welcome
Guest, or Intruder?" Accounting Review (Oct. 1962), pp. 617-20.
42. Paton, Accounting Theory, pp. 428-29.
43. Paton, "Depreciation," p. 4.
44. W.A. Paton, Accounting (New York: Macmillan Co., 1924).
106 Stephen A. Zeff

increase in value in reasonably liquid assets there is ground for

urging that such appreciation is quite as substantial a basis for the
recognition of income as an increase in value brought about
through the sale of original assets/' Paton hastens to point out that
since the Treasury Department "does not permit the ordinary inves¬
tor in securities to treat appreciation as taxable income ... it is
probably advisable in most cases for the investor not to set up
appreciated values in his regular accounts."45 Thus enters the Law,
which plays a major role in Paton's writings in the 1930s.
The appreciation of plant assets is of "dubious significance as far
as the asset accounts are concerned, and is of still more doubtful
character as an income determinant," and only with respect to land
which is highly marketable does "genuine appreciation [have] some
significance as true income."46
Elsewhere Paton mentions several objections to crediting appre¬
ciation to income:

. . . appreciation is not in general viewed as income from the legal point

of view, . . . enhancement does not measure liquid assets which may be
disbursed as dividends, and . . . appraisal values may be used illegiti¬
mately to conceal operating deficits and other losses.47

He does not undertake to dispute these objections—which are, in

company with similar statements in his 1920 article and Accounting
Theory, difficult to associate with the Paton of 1917-18.
Paton is still willing to countenance the upward revaluation of
fixed assets and inventories so long as the credit is made to Surplus,
not income.48 But his support for the practice falls well short of
advocacy, and when objections and limitations are raised, they are
seldom rebutted.
In Accounting, Paton does not discuss the accounting effects of
changing general prices. Nor does he make an issue of the preserva¬
tion of physical capacity during periods of advancing prices, which
was a pivotal concern in his 1920 article and Accounting Theory.
Allowing for the fact that Accounting is an introductory text and
Accounting Theory is a major treatise, the tenor of Paton's remarks
on appreciation is in most respects similar in both. Perhaps the
noteworthy difference is the somewhat greater attention accorded to
legal prescriptions in Accounting than in Accounting Theory. The

45. Ibid., pp. 624-25.

46. Ibid., p. 625.
47. Ibid., p. 367.
48. Ibid., pp. 352-53, 368, 373-74.
 Paton on the Effects of Changing Prices 107

differences between Accounting and the 1917-18 Principles of Ac¬

counting are, of course, graphic.

Four Articles and the AAA Statement

in the Early to Mid-1930s

By the 1930s, Paton's enthusiasm for replacement cost accounting

had diminished still further, and he began to warm to the recom¬
mendation that supplementary disclosures should report the effect
of changing prices on the firm. The reasons were several. One, he
was disturbed by the haphazard and unsubstantiated upward reval¬
uations in the 1920s and downward revaluations in the early 1930s.
He writes in 1934,

One of the things which tends to discourage hopes of the possibility of

improving accounting via the route of revaluation of fixed assets is ob¬
servation of the way in which revaluations are handled in actual practice.
The revaluations downward in recent years, for example, have in most
cases been entirely unscientific and often arbitrary in the extreme.49

In 1932 he writes.

And it is equally apparent that the [write-down] procedures being fol¬

lowed [in the early 1930s] are in many cases just as unsystematic and
improper as were the procedures followed in many of the earlier write¬
ups.50

Also in the early 1930s, Paton began to be preoccupied with the

legal boundaries within which the formal accounting system must
operate. Paton's awareness of the growing importance of the federal
income tax law coupled with the intense interest being shown dur¬
ing the Depression years in the legal definitions of income, capital,
surplus, and dividends may have contributed to this concern.51 The
relative interests of different classes of investors in troubled corpo¬
rations appear to have been a live issue in the early 1930s. Paton
may also have been thinking of references to accounting figures in
bond contracts. He reserved some of his most picturesque writing
to describe the heavy hand of the law:

49. William A. Paton, "Aspects of Asset Valuations," Accounting Review (June

1934), p. 127.
50. W.A. Paton, "Accounting Problems of the Depression," Accounting Review
(Dec. 1932), p. 264.
51. The attention being given to such questions may be discerned from reading
issues of the Accounting Review during the early and middle 1930s.
108 Stephen A. Zeff

It must also not be forgotten that our legal institutions care nothing for
purchasing power and treat all dollars, regardless of changing price lev¬
els, as equivalents. And the accountant, called upon 'to report and trace
the effects of a maze of specific legal relationships and contracts, cannot
ignore the legal structure, and hence is circumscribed and harassed in his
efforts to become a statistical economist.52

In a 1931 article, Paton reveals the philosophical underpinning for

the use of current costs in financial statements. He argues that

to be an effective cost a cost factor must be significant in the competitive

price-making process, or, to put the matter more definitely from the
standpoint of accounting valuation, to constitute a legitimate asset in the
balance sheet an element of cost incurred must be of such a character that
the presence of the factor or condition represented by such cost makes it
possible for the enterprise to avoid incurring another similar cost which
would otherwise be required in the production of revenue.53

He concludes that by "effective cost" the economist means "the po¬

tential cost of replacement."54 (In this period, Paton's references to
replacement cost were to eventual replacement, not to simulated im¬
mediate replacement.) And "It is only in the case of a continuous
flow of standard units that replacement cost can be held to have any
marked influence on value."55 The same article is one of the rela¬
tively few places in Paton's writings where he discusses in some
depth a combination of replacement costs and general price-level
changes, the change in the former in relation to the latter yielding the
"real" (or "true") enhancement or decrement in economic status.56
But experiences during the Depression convinced Paton that this
philosophy did not deal adequately with the times. By the end of
1932, he writes,

In fact in a time [of overcapacity, replacement cost] has been temporarily

eclipsed as a price-determining factor, for the individual concern or for
business in general. . . .

In the first place cost of replacement means very little if anything in

the case of obsolete or semi-obsolete property which is not being repro¬
duced in the form of a constant flow of new plants and new machines.
Second, the complex plant property of modern times is so specialized

52. Paton, “Aspects," p. 126.

53. W.A. Paton, “Economic Theory in Relation to Accounting Valuations," Ac¬
counting Review (June 1931), p. 91. Compare the later article, David Green, Jr., “Mo¬
ral to the Direct Costing Controversy?" Journal of Business (July 1960), pp. 218-26.
54. Paton, “Economic Theory," p. 94.
55. Ibid., p. 95.
56. Ibid.
 Paton on the Effects of Changing Prices 109

that much of it, even in the case of the newest and best types, will never
be reproduced in precisely the same form.57

In 1934, he seemed to question the principle still further:

Nevertheless it must be remembered, first that the operating expenses of

the particular concern in which the accountant is interested will not as a
rule be the focal point in price determination, and, second, that the
influence of the depreciation charge upon price is obscure and not sus¬
ceptible of statistical isolation.58

Only in the long run, or in periods of capacity operation, appar¬

ently, might the depreciation charge be price-determining, and Pa¬
ton did not believe that the prospect of such a remote future effect
was worth recording in the accounts.59
Paton was deeply troubled throughout the period. On the one
hand, he wanted to argue on principle that historical costs consti¬
tute obsolete information, but on the other hand he felt constrained
by institutional and economic considerations. It must have been
painful for him to admit in 1934 that "barring exceptional circum¬
stances the cost basis of valuation, with all its limitations, is not
seriously defective for accounting purposes."60 But then he adds his
clarion call of the 1930s:

These days of monetary tinkering, however, may be the "exceptional

circumstances" which justify our giving more attention to devices and
methods designed to supplement, in a useful way, our present types of
records.61

He favored supplementary disclosures of the effects of changes in

both replacement costs and general price levels,62 and in a second
1934 article, he criticized accountants for being so "sluggish" in
following the lead of Sweeney and others toward the preparation
of supplementary analyses of the impact of inflation on enterprise
affairs.63
By 1936, therefore, when he joined such arch historical costers as
Eric L. Kohler, A.C. Littleton, and Howard C. Greer in the draft¬
ing of an accounting principles statement on behalf of the newly

57. Paton, "Depression," p. 266.

58. Paton, "Aspects," p. 126.
59. Ibid. See also p. 123.
60. Ibid., p. 128.
61. Ibid. Emphasis added.
62. Ibid.
63. William A. Paton, "Shortcomings of Present-Day Financial Statements," Jour¬
nal of Accountancy (Feb. 1934), p. 130.
110 Stephen A. Zeff

reorganized American Accounting Association, Paton could fairly

be said to be a historical coster himself in the basic financial state¬
ments. Nonetheless, the committee's assertion that "accounting is
thus not essentially a process of valuation, but the allocation of
historical costs and revenues to the current and succeeding fiscal
periods" may not have been a genuine reflection of Paton's funda¬
mental view of accounting.64

The committee's statement was a thoroughgoing paean to histori¬

cal costs, although it conceded that "interpretation [of financial
statements] is often impossible without extensive supplemental
data, including the provision of common denominators through
which comparisons may be effected."65
Elsewhere in the statement, the committee writes, "An extreme
change in the value of money might vitiate the usefulness of cost
records but there seems to be no sound reason for repeated adjust¬
ments of asset values for the ordinary changes in price levels com¬
monly experienced from one generation to another."66 This asser¬
tion is in line with Paton's separate writings on the desirability of
preserving the record of cost in the accounts. Finally, the committee
criticizes the "uncoordinated" appraisals of fixed assets in language
that could well have been Paton's.67
Thus, by the middle of the 1930s, Paton had transformed his
radical call for financial-statement reform of 1917-18 to a recom¬
mendation that corporations supplement their financial statements
with informative disclosures. The questionable booking of write¬
ups in the 1920s and write-downs in the 1930s, the special problems
wrought by the economic stagnation of the Great Depression, and
his growing concern over the legal constraints within which ac¬
counting must be performed, collectively persuaded Paton that the
formal accounting records and resulting financial statements were
better kept on a cost footing than on a replacement-cost basis.
When reporting to managers, Paton continued to prefer revalua¬
tions based on replacement cost to restatements in line with chang¬
ing general price levels—a position from which he has never

64. “A Tentative Statement of Accounting Principles Affecting Corporate Re¬

ports," Accounting Review (June 1936), p. 188. Under the title, "A Tentative State¬
ment of Accounting Principles Underlying Corporate Financial Statements," it was
widely distributed as a reprint.
65. Ibid.
66. Ibid., p. 189.
67. Ibid.
 Paton on the Effects of Changing Prices 111

wavered.68 Indeed, while he has recommended at different times

that replacement costs might be reflected in the accounts and in the
financial statements, he has never proposed that the effects of
changes in general price levels be shown anywhere but in supple¬
mentary presentations. But in the 1930s, when economic conditions
failed to substantiate the belief that replacement costs were price-
influencing, Paton declined even to recommend that replacement
costs be entered in the accounts. His conclusion during that era was
that either replacement costs or general price-level effects should be
displayed in supplementary disclosures.

Essentials of Accounting (1938)

Essentials of Accounting was Paton's first textbook after Accounting

(1924). In Essentials, Paton continues to espouse a cautious policy
toward valuations other than historical cost. In the majority of in¬
stances, the financial statements should be based on historical cost,
but Paton encourages supplementary disclosures to inform manage¬
ment of the impact of price fluctuations.69 Nonetheless, in “periods
of sharply changing prices/' a departure from cost valuation of in¬
ventory “has some merit where the period of production is suffi¬
ciently long, and the selling market is sufficiently sensitive, to make
possible modifications of policy based on observation of the changing
costs of goods on hand." Paton adds, however, without rebutting
the point, “The most common objection offered to the consistent
use of replacement cost in pricing inventory is that such valuation
will result in effect in the recognition of unrealized profit where
replacement cost is higher than actual cost."70
In Principles of Accounting (1918), Paton and Stevenson were op¬
posed to the inclusion in income of the margin between the net
realizable value and replacement cost of merchandise inventories.
In Accounting, published in 1924, Paton objected to crediting in¬
come with the spread between the replacement cost and historical
cost of plant assets, but he seemed to countenance the inclusion in
income of this spread in regard to "reasonably liquid assets,"
which excluded merchandise inventories. In 1938, he reaffirmed
this position.

68. Paton, “Aspects," p. 126.

69. W.A. Paton, Essentials of Accounting (New York: Macmillan Co., 1938), pp. 482,
813.
70. Ibid., pp. 482-83.
112 Stephen A. Zeff

By 1938, Paton was prepared to consider a formal recognition in

the accounts of the replacement cost of fixed assets. It is in Essentials
that he reveals his new tactic. After stipulating that “the cost of
replacement is truly significant [for business managements] only in
the case of standard plant units which will presumably be replaced
upon retirement substantially in kind,” Paton recites "numerous
difficulties in the way of achieving results which are worthwhile by
attempting to shift the measure of depreciation from cost to replace¬
ment cost":

Continuous revaluation is a costly process; appraisal results at the best

are estimates; current costs of a particular date are not likely to measure
the actual cost of replacement upon retirement; depreciation is only one
cost and a modification of this element will not change the operating
picture materially except where the movement of costs has been very
sharp; [and] in general replacement cost is not recognized as a basis of
depreciation for income-tax purposes or in other legal connections.71

If revaluations are nonetheless entered in the accounts, the "recorded

cost figures [should not be] obscured." The gross revaluation of plant
should be shown in a "distinct" account, the corresponding credit
should be "segregated" from the ordinary capital and proprietary
accounts, and as subsequent depreciation accruals are recorded on
the basis of the revalued assets, a transfer should be made from the
proprietorship account originally credited with the revaluation "to
the income or [earned] surplus account as a correction." Thus,
Earned Surplus, if not Net Income, would be shown as if the revalua¬
tions had not been recorded. "It is universally agreed," Paton as¬
serts, incorrectly, "that no part of the unrealized appreciation or
declination should be reflected in the income statement."72 His re¬
commended treatment of plant revaluations, including the "wash¬
out" of subsequent depreciation accruals in the income statement,
was later labeled by Paton a "compromise procedure."

71. Ibid., pp. 542-43.

72. Ibid., p. 543. In 1939, Kenneth MacNeal published his controversial Truth in
Accounting (Philadelphia: University of Pennsylvania Press) in which he advocated
the inclusion of unrealized appreciation or declination of fixed assets in the income
statement, but said these amounts should be carried to Capital Surplus, not Earned
Surplus. He would, however, transfer the unrealized price appreciations and declina¬
tions on inventory to Earned Surplus (see chap. XIV). Sweeney recommended that
each year's installment of real unrealized appreciation (the amount of unrealized ap¬
preciation which remains after allowing for changes in the general price level) be
shown in the income statement as "unrealized net income," which would be trans¬
ferred to "Unrealized Surplus," rather than the Retained Earnings-like account, "Real¬
ized Surplus." Henry W. Sweeney, Stabilized Accounting (1936), pp. 18-19, 50-52.
 Paton on the Effects of Changing Prices 113

Finally, although Essentials was an introductory textbook, Paton

devotes the last six pages to the impact of inflation on the accounts.
He returns to earlier arguments:

In attempting to deal with this problem the accountant is embarrassed

by the fact that the law in general entirely fails to acknowledge the
phenomenon of a variable unit of monetary measure. The accountant, in
other words, is coerced in some measure by the framework of contracts
and legal institutions by which the business enterprise—his immediate
field of activity—is surrounded and sustained; he must report contractual
earnings, amount subject to income tax, amount available for dividends,
capital stock, accumulated surplus, etc., in the first instance at any rate,
in accordance with impinging legal requirements. At the same time the
accountant is coming to be depended upon more and more as an inter¬
preter of the essential economic conditions of the enterprise for the pur¬
poses in particular of assisting in advising investors and framing mana¬
gerial policies. And in these connections it is plain that he cannot fulfill
his function adequately and remain blind to the limitations of the con¬
ventional accounts and reports in the face of a varying dollar.73

He discusses the impact of inflation on an enterprise's working

capital and long-term liabilities, and illustrates in numerical format
(for the first time) the conversion of a comparative balance sheet
into dollars of a common index. "What is needed in most situations
is a special report supplementing the periodic statements, [but] ex¬
cept in periods where the change in dollar value is sharp and pro¬
longed as a result of emergency conditions and accompanying
monetary tinkering the case for giving explicit attention to the
problem is not very strong."74 Here he uses, for the first time, the
expression "common-dollar basis" to describe the method em¬
ployed.75 Essentials may well have been the first American account¬
ing textbook to discuss and numerically illustrate general price-level
accounting.

"Recent and Prospective Developments in Accounting

Theory" (1940)

In April, 1940, Paton delivered the third Dickinson Lecture at the

Harvard Graduate School of Business Administration.76 Entitled
"Recent and Prospective Developments in Accounting Theory," it

73. Paton, Essentials, p. 813.

74. Ibid.
75. Ibid., p. 817.
76. Dickinson Lectures in Accounting (Cambridge, Mass.: Harvard University Press,
1943), pp. 85-131.
114 Stephen A. Zeff

contains only a few pages on the effects of changing prices on

accounting. Paton observes that the trends of specific and general
prices "are regularly in the same direction/' following which he
offers a sly dictum which he became fond of repeating in later
years.
It is this fact which affords some justification for the paradoxical state¬
ment that "the proponents of replacement cost valuations are in effect
defending adherence to the original cost basis." Interpreting cost in¬
curred in the sense of economic power committed to the enterprise, there
is a measure of truth in this observation.77

Whatever the validity of this statement, it was undeniably a clever

debating tactic.
He continues to argue that the replacement-cost approach is more
useful than the purchasing-power approach to managers and own¬
ers of enterprise. In a statement which is prophetic of the American
accounting literature of the 1960s and 1970s, Paton writes, "Busi¬
ness management is concerned in some degree with the mainte¬
nance of the integrity of investment measured in purchasing power
but is perhaps even more devoted to the protection of productive
capacity and scope of activity."78
He nonetheless persists in his opposition to any role for plant
valuations in the determination of income. Among the reasons
given, he contends that the use of replacement cost is "of marked
significance only in the case of standard, up-to-date facilities,
which will presumably be replaced substantially in kind." Return¬
ing to a theme in his "Aspects" (1932) and "Depression" (1934)
articles, Paton adds, "Neither the original cost nor the replacement
cost of facilities of an obsolescent character are potent factors in the
economic process." Another reason given, which elaborates on a
point made briefly in Essentials of Accounting two years before, is
Paton's belief that depreciation is in most cases a relatively small
fraction of total operating costs, and thus the effect of a change from
original cost to replacement cost would be of little practical conse¬
quence. Following World War II, however, Paton cited the size of
depreciation in relation to net income as justification for a change in
the basis of depreciation accounting.
In Paton's view, the most important argument against the use of
replacement costs in the determination of net income is its "con¬
flict with the legal point of view." In language reminiscent of his

77. Ibid., p. 126.

78. Ibid., pp. 126-27.
 Paton on the Effects of Changing Prices 115

"Aspects” (1934) article and Essentials, he writes that "the accoun¬

tant is inevitably coerced in considerable measure by the legal
framework, and in his search for ways of improving the reporting
of operating data for managerial purposes he should not lose sight
of the legal criteria of the general income statement."79
The legal impediments to changing the way in which net income
is determined have continued to be an influential factor in Paton's
thinking. In his Dickinson Lecture, Paton unveils his adaptation of
the quasi-reorganization to deal with undervalued fixed assets, an
accounting adjustment which he urged on numerous occasions in
the 1940s and 1950s as a means of supplanting historical cost rec¬
ords by infrequent upward revaluations of nonmonetary assets.80
That quasi-reorganizations were acceptable under certain condi¬
tions for downward revaluations persuaded Paton that, by analo¬
gous reasoning, the same accounting device could be extended to
upward revaluations. Even if upward quasi-reorganizations were
not available to a company, it might, according to Paton, book
replacement costs without obscuring recorded costs, with the pro¬
viso that the subsequent depreciation based on the appraisal incre¬
ments would appear in the income statement as information only,
without reducing the reported net income below what would have
been shown by the use of historical cost. This is the same method
that he introduced and discussed in Essentials.

An Introduction to Corporate Accounting Standards (1940)

Paton's posture on the accounting response to changing prices

was faithfully reflected in his celebrated 1940 monograph with A.C.
Littleton, An Introduction to Corporate Accounting Standards.81 The
monograph was an outgrowth of the tentative "Statement of Ac¬
counting Principles Underlying Corporate Financial Statements,"
issued by the executive committee (of which Paton was a member)
of the American Accounting Association in 1936.
At several places in the monograph, the authors draw attention to
the importance of legal considerations, thus obliging the accountant
to employ traditional historical costs in the reckoning of income.82

79. Ibid., pp. 127-29.

80. Ibid., p. 129.
81. W.A. Paton and A.C. Littleton, An Introduction to Corporate Accounting Stan¬
dards (Chicago: American Accounting Association, 1940).
82. Ibid., pp. 4, 9, 62, 123, 135, 136, and 141; but see pp. 10-11.
116 Stephen A. Zeff

Indeed, they categorically exclude appreciation on nonmonetary as¬

sets from income:

Appreciation, in general, does not reflect or measure the progress of

operating activity; appreciation is not the result of any transaction or any
act of conversion; appreciation makes available no additional liquid re¬
sources which may be used to meet obligations or make disbursements
to investors; appreciation has little or no legal standing as income.83

A monograph in which “verifiable, objective evidence" is pro¬

posed as a basic concept is not a place where one would expect to
find an argument for replacement costs in the principal financial
statements.84 Paton and Littleton acknowledge that managers
should be made aware of the impact of changing replacement costs
where such costs might affect business decisions, but they suggest
that this information can be made available to management without
altering the accounts. They enumerate several limitations of the
replacement-cost approach, all of which would have been well
known to readers of Paton's previous writings.85 The set of entries
by which historical-cost measurements of plant cost and deprecia¬
tion would be accompanied by appraisal increments or decrements,
but without altering the historical-cost net income, is discussed at
length (the same approach was proposed in Essentials of Accounting
and in the Dickinson Lecture). Paton's increasing interest in
Sweeney's "stabilized accounting" is further evident in the mono¬
graph, where extensive discussion is devoted to a means by which
recorded costs might be converted to "common dollars."86 In regard
to both replacement costs and common-dollar accounting, the au¬
thors conclude that supplementary disclosures would suffice to
adapt the formal accounts to conditions of changing prices. It would
appear that the imperative of "verifiable, objective evidence" to¬
gether with the authors' pervasive preoccupation with legal consid¬
erations serves to defeat any argument which might displace his¬
torical costs in the formal financial statements. Replacement costs
would perhaps be allowed in the financial statements only as bits of
information, so long as they do not affect the measurement of net
income.
Of interest, however, is the authors' reference (doubtless a Pato-
nian initiative) to the possibility, albeit mildly suggested, that the

83. Ibid., p. 62.

84. Ibid., pp. 18-21.
85. Ibid., pp. 131-34.
86. Ibid., pp. 139-41.
 Paton on the Effects of Changing Prices 117

difference between the incurred cost and book value of a corpora¬

tion's reacquired shares might serve as a basis for the revaluation of
the corporate assets.87 The suggestion is mentioned in the most
inoffensive terms, and the authors seem content not to carry out its
implications.

Advanced Accounting (1941)

Advanced Accounting was Paton's first textbook designed for stu¬

dents above the introductory level. It may be distinguished from
most other intermediate and advanced texts by Paton's determina¬
tion to discuss and debate virtually all of the controversial issues
of the day. Advanced Accounting was hardly a compendium of
mechanics.
For inventories, Paton peremptorily rejects valuation at net selling
prices and shows little enthusiasm even for the use of replacement
costs.88 Both of these alternatives to historical cost, he argues,
would give rise to the recognition of unrealized profit. He does not
discuss the alternative of crediting stockholders' equity instead of
income for the amount of any such appreciation, perhaps because
"for most trading concerns, it is fair to say, there is little to be
gained [in the income statement] in the way of useful reporting by a
shift from recorded cost to current buying price, particularly when
cost is estimated by the first-in, first-out method." Moreover, "re¬
placement cost as an independent valuation base is not generally
approved by the income-tax authorities, and is viewed as noncon¬
servative by many accountants, bankers, and business managers."
When he is not disposed to criticize extant practice, Paton takes
refuge in the word of other authorities.
In regard to the valuation of plant, Paton rejects valuation based
on a discounting of the prospective earning power of the assets.89
Two principal reasons are that such estimates "are none too de¬
pendable under the most favorable conditions" and that the income
streams are not uniquely linked to particular assets, but to a compo¬
site of assets. In discussing valuation at replacement cost, Paton
argues once again that "the influential costs in the economic pro¬
cess are those that are reflected in the immediate level of prices."

87. Ibid., p. 116.

88. W.A. Paton, Advanced Accounting (New York: Macmillan Co., 1941), pp. 151-
52, 160-61.
89. Ibid., pp. 323-24.
118 Stephen A. Zeff

But he cautions that the adoption of this view does not lead inexo¬
rably to "a continuous process of plant appraisal and a continuous
revision of plant accounts." It would be "inexpedient" to embark
on such a "systematic program" in the absence of a "major and
persistent change" in prices. Furthermore,

it must not be forgotten . . . that the price system is not uniformly sensi¬
tive throughout, and that for considerable periods selling prices may not
move in harmony with changing costs of production. Selling prices, more¬
over, are not in general fixed by cost to the particular concern—whatever
the basis on which such cost may be computed.

Finally (as in the 1940 monograph), he suggests that business man¬

agement can take into consideration such price movements without
necessarily "revamping the plant accounts." He concludes by strik¬
ing a familiar chord:

Particularly in view of income-tax regulations and other aspects of the

legal framework within which a business enterprise must operate, the
original cost of plant and the depreciation charges based on such cost are
data which the accountant must continue to make available. From an
accounting standpoint, that is, replacement costs are not a substitute for
recorded costs, although they may represent significant supplementary
facts.

Later in the same chapter, he claims that the failure of recorded

cost to reflect replacement cost "is not the most serious weakness of
the cost basis." Of greater consequence is that "the original dollar
cost [of nonmonetary assets] is not a dependable expression of ac¬
tual cost in the sense of economic sacrifice or committed purchasing
power." Paton argues forcefully for an accounting recognition of
changes in the value of the monetary unit and blames "the legal
framework within which business transactions are conducted" for
the lack of progress in this direction.90 An entire chapter is devoted
to "Common-Dollar Reporting," constituting (up to that time) the
most extensive discussion of general price-level accounting in any
American accounting textbook. In this 18-page chapter, he illus¬
trates the conversion of the balance sheet and income statement and
discusses the accounting treatment to be given to purchasing power
gains and losses. Like Sweeney, Paton contends that the gains and
losses on monetary items are unrealized so long as the monetary
assets are held or the monetary liabilities are unpaid.91 By 1955,

90. Ibid., p. 329.

91. Sweeney, Stabilized Accounting, pp. 20-22, 33; Paton, Advanced Accounting, pp.
739, 748.
 Paton on the Effects of Changing Prices 119

Paton modified this stance in regard to current monetary assets and

liabilities.
Paton illustrates alternative accounting treatments for booked ap¬
preciation or "declination." His preference is to credit a special
stockholders' equity account for the amount of the net appreciation
(i.e., after allowing for accumulated depreciation on the apprecia¬
tion) and then to "reserve" an amount of Surplus (Retained Earn¬
ings in today's parlance) equal to the accumulated depreciation on
the appreciation. The object of the latter transfer is to "provide" for
the understated depreciation of prior years. When depreciation is
later recorded on the appreciated plant, an amount equal to the
depreciation on appreciation is transferred to the income statement
from the special stockholders' equity account. In this manner, the
net income is undisturbed by the revaluation of the plant. At the
end of the year, an amount equal to the year's depreciation on
appreciation is transferred from the Surplus account to a surplus
reserve. The effect of these entries is to display the revaluation in
the balance sheet (while not erasing the historical costs) and to
confine the revaluation in the income statement to disclosure of the
year's depreciation on appreciation without thereby affecting net
income. The unreserved Surplus account is nonetheless reduced as
if net income had been lower by the amount of the depreciation on
appreciation. "Declinations" would be treated, in a similar manner
but with opposite effects, as appreciations. In later years, Paton
came to call this treatment "a compromise procedure," as it recog¬
nized revaluations without obscuring historical costs or affecting
the amount of the net income.92 Paton defended this compromise as
follows:

Adoption of a policy of charging appreciation to operations, with no

accompanying adjustment of profit and surplus figures, [would resultl in
a record of earnings that fails to satisfy legal requirements and may cause
misunderstanding and definite impairment of the rights of particular
classes of investors.93

Paton also apportioned the appreciation as between the real un¬

realized gain and the nominal unrealized gain. The latter was
treated as a restatement of the Capital and Surplus accounts, while
the former (being the residue after reducing the aggregate change in
cost by the movement in the general price level) was denominated

92. Paton, Advanced Accounting, pp. 343-47.

93. Ibid., p. 349.
120 Stephen A. Zeff

as unrealized gain—and was credited to the special stockholders

equity account.94
Paton likewise argued against the inclusion of land appreciation
in income, "as the enhancement does not represent realized or dis¬
posable income."95 Here the role of income as a dividend base
intrudes.
In a discussion of "Appreciation and Income" in the chapter on
"Income Determination—Revenue," Paton rehearses the general ar¬
guments against booking unrealized appreciation as income. He
summarizes his position as follows:

Many accountants agree that where enhancement in value is unmis¬

takable and is present in large amount it is not objectionable to give
recognition thereto in the accounts and statements, although there is no
consensus of opinion as to the details of how this should be done. Very
few, however, are willing to view unrealized appreciation as genuine
business income, no matter how apparent and fully validated the in¬
crease in value may be.
The legal attitude is of decided consequence, if not controlling. The
accountant is under obligation to report income in a manner that is
acceptable in the eyes of the law, although this obligation does not pre¬
vent the disclosure of other significant figures.96

In Advanced Accounting, Paton repeats a suggestion which first

appeared in the 1940 monograph, that a reacquisition by a corpora¬
tion of a significant slice of its outstanding stock at a price different
from its book value might be an occasion for recognizing an im¬
plied undervaluation or overvaluation of the corporation's assets.97
This procedure would be parallel to a recommendation routinely
made in leading textbooks in regard to partnerships, when the
ownership interest of one of the partners is purchased by the
partnership at a price above or below its book value—justifying in
some circumstances the accounting recognition of an increase or
decrease in the partnership goodwill. Thus, if a corporation reac¬
quires a portion of its outstanding stock at a price lower than its
book value, rather than recognizing a "profit from retirement," it
would make an entry to remove the implied overvaluation of the
company's assets. If 10 percent of the company's stock were pur¬
chased for $10,000 less than its book value, the overvaluation of ten
times the discrepancy, or $100,000, would be eradicated as follows:

94. Ibid., pp. 332-33.

95. Ibid., p. 377.
96. Ibid., p. 452.
97. Ibid., pp. 543-45.
 Paton on the Effects of Changing Prices 121

Estimated Loss of Capital and Surplus $100,000

Allowance for General Revaluation
of Resources $100,000

Although individual asset accounts may be found to be overvalued,

the Allowance account may be retained as “a blanket contra to
resources.”
Paton would confine such revaluations to reacquisitions "carried
out on terms which are equally fair to both parties—the corporation
representing the remaining stockholders on the one hand and the
retiring stockholders on the other.” Paton continues:

Assuming good accounting and ignoring wild swings in security

prices it may be said that the excess of the price of shares over book
values rests primarily on two factors: one, the amount of goodwill and
other intangibles which reflect the existence of superior earning power;
two, appreciation of existing assets which have not been traced in the
accounts. Similarly, where book value exceeds market price under the
assumed conditions the difference may largely be explained by reference
to lack of earning power and declinations in asset value due to price
movements not covered by depreciation accruals and other offsets.

Notwithstanding the evident appeal of this revaluation procedure

to one whose deeper sympathies were with the expression in the
accounts of current prices, Paton, plagued by doubts so evident in
his writings throughout the 1930s and early 1940s, recoiled from a
recommendation that the idea actually be implemented:

While [this revaluation of the stock equity] has intrinsic merit it can
hardly be explicitly employed in accounting for stock retirements. Ignor¬
ing considerations of law and assuming that a practicable procedure
could be worked out—without impairing the usefulness of accounts in
other directions—the fact remains that the retirement of a portion of the
outstanding stock could seldom be rated as a compelling reason for a
complete revaluation of the equity. Certainly the acquisition of a few
shares, at what may be a very temporary price, would not be controlling,
and even the purchase of a substantial block at an established market or
negotiated price would scarcely justify revaluation, particularly where
the accounting of the corporation has been on an acceptable basis.

His reasons for declining to recommend the procedure, especially

the vague reference to "an acceptable basis” of corporation account¬
ing, seem reluctant and unconvincing.
Paton has advised me that this revaluation procedure occurred to
him after reading Warner H. Hord's "The Flow of Property as a
Basis of Internal Control" in the Accounting Review of September
1939.
122 Stephen A. Zeff

Finally, Paton carries forward the upward quasi-reorganization,

which he first exposed to view in his Dickinson Lecture a year
earlier. Journal entries are proposed for the "fresh start," but Pa-
ton's support for the procedure is restrained, owing to the omnipres¬
ent legal factors:
On the score of simplicity the procedure under consideration is all to
the good. From the standpoint of adequate recognition of the cost basis
and full disclosure of pertinent data, on the other hand, the immediate
capitalization of appreciation is not satisfactory. Adoption of a policy of
charging appreciation to operations, with no accompanying adjustment
of profit and surplus figures, results in a record of earnings that fails to
satisfy legal requirements and may cause misunderstanding and definite
impairment of the rights of particular classes of investors.98

Advanced Accounting may be seen as the most important and com¬

plete statement of Paton's views on the impact of changing prices
on accounting since the publication in 1924 of Accounting. It re¬
flects, on the whole, a caution induced by Paton's continuing preoc¬
cupation with the legal confines of accounting. By 1941, he had
developed several innovations by which changing prices might find
their way into the accounts, but it was not until the postwar period
that he became their full-throated advocate.

Response to the Postwar Inflation

Paton was a member of the Committee on Accounting Procedure of

the American Institute of Accountants (as it was then known) be¬
tween 1939, when it began issuing Accounting Research Bulletins, and
1950. In 1945, the Committee unanimously recommended that quasi¬
reorganizations be reflected when asset values were understated as
well as overstated. Quasi-reorganizations had been accepted practice
for many years when corporations were facing severe financial diffi¬
culties and their assets were significantly overvalued. But the Securi¬
ties and Exchange Commission (SEC) firmly registered its opposition
to write-ups under quasi-reorganizations, and the Institute's Com¬
mittee declined to press the matter further.99
In 1947, the Committee issued Accounting Research Bulletin No. 33,
in which it opposed the incipient practice of recording depreciation
based on replacement cost. Paton did not vote on that bulletin, but
in 1948, when the Committee issued a letter reaffirming Bulletin No.

98. Ibid., p. 349.

99. Stephen A. Zeff, Forging Accounting Principles in Five Countries: A History and
an Analysis of Trends (Champaign, Ill.: Stipes Publishing Co., 1972), p. 156.
 Paton on the Effects of Changing Prices 123

33 in the face of criticism from within the profession, Paton dis¬

sented. It was reported that Paton and the three other dissenters
believe . . . that inflation has proceeded to a point where original dollar
costs have already lost their practical significance and that where depre¬
ciation is an important element of cost the advantages which would
result from a basic change in accounting treatment outweigh the possible
disadvantage against it.100

The postwar inflation convinced Paton that measures had to be

taken—within the legal framework—to restore the financial state¬
ments as meaningful economic indicators.
In “Cost and Value in Accounting," a March 1946 article,101 Paton
briefly refers to the (two-way) quasi-reorganization as “the neces¬
sary safety valve in a structure of accounting generally based on
cost. It furnishes a means under which a corporation may restate its
accounts, and provide a new basis of income measurement, without
undergoing an actual legal reorganization."102 In the article, Paton
emphasizes that historical costs are not to be treasured in them¬
selves, but are merely the initial indications of value, later subject
to revision as conditions change. He writes.
In fact cost is significant primarily because it approximates fair value at
date of acquisition. Cost is not of basic importance because it represents
an amount paid; it is important as a measure of the value of what is
acquired. . . . The problem in accounting is primarily the question of
how cost figures should be dealt with, and when and how they should
be revised in view of changing conditions.103

Here Paton begins to diverge from the Paton-Littleton devotion to

the fundamental importance of price-aggregates in accounting reck¬
onings.
In a March 1948 note entitled "What Is Actual Cost in Depreciation
Accounting?" Paton criticizes Bulletin No. 33 for discouraging efforts
at dealing with the kinds of changed conditions to which he referred
in his March 1946 article.104 He proposes two possible reforms: (1)

100. "Institute Committee Rejects Change in Basis for Depreciation Charges,"

Journal of Accountancy (Nov. 1948), p. 381.
101. William A. Paton, "Cost and Value in Accounting," Journal of Accountancy
(March 1946), pp. 192-99. An earlier, but similar, version of this article was drafted
by Paton in 1944 for a possible accounting research bulletin to be issued by the
Institute's Committee on Accounting Procedure. The proposed bulletin underwent
revision within the Committee but did not survive final scrutiny.
102. Ibid., p. 199.
103. Ibid., p. 193.
104. W.A. Paton, "What Is Actual Cost in Depreciation Accounting?" Journal of
Accountancy (March 1948), pp. 206-7.
124 Stephen A. Zeff

supplementary disclosures to financial statements designed to

show the limitations of the conventional, legalistic-presentation, in
which—erroneously—all dollars are assumed to be homogeneous,"
and (2) a "fresh start" or quasi-reorganization which "should be
defined in such terms as to permit application to all situations in
which recorded cost data are so far from economic realities that the
resulting reports are misleading, regardless of the arithmetical direc¬
tion of the adjustment." Ever the skillful debater, Paton, who railed
in the 1940s against the infamous "original cost" of the utility regula¬
tory agencies and against federal income taxation based solely on
unmodified historical cost, argues that "those who support replace¬
ment-cost depreciation are supporting actual cost more effectively
than those who advocate continued adherence to depreciation based
on the so-called 'actual cost.' " He and others hoped that an argu¬
ment by analogy (a favorite Patonian ploy) would persuade account¬
ing authorities, chiefly the SEC, that quasi-reorganizations could be
extended to cover upward revaluations. In that endeavor the Insti¬
tute's Committee on Accounting Procedure failed in 1945 and failed
again in 1950.105 From its inception in 1934 until the early 1970s, the
SEC was an unyielding foe of departures from traditional historical
cost in the measurement of income. Only the long-accepted use of
downward quasi-reorganizations was countenanced by the SEC
when corporations were in financial straits.
In "Accounting Procedures and Private Enterprise," an April 1948
article, Paton suggested three alternative approaches to dealing with
the impact on accounting of changing prices:106 (1) Supplementary
disclosure of the pertinent data. (2) A "compromise procedure" for
the booking of replacement cost of plant, the recording of deprecia¬
tion based on the replacement cost, but determination of net income
based on conventional, historical cost principles after the deprecia¬
tion on appreciation was washed out against a stockholders' equity
adjustment account. (This alternative was first exposed in Essentials
of Accounting and was also discussed on pages 815-16 in the third
edition of the Accountants' Handbook, which Paton edited and largely
wrote in 1943.)107 (3) A quasi-reorganization.108 He insisted that any

105. Zeff, Forging Accounting Principles, pp. 156-57.

106. W.A. Paton, “Accounting Procedures and Private Enterprise," Journal of Ac¬
countancy (April 1948), pp. 278-91.
107. W.A. Paton, ed., Accountants' Handbook (3d ed.; New York: Ronald Press,
1943). It is evident from the editor's preface on p. x that Paton himself wrote the
section on "Plant Appraisals," pp. 787-833.
108. Paton, "Accounting Procedures and Private Enterprise," pp. 289-90.
 Paton on the Effects of Changing Prices 125

accounting adjustments should “be handled in an orderly, system¬

atic manner and with complete disclosure. Arbitrary increasing or
decreasing of the depreciation charge, with no concurrent revision
of the basic figures on which depreciation is computed, cannot be
supported by the accountant." Here he evidently refers to the ef¬
forts in the late 1940s of several large companies (notably United
States Steel Corporation, E.I. du Pont de Nemours & Company, and
Allied Chemical & Dye Corporation) to record replacement-cost de¬
preciation in their published reports. In this respect, Paton reiter¬
ates the view he expressed in “Depreciation, Appreciation and Pro¬
ductive Capacity," published in 1920.
In urging accounting reform, Paton was alarmed by what he re¬
garded as a growing trend toward socialism. American industry
was being accused of profiteering and earning excessive rates of
return, and Paton feared a crumbling away of the free enterprise
system in the wake of such criticism. He foresaw "a great confisca¬
tion of private property if the state were to take over business
concerns at present book values."109 He was worried that accoun¬
tants would concentrate their reform on the income statement, ig¬
noring an understated balance sheet and the resulting overstated
measures of earning power. He believed that sound accounting
procedures were essential to a preservation of private enterprise.
On several occasions, Paton criticized his brethren for playing into
the hands of the socialists. In an address given in June 1949, he
said.

What is important is that the aggregate of the earnings of American

stockholders are being seriously overstated at a time when private prop¬
erty—particularly venture capital in the corporate field—is under attack
and undergoing inequitable treatment tax-wise, as features of a strong
current in the direction of state socialism. This is certainly a bad time in
which to be lending aid and succor to those who are trying to hamstring
and destroy private enterprise and our market economy by means—
among other things—of accounting records and reports that seriously
overstate the actual earnings of venture capital. This is a sad state of
affairs and accountants should hang their heads as a result of their part in
the process of encouraging misunderstanding and the unwarranted at¬
tack on the stake that the millions of common stockholders have in
American business.110

109. Ibid., p. 290.

110. William A. Paton, "Measuring Profits under Inflation Conditions: A Serious
Problem for Accountants," Journal of Accountancy (Jan. 1950), p. 21.
126 Stephen A. Zeff

Paton's great concern over the impact of overstated profits on deci¬

sions involving dividends, wages, prices (including especially those
determined by utility regulatory commissions on the basis of
"original cost"), and income taxes, evidently convinced him that
informative disclosures accompanying a reckoning based on un¬
modified historical cost were not an adequate solution in most in¬
stances.111 The figures in the financial statements themselves would
need to be modified—either by means of his "compromise proce¬
dure" or a full-fledged "fresh start."
Paton spoke out frequently during the late 1940s and early 1950s
on the dire effects of overstating companies' reported profits.112
Although he evidently favored the quasi-reorganization approach
when current replacement costs differed significantly from histori¬
cal costs, he also complained of the confusion wrought by changes
in the general price level.113 He attacked on two fronts: relative
price changes and economy-wide inflation. The tactical advantage
of a general price-level reform would have been its evident objec¬
tivity and the commonsense appreciation of the diminishing pur¬
chasing power of the dollar. In testimony before the Subcommittee
on Profits of the Joint Committee (of the U.S. Congress) on the
Economic Report, in late 1948, Paton suggested that "we might call
the 1948 unit the zollar, and this would encourage clear thinking
when we were comparing the present value of money with, say
the 1940 dollar."114 Paton believed that the illusion that different
years' dollars could meaningfully be added or subtracted was forti¬
fied by a continuing use of the same term to describe the mone¬
tary unit. He was later fond of saying that while accountants
would carefully allow for a spread of 2 or 3 percentage points
between the American and Canadian dollars, they would ignore
altogether changes of 5 and 10 percentage points in successive

111. For Paton's views on the controversy over "original cost," see William A.
Paton, "Accounting Policies of the Federal Power Commission—A Critique," Journal
of Accountancy (June 1944), pp. 432-60.
112. In Shirtsleeve Economics: A Commonsense Survey (New York: Appleton-Cen-
tury-Crofts, Inc., 1952), Paton makes many of his points in nontechnical terms. See
pp. 232-58.
113. See, for example, the address by W.A. Paton in Accounting Problems Relating
to the Reporting of Profits (New Wilmington, Pa.: Economic and Business Foundation,
1949), pp. 10-11.
114. U.S., Congress, Corporate Profits, Hearings before the Sub-Committee on Prof¬
its of the Joint Committee on the Economic Report, 80th Cong., 2d sess., 1948, p. 62.
 Paton on the Effects of Changing Prices 127

American dollars.115 In pressing his general price-level case, Paton

repeatedly cited Henry W. Sweeney's path-breaking 1936 book,
Stabilized Accounting.
In his congressional testimony in December 1948, Paton also
modified his view of the late 1930s that depreciation was, after all,
but a relatively small fraction of total expenses—implying that an
adjustment of depreciation would not have a material effect on net
income. In 1948, he said, “It must not be forgotten that although in
many industrial companies the reported depreciation cost figure is
not a large fraction of total expenses it may be an important figure
when compared with net income.''116
The base against which materiality should be judged was now net
income, not total expenses.
Paton was greatly disturbed over the misuse of accounting infor¬
mation. In his speeches, monographs, and testimony, he dealt in
particular with the need to reform the tax base (he recommended
replacement-cost depreciation) and to predicate wage increases on
sound measures of profit.117

Essentials of Accounting (1949)

In the discussion of the valuation of plant, the principal differ¬

ence between this second edition of Essentials and its predecessor of
1938 is the inclusion in the former of a reference to the upward
quasi-reorganization. In the two editions, the treatments of general
price-level-adjusted financial statements are virtually identical. But
in 1949 Paton includes for the first time in Essentials a “fresh start"
revaluation of assets arising from the reacquisition by the corpora¬
tion of a significant block of its own shares. Carrying forward his
discussion in Advanced Accounting (1941) and a commentary in the
third edition of the Accountants' Handbook (1943) (where, on page
1010, Paton discloses his debt to Hord), Paton writes in Essentials:

115. See, for example, William A. Paton, with the assistance of William A. Paton,
Jr., Corporation Accounts and Statements: An Advanced Course (New York: Macmillan
Co., 1955), p. 531.
116. U.S., Congress, Corporate Profits, p. 63.
117. Regarding the latter, see his testimony before the Presidential Steel Board, In
the Matter of the United Steelworkers of America—CIO and Republic Steel Corporation et
al. (Aug. 19, 1949) and before the Steel Panel of the Wage Stabilization Board, In the
Matter of the United Steelworkers of America—CIO and Various Steel and Iron Ore
Companies (Feb. 14, 1952).
128 Stephen A. Zeff

Assuming that corporation and stockholder are fully informed and acting
intelligently on a commercial basis acquisition at a figure materially be¬
low book value implies that such value is overstated from the standpoint
of going-concern value. . . . Similarly acquisition of shares at a figure
materially above book value implies that the going-concern values of
specific assets are in excess of book amounts or that unrecognized intan¬
gibles, reflecting superior earning power, are present.118

This view, which is compared above with a parallel procedure in

partnership accounting, also may be shown to be analogous to
the allocation of the excess of acquisition cost of a corporation's
interest in a subsidiary over the underlying book values of the
subsidiary's net assets. Paton's recommended revaluation of as¬
sets on the occasion of reacquisition of shares under certain con¬
ditions seems to have been generally ignored by his academic
colleagues as well as by corporations and practicing accountants. I
am aware of only a single comment by other writers on this
Patonian innovation.119 Nonetheless, Paton contends that this
version of an accounting "fresh start" should serve to indicate the
unreasonableness of the common assumption that a gain to the
corporation emerges when shares are acquired at less than book
value, or that a loss is suffered when shares are acquired at more
than book value.120
It is curious that Essentials is largely devoid of the enthusiastic
advocacy of accounting reform which typified Paton's every speech
and article of the period. The reforms are indeed mentioned and in
some instances illustrated in the book, but the passionate argument
which one finds in Paton's other writings, including some of his
later textbooks, is missing here.

Asset Accounting (1952)

Asset Accounting was the first of two volumes (the other was Cor¬
poration Accounts and Statements [1955]) which collectively consti¬
tuted the revision of Advanced Accounting (1941). Here one finds
Paton the advocate and debater in full cry. Fresh from the battles of

118. William A. Paton, Essentials of Accounting (New York: Macmillan Co., 1949),
pp. 710-11.
119. The lone comment appears in the finance literature. See Charles D. Ellis and
Allan E. Young, The Repurchase of Common Stock (New York: Ronald Press Co.,
1971), pp. 128—32. I am grateful to William A. Paton, Jr. for this reference.
120. Ibid., p. 712.
 Paton on the Effects of Changing Prices 129

the late 1940s and early 1950s, Paton carries the fight into this inter¬
mediate-level textbook.121
For internal managerial accounting, Paton recommends that all
calculations be based on current prices, not historical costs.122 In
regard to published financial statements, Paton admits that it was
“very discouraging" that corporations were doing so little to ex¬
plain the limitations of conventional, unadjusted figures.

Corporate earnings have been overstated by many billions in recent

years, and such figures have been widely used to fan the fire of misrep¬
resentation of the facts of business operation, but neither accountants
nor business managements have done much of anything to put the fig¬
ures being issued in proper perspective. Public accountants continue to
certify financial statements, and insist that they fairly present the pic¬
ture of earnings and financial position, when they are in fact downright
misleading.123

This is a bold indictment to appear in a primer on financial

accounting.
In a 25-page chapter on "Adjusting Plant Costs and Deprecia¬
tion," Paton presents and illustrates his three alternative ap¬
proaches to dealing with the impact of changing prices: supplemen¬
tary statements showing plant cost, depreciation charges, and
stockholders' equity on a current basis; the "compromise proce¬
dure"; and the "fresh start." For the first time, Paton illustrates the
income statement which would result from adoption of the "com¬
promise procedure" (although on page 816 of the Accountants' Hand¬
book (1943) Paton suggests how the statement might appear). His
two subtotals would seem to be parallel to Edwards' and Bell's
"current operating profit" and "realized profit";124 Paton, however,
would credit Edwards' and Bell's "realizable gains" directly to
stockholders' equity. Paton's "compromise procedure" income
statement would appear as follows:125

121. William A. Paton, with the assistance of William A. Paton, Jr., Asset Account¬
ing: An Intermediate Course (New York: Macmillan Co., 1952). Internal evidence
strongly suggests, and correspondence with William A. Paton, Jr. confirms, that
Paton the elder was the de facto author of this book and of Corporation Accounts and
Statements (see below). Accordingly, Paton, Sr. will be cited as the author of these
works.
122. Paton, Asset Accounting, pp. 333-34.
123. Ibid., p. 334.
124. Edgar O. Edwards and Philip W. Bell, The Theory and Measurement of Business
Income (Berkeley, Calif.: University of California Press, 1961), esp. chaps. 4-6.
125. Paton, Asset Accounting, p. 343.
130 Stephen A. Zeff

Sales and other revenues $500,000

Expenses, taxes, and other charges
(including depreciation on a current cost
basis, $12,500) 465,000
Net earnings [Edwards' and Bell's "current
operating profit"] $ 35,000
Increase in plant cost absorbed in year's
expenses 7,500
Net earnings on unadjusted basis [Edwards'
and Bell's "realized profit"] $ 42,500

In the course of discussing the "compromise” and "fresh-start"

procedures, Paton offers the following criterion for their use: "Only
where a major change [in prices] has occurred, that bids fair to be
quite persistent, should restatement be undertaken, and this hap¬
pens only during periods of serious inflation and deflation."126
Asset Accounting is rich in colorful expression and is a vintage
example of vigorous Patonian argument.127 In this chapter alone,
we are advised, for example, that the "fresh-start" procedure is
"simpler than the compromise approach, which involves riding
two horses at once.”128 The efforts by several large corporations in
the late 1940s to record replacement-cost depreciation are derided
as "not very well thought out," "incomplete," "somewhat slip¬
shod," and "half-hearted tinkering with the income statement
alone, with no systematic, well-grounded adjustment across the
board."129 The reader is informed that "in the depression days of
the early '30s, for example, one of the favored indoor sports in
financial management was that of writing down plant." Paton was
unflinching in his criticism of those who would applaud write¬
downs in the early 1930s "no matter how sloppy the procedure
and questionable the amount," but who would oppose postwar
write-ups which were "carefully substantiated and well worked
out on the procedural side."130 Paton's thinly veiled denunciation

126. Ibid., pp. 349-50.

127. Doubtless one of the more graphic characterizations in accounting textbooks
of managerial misunderstanding of accounting information occurs in Paton's discus¬
sion of the interpretation sometimes placed on fully amortized fixed assets: "Of
course the owner of the amortized facilities can shut his eyes to their economic
significance, if he doesn't know any better, but so can the owner of similar assets
recently purchased, if he is determined to make an ass of himself." Ibid., p. 309.
128. Ibid., p. 346.
129. Ibid., p. 351.
130. Ibid., p. 354.
 Paton on the Effects of Changing Prices 131

of the SEC's undeviating opposition to departures from conven¬

tional practice leaps from the page:

Here is a conspicuous example of the fatuous worship of "decrease" and

the indiscriminate condemnation of "increase" long characteristic of
many people in financial and accounting circles. It's a ridiculous attitude,
with no redeeming savor of professional integrity or competence, and
deserves thoroughgoing repudiation by all who are interested in ac¬
counting as a means of providing useful and valid compilations and
interpretations of the facts of business operation.131

Paton declined to recommend any of the three alternative ap¬

proaches as possessing the greatest merit, although he was already
on record in favor of the "fresh-start" procedure.132
In a separate chapter, Paton discusses the problems which must
be solved in the valuation of plant. For going-concern purposes, he
prefers replacement cost to a capitalization of future earnings, as the
latter is seldom practicable.133
Finally, Paton continues to espouse the accounting recognition of
"marked appreciation" in the price of shares, although the reader is
encouraged to infer that the credit would be to a special stock¬
holders' equity account, and not to income or to Retained Earn¬
ings.134 This position is unchanged from that in Advanced Account¬
ing (1941).
On the whole. Asset Accounting shows a clear preference for
modifying conventional accounting for plant costs and depreciation
charges. Paton was greatly disturbed by allegedly misleading infer¬
ences being drawn from conventional financial reports, and he had
launched a vigorous campaign for reform.

Corporation Accounts and Statements (1955)

As the Patons observe in the preface of Corporation Accounts and

Statements, "It would not have been altogether inappropriate to have
adopted the title 'Equity Accounting' " for the book, as it deals with
"the array of problems having to do with the rights of common
stockholders, senior investors, and other interested parties.13S

131. Ibid.
132. In addition to his remark in Accounting Problems Relating to the Reporting of
Profits (see above), he took this position in "Measuring Profits Under Inflation Con¬
ditions," p. 25.
133. Paton, Asset Accounting, p. 360.
134. Ibid., p. 130.
135. Paton, Corporation Accounts and Statements, p. v.
132 Stephen A. Zeff

Paton's discussion of the accounting recognition of asset value

appreciation or declination based on a divergence between the
book value and acquisition price of treasury shares is treated more
elaborately and with greater enthusiasm than in Advanced Account¬
ing (1941). Such acquisitions are divided into two types: those
based on an "uneven-deal" assumption and those supported by an
"even-deal" assumption. Acquisitions of the uneven-deal type rep¬
resent bargains or are otherwise "onesided and unfair" and would
not yield sound inferences about undervalued or overvalued as¬
sets.136 Where, however, "the terms of the acquisition transaction
are fair to both parties"—the "even-deal" assumption, which Paton
regards as the much more reasonable interpretation—a revaluation
of assets could be defended and would be followed, Paton suggests,
by a "fresh-start" accounting for any subsequent depreciation
charges.137 Paton's argument for a revaluation based on the "even-
deal" interpretation is energetically advanced:

Where ... a substantial fraction of the outstanding shares is acquired

(say 5% or more), and the amount paid has been carefully determined in
the light of prevailing conditions, and, moreover, there is a substantial
difference between book value and the determined market value, it can
be urged that a restatement of recorded values, in harmony with the
settlement made for the acquired shares, would be fully justified.138

Paton continues to oppose the inclusion in income of unrealized

appreciation, although his reasoning has changed from that in Ad¬
vanced Accounting. In both books, he cites six "usual arguments"
against this practice. As recited in Corporation Accounts and State¬
ments, they are:

1. Appreciation of existing assets does not bring about any increase in

liquid funds and hence cannot be made a basis for dividend dis¬
bursements or payments for any other purpose.
2. An increase in the market value of property intended for sale may be
offset by a later decline and hence may never be realized; an increase
in the value of property held for use in the business, even if persis¬
tent, will be realized only through the process of consumption in
production or unexpected liquidation.
3. Appreciation is based on estimate and hence the amount of the en¬
hancement is uncertain.

4. Reporting of appreciation in the income statement, even if the item

136. Ibid., p. 191.

137. Ibid., pp. 198, 202.
138. Ibid., p. 204.
 Paton on the Effects of Changing Prices 133

were clearly labeled and excluded from operating results, would not
be conservative practice.
5. Appreciation is likely to be largely a reflection of a change in the
general level of prices, and thus represents an adjustment of capital
rather than true income.
6. Appreciation has no legal standing as income.139

In both books, Paton regards the first argument as "decisive” for

fixed assets but possessing "less force" for current assets. Yet even
in the case of current assets, Paton continues, the appreciation may
not be in the form of disposable funds. Also in both books, the
second point is said to have merit. "Obviously," he writes, "appre¬
ciation of asset value is not on a level with accounts receivable,
resulting from delivery of product, as evidence of the presence of
revenue. Aside from the possibility of failure to collect, usually of
minor importance, revenue is conclusively and finally determined
by sale of product, whereas realization of appreciation depends on
future events." In both books, Paton views the third and fourth
arguments as "less serious," since "no one has ever advocated
booking appreciation except where the evidence is convincing
[sz'c!]," and accountants have, in his opinion, placed undue reliance
on conservatism.140
In regard to the fifth and sixth arguments, however, Paton takes
different positions in the two books. In Advanced Accounting, Paton
observes that the argument is "important" when the change in
specific prices is matched by movements in the general level of
prices, but this point "is offset somewhat by the fact that conven¬
tional earned income likewise may not represent an increase in
effective purchasing power."141 In the 1955 book, however, Paton
says that the fifth argument "has special importance in view of the
inflationary condition which has prevailed in recent years," but
after repeating the countervailing point of 1941, he concludes that
"it would not improve matters to adopt the policy of regarding
nominal unrealized appreciation as a form of income."142 By "mat¬
ters," Paton could well be referring to the many misleading inter¬
pretations that he had been attributing to grossly overstated corpo¬
rate profits. Thus, Paton's new stance on the fifth argument may be
traced to changed economic conditions and especially to Paton's

139. Ibid., p. 298.

140. Ibid., pp. 298-99.
141. Paton, Advanced Accounting, p. 452.
142. Paton, Corporation Accounts, p. 299.
134 Stephen A. Zeff

great concern over the consequences of overstated profits. His

altered view of the sixth argument reflects a lesser preoccupation
with legal arguments which tended to dominate his thinking in the
1930s and early 1940s. In Advanced Accounting, Paton writes, "The
legal attitude is of decided consequence, if not controlling."143
Fourteen years later, in Corporation Accounts and Statements, he
says, "The legal attitude ... is of consequence, although not neces¬
sarily controlling."144
Finally, Paton devotes an entire chapter to what he now calls
"Uniform-Dollar Accounting." It is by far his most extensive treat¬
ment, in textbooks or articles, of general price-level accounting, and
is twice the length of the comparable chapter in Advanced Account¬
ing. He again exhibits his desire not to be confined to what the law
requires. He writes:

There is more involved [in accounting] than prevailing legal rules and
concepts. More and more the accounting process is being relied upon as
the major means of compiling and interpreting the economic data of
business enterprise, for operating management, fdr current and prospec¬
tive investors, for legislative bodies interested in basic problems of pol¬
icy, and in other connections. And accounting cannot possibly fulfill its
obligations in these directions by remaining blind to the limitations of
raw accounting data in the face of a major and sustained decline in the
purchasing power of the dollar. Moreover, there is no reason why ac¬
counting should not lead the way to changes in the legal structure as it
relates to business enterprise in so far as new conditions provide a valid
basis for such changes.145

Paton's chapter on "uniform-dollar reporting" is not a neutral,

bland presentation. It is vigorous, undiluted advocacy, drawing on
Paton's considerable skills as a debater. All of the color found in the
best examples of Patonian rhetoric is present here in abundance. He
marshals arguments of diverse kinds—-from intrinsic deficiencies in
accounting measurement to the necessity of replying effectively to
"the enemies of private enterprise"—in order to fortify the conclu¬
sion that unconverted historical costs, in periods of significant in¬
flation, yield misleading information.146 Paton concedes that figures
expressed in specific current costs would be more useful to internal
management than historical costs which are expressed in converted
dollars. But as a "year-by-year appraisal of all the assets of the

143. Paton, Advanced Accounting, p. 452.

144. Paton, Corporation Accounts, p. 299.
145. Ibid., pp. 526-27.
146. Ibid., p. 540.
 Paton on the Effects of Changing Prices 135

enterprise [is] ordinarily impracticable," converted historical cost

can be strongly defended."147 Since his first utterances on the
impact of changing prices in Principles of Accounting (1916), Paton
has preferred the specific cost solution. For tactical and pragmatic
reasons, however, he has often proposed the conversion of histori¬
cal cost into dollars of current purchasing power, but only in sup¬
plementary statements. But there was another argument for "uni-
form-dollar reporting": to place multi-earnings summaries and
other inter-year financial presentations on a comparable basis. As
corporations began to point to trends in earnings and other finan¬
cial statistics, Paton believed that the "raw figures of the successive
years are literally not comparable, and showing them in juxtaposi¬
tion without explanation [or conversion] is almost sure to lead to
misinterpretation."148
In Corporation Accounts and Statements, Paton abandons his sup¬
port of Sweeney's view that the gains and losses on monetary items
should be treated as realized only to the extent that the monetary
assets are utilized or the monetary liabilities are settled. Paton now
regards the gains and losses on current monetary items as being
realized when the relevant index changes, but the gain or loss on
long-term debt is still unrealized until it is paid. His logic is as
follows:

The recognition of a current liability has long been regarded as having

substantially the same force as an expenditure, and the current receiv¬
ables have long been regarded as providing an adequate foundation for
the booking of revenue. And when it comes to cash on hand—purchas¬
ing power as such—it is hard to see justification for the contention that a
change in value is not realized until there is an actual disbursement.149

One finds, therefore, that Paton incorporated in Asset Accounting

and Corporation Accounts and Statements the essence and full force of
the arguments he made during the late 1940s and early 1950s in
speeches, congressional testimony, submissions to government pan¬
els, and articles. Both books are faithful to the view that positive
action must be taken to accommodate accounting to an environment
of changing prices. In a sense, Paton came full circle from his undi¬
luted idealism of 1917-18, through a long period of doubt and reas¬
sessment of his position in the face of changing economic condi¬
tions, to a renewed call for accounting reform. To be sure, the

147. Ibid., p. 537.

148. Ibid., p. 554.
149. Ibid., p. 550.
136 Stephen A. Zeff

latter-day Paton was much influenced by his fundamental belief in

the efficacy of the free enterprise system and in the role which he
believed that accounting must play to counter the misguided critics
of private enterprise. This is not to suggest that Paton s specific
policy recommendations of the late 1940s and 1950s were identical
to those of the 1910s. They were not. Following the 1910s, he never
again embraced the general proposition that income should be cred¬
ited with unrealized appreciation. Furthermore, the Paton of the
1940s and 1950s was more alive to the tactics of devising a form of
revised accounting that might appeal to the authorities and the
accounting establishment. An apt example would be his tailoring of
the quasi-reorganization, a device originally intended to deal exclu¬
sively with overvalued assets, to the circumstance of undervalued
assets. Doubtless his twelve years of service on the Committee on
Accounting Procedure heightened his sensitivity to the tactics of
accounting reform. The Paton of the 1940s and 1950s was a more
clever and resourceful advocate than was the idealist of the 1910s.

Conclusion
One is able to discern four periods in the evolution of Paton's
writings on the impact of changing prices on accounting. The first
period, 1916-18, is typified by youthful idealism. Paton is confident
of his position and confidently dismisses most criticisms. He sees
an integral role in the financial statements for replacement costs,
but by the end of this period he only dimly perceives the possibili¬
ties of general price-level disclosures.
The years from 1920 to near the end of the 1930s were ones of
reassessment and reconsideration of the place of replacement costs
in financial accounting. At the outset of the period, he began to
retreat from his earlier advocacy of crediting appreciation to in¬
come. By the mid-1930s, he seemed to conclude that replacement
costs are better reported in supplementary disclosures than in the
bojdy of the financial statements. Concerns over the unscientific
appraisals during the 1920s, the economic stagnation of the early
1930s, and the growing importance of legal considerations in busi¬
ness operation seem to have been influential. During this period,
however, the possible role of general price-level disclosures ap¬
peared to grow in importance, and the publication in 1936 of
Sweeney's Stabilized Accounting was doubtless a significant event.
From the late 1930s to the end of World War II, Paton devised a
number of accounting stratagems by which replacement costs might
find their way into the financial statements without doing violence
to accounting conventions. That this period coincided with the first
 Paton on the Effects of Changing Prices 137

six years of Paton's service on the Institute's Committee on Ac¬

counting Procedure may not have been altogether an accident. He
proposed upward quasi-reorganizations, revaluations based on the
discrepancy between the book value and reacquisition price of large
blocks of treasury shares, and a "compromise procedure" for book¬
ing replacement costs without passing their effects through net in¬
come. During the same period, his advocacy of the supplemental
disclosure of general price-level effects accelerated, and by the end
of the war, the positions which he had espoused in his famous 1940
monograph with A.C. Littleton, An Introduction to Corporate Ac¬
counting Standards, evidently were undergoing a change.
Following the war, his arguments in favor of an accounting recog¬
nition of replacement costs or of supplementary general price-level
disclosures, or of a combination of the two, veritably reached the
fever pitch of a campaign, and in his articles, speeches, and even
textbooks, he passionately called for accounting reform. The post¬
war attack on the excessive profits of large American corporations,
coupled with the trend which he detected toward socialism, galva¬
nized his energies to defend the free enterprise system and to urge
accountants. Congress, the SEC, and government wage tribunals to
recognize that conventional accounting practices did not accurately
reflect profits during inflationary times.
Through all of his active career, Paton has been highly attuned to
the economic circumstances in which corporations find themselves.
Although historical analysis does not admit of rigorously defensible
interpretations of the factors which might have motivated a writer
at different points in his career, the evidence adduced in this paper
strongly suggests that some kind of accounting reform to deal with
the effects of changing prices was never far from Paton's mind even
in the 1930s. Prior to the post-World War II period, his was a lonely
crusade, and his writings show that he was sensitive both to the
problems wrought by changing economic times as well as to criti¬
cisms by others of a replacement-cost solution (e.g., that the ques¬
tionable appraisals of the 1920s might reappear). Without sacrific¬
ing his basic belief that historical cost accounting per se does not
deal adequately with changing prices, he retreated in the 1930s
until he had devised procedures that could give effect to his beliefs
while overcoming the earlier criticisms. In this light therefore, Pa¬
ton's succession of changed positions on the accounting response to
changing prices reflects more his attempt to develop defensible
procedures to achieve his fundamental objectives than a rethinking
or recanting of his objectives and basic beliefs.
March 1978
Relationships among Income
Measurements
Norton M. Bedford

Consider the proposition that the base, the roots, the hidden
core, the unobserved essence from which we have derived account¬
ing income measures rests not in economics but rather in psychol¬
ogy and the physiology of the brain, with the consequence that
prevailing measures of income do not fit well with pre-established
economic constitutive definitions of income.1 In the classical sense
that a measurement is an assignment of a number to a "thing" such
as income, the inference is that accounting income measures suffer
from the prevalence of an ambiguous and ill-defined constitutive
notion of economic income. The failure of efforts to articulate an

1. The notion of a "constitutive concept" refers to the existence of a physical or

psychological phenomenon having the capacity to be observed or felt by a recipient.
It differs from an "operational concept," which refers to a measurement or a process.
For example, an increase in the wealth of an individual if observable or sensed
would represent a constitutive thing, whereas a process of assigning a number to
that thing would reflect an operation; and the operational measure may or may not
be a perfect surrogate for the constitutive element of wealth. Given the possibility
that constitutive concepts and operational concepts may diverge as the sophistication
of the measurement process develops, it seems preferable that we distinguish be¬
tween the thing (constitutive concept) and the process by which it is measured or
attained (operational concept) if we are to improve the quality of accounting infor¬
mation. The definition of income as the amount a person can spend and be as well
off at the end of a period as at the beginning of the period is a constitutive defini¬
tion. The definition of income as the measure by which revenues exceed expenses,
both defined in terms of recording procedures, would be an operational definition.
See also Norton M. Bedford, Income Determination Theory: An Accounting Framework
(Reading, Mass.: Addison-Wesley, 1965).

139
140 Norton M. Bedford

unambiguous description of the substance of income leads to the

conclusion that the content (meaning) of an income measure de¬
pends upon the measurement operations involved in bringing it
forth.
From this arises the alternative proposition that the base, the roots,
from which accounting income measures derive must rest neither in
psychology nor in economics but in a set of measurement rules. And
from this alternative base of observable rules springs forth the practi¬
cal operational view that accounting income measures—the observ¬
ables—first arise, and that recipients of the accounting measure then
infer a constitutive characteristic to the measure.
Given the existence and somewhat arbitrary use of alternative
accounting principles for developing accounting income measures,
the alternative base seems quite realistic. It is axiomatic that differ¬
ent measures cannot all be true representations of the same income.
Different constitutive concepts of income are implied by the differ¬
ent measures, and the constitutive meaning of any income measure
can only be determined by knowing the measurement processes
used in producing the measure. The rise of the alternative base
implies that while a nebulous constitutive notion of income may
have initiated attempts to assign numbers to the constitutive
"thing” known as income, the haziness of the notion has resulted
in a variety of measures each purporting to be the correct measure
of income. Empirical support for the alternative base exists in ex¬
aminations of some of the income measures, for they reveal that
different constitutive notions of income underlie the different in¬
come measurements. From this, intuitive logic brings forth the be¬
lief that the nature of accounting income depends upon the way it
is measured, because accounting income is merely the result of the
operations used by accountants to develop the measure.
This is not to say that constitutive income plays no role in mod¬
ern society. On the contrary it is the motivating force that causes
much of the activity in a society, for the bulk of man's endeavor is
directed to the acquiring of income. Not only must this desire for
constitutive income be prevalent in society if it is to develop a
civilization, but it must prevail at all levels of society if that society
is to progress. Unless the masses of a society aspire to some type of
constitutive income, the accumulated evidence of history is that the
society falls far short of those in which the income motivation is
strongly embedded. Striving by an elite at the top is not enough.
The problem with constitutive notions of income is that there are
so blessed many of them that one can seldom say which notion is
 Relationships among Income Measurements 141

entertained by a particular person at any particular time. All of this

tends to preclude the measurement of constitutive income, and from
this ambiguity the operational concept of income has emerged.
Operational income operates within limits, however, for it loses
creditability if it produces accounting income measures that exceed
the range of the constitutive concepts gathered together under the
umbrella of a broad, nebulous, economic-based constitutive con¬
cept of income. Currently the variation among income measures
that can be developed by the use of alternative accounting princi¬
ples suggests that a wide range of constitutive meanings attaches to
accounting measures of income.
The difference between the two bases—the constitutive versus
the operational—for accounting income measures is fundamental.
According to the first, income is a constitutive "thing" of sub¬
stance, possibly an increase in wealth—whatever that is—and it is
the task of accounting to assign symbols (including numbers) to
that "thing" so that information about it can be communicated
between various parties. The alternative operational base, on the
contrary, assumes that constitutive income is a subjective phe¬
nomenon, an individual feeling, that can never be measured as a
general concept. It is situation specific. Accounting measures of
income measure nothing of substance. Rather they represent bits of
information which must be interpreted to have substance; and,
while common interpretations of the substance may be made by
one group of users, the income measure precedes the interpretation
or determination of a substance or constitutive notion of income.
On the basis of the foregoing underlying philosophical notions, it
is worthwhile to examine the five extant operational views of ap¬
propriate accounting income measurement models: historical cost,
price-level adjusted historical cost, current replacement cost, current
exit value, and present value of expected cash flows. Such a survey
reveals some unusual relationships among the five. Before examin¬
ing these operational accounting income measurement models,
however, further insight and a stronger base for analysis will be
provided by a critical review of the constitutive concepts of income:
psychic, real, and money income, along with extensions of each.
Based on the works of Fisher and Hicks, the following framework of
income concepts will reveal distinctions among concepts of income.
Figure 1 suggests that the three basic income concepts are func¬
tionally related to each other but that each is ambiguous in itself.
Thus, psychic income may refer to the satisfaction an individual
obtains by consuming means to satisfy human wants, or to the
 Basic Constitutive Income Concepts
Norton M. Bedford
142

Fig. 1. Interrelationships among income concepts.

 Relationships among Income Measurements 143

increase in an individual's capacity to satisfy wants. Since a sub¬

stantial portion of such capacity to satisfy wants comes from goods
and services used, a type of real income, a connection between real
and psychic income is intuitively obvious. Real income, however,
is also ambiguous, for among other distinctions it may refer to
goods and services acquired or to those used. Presumably the dif¬
ference between goods and services acquired and those used would
represent "investments" or at least deferred consumption. Further,
assuming that the lag due to credit extensions is constant, the
amount of goods and services acquired will be related to money
spent by an entity, a type of money income, and this ties real
income to money income in some fashion. Because all money re¬
ceived without an obligation to repay in some way may be either
saved or spent, the notion of money income is ambiguous to the
extent that it varies between these two extremes.
From the foregoing basic income concepts, the notion of ac¬
counting income has been derived and used as an amount of mea¬
sured income to motivate economic activity. As an accrual concept,
without adjustment for price-level changes, it is closely associated
with both increases in money income and with real income. To
the extent that expenditures for capital goods are not treated as
expense by accountants but depreciated over the life of the goods
acquired, accounting income is associated with increases or de¬
creases in real income. But to the extent that items such as re¬
search and development expenditures are treated as expense, or
depreciation is expressed in terms of historical cost, accounting
income is related to money income. Being neither "real" nor
"money," accounting income serves as a general index of various
constitutive notions of income, or as an information item that
motivates economic activity.
The view that accounting income is merely a bit of information, or
news of limited scope, has led to the proposition that the accounting
concept of income should be replaced with the broader notion of
"economic motivating information." The expansion of accounting
income disclosures into accounting information disclosures that
would motivate and direct the economic activity of a society has
received little more than speculative attention in accounting theory,
but the accounting income concept could well be expanded to in¬
clude it in the future. Because accounting income is not a measure of
any unambiguous constitutive notion of income. Figure 1 indicates
that it represents operational income concepts and can be associated
with at least five income measurement models.
144 Norton M. Bedford

Associated with the various notions and concepts of income are

different criteria for classification, valuation, and recognition which
have led to anomalies in economic constitutive concepts of income:
situations where capital gains and losses have been excluded from
the income concept at times and included at others, where holding
gains have sometimes been distinguished from operating gains and
sometimes combined, and where both price-level adjusted and un¬
adjusted concepts of income have been advocated.
The problem with the constitutive economic distinctions is that
they are based on a rather limited view of human psychological and
physiological needs. This is a singular situation, for it is intuitively
obvious, and empirically supported, that human wants and needs
vary from person to person, from situation to situation, and from
time to time; and they even conflict with each other in a single
situation, at the same time, and for the same person. This being so,
the economic income attributes of a thing or event must vary, just
as individual human wants and needs vary. By inserting the collec¬
tive wisdom of the market price into the income notion, it became
possible to develop single-valued income measures; but just as any
average seldom fits any particular need, so single-valued income
measures represent a limited view of individual income.
Possibly awareness of this state of affairs contributed to Professor
Kenneth Boulding's doubt as to whether or not economics should
exist as a discipline. The confusion seems to have arisen because
economists have dealt with concepts of constitutive income, an in¬
coming of something of substance or constituting something real.
Even in that sense, however, the economic analyses of income have
never settled the question of whether income is the amount that can
be consumed (spent) during a period without impairing capital
(capital maintenance), or whether it is the amount that can be spent
without impairing the ability to spend the same amount in future
periods (spending power maintenance).
Given this variety of income concepts, it seems appropriate to
clarify the relation between various aspects of the various constitu¬
tive income concepts that accountants have attempted to treat by
using alternative accounting principles. The relation between the
following income notions will be examined:
1. The nature of income—psychic versus real money income
2. The notion of periodic income—subjective versus realizable versus
realized income
3. The elements of income—acquisition versus holding versus produc¬
tion versus distribution income
 Relationships among Income Measurements 145

4. The recipients of income—entity versus equity holders' versus junior

shareholders' income
5. The use of income measures—operating income versus variable in¬
come versus disposable income

Subsequently, to complete the topic under discussion, relations be¬

tween the five operational income measurement models (historical
cost, general price-level adjusted historical cost, current replacement
cost, current exit values, and discounted cash flow) will be set forth.

The Nature of Income

The relation between psychic, real, and money income exists be¬
cause they all flow from the basic concept of psychic income. Be¬
cause psychic income is not easily measurable, the more objectively
determinable notion of real physical income is substituted as a sur¬
rogate for psychic income. In this relationship, a measure of real
income is held to be an index of psychic income. Because measures
of real physical income do not provide for an income evaluation of
the net effect of a decrease in one physical real income item and an
increase in another, money income is used as a homogenizing attri¬
bute common to all real income items in such a fashion that gains
and losses are additive.
In this function money income is a surrogate for real income.
Since the accounting measures the falls between real and money
income, operationally accounting income is a surrogate for a surro¬
gate (real income) for the basic psychic income concept. While this
relation between the three basic income concepts may be ques¬
tioned, it seems obvious that unless income is assumed to be re¬
lated to psychic income it can hardly claim to play a significant role
as a meaningful motivational force in society. The surrogate rela¬
tionship must hold, but it holds loosely and it varies. If the surro¬
gate relationship were constant and a 10 percent increase in ac¬
counting income always meant an 8 percent increase in real income,
and in turn a 5 percent increase in psychic income, constitutive
interpretations of accounting income would be enhanced. But when
the surrogate relation among the income concepts varies, given the
tendency for functional fixation to develop among users of account¬
ing income measures, lags occur in the functioning of the relations
between the three constitutive income concepts. In the large and in
the main, however, the surrogates do reflect the relation between
the three constitutive concepts of income, and this relationship
holds for group as well as individual income measures.
146 Norton M. Bedford

The Notion of Periodic Income

The relation between subjective, realizable, and realized income
is a relation between periodic income concepts. Whether psychic,
real, accounting, or money income is being measured, the question
of the proper amount to allocate to a particular period depends
upon whether subjective, realizable, or realized income is to be
measured. In the ultimate sense, over the entire life of an entity, all
three concepts should result in the same total income; but for any
one period, significant variations in the income measure would be
expected.
This relation between the periodic income concepts derives from
the level of objectivity or degree of confidence that the measured
income will ultimately be realized. Thus realized income is the ulti¬
mate test that income exists, leaving subjective and realizable in¬
come as predictions of the amount of realized income that will
result from the activities of an entity in a given period. While the
management of an entity may believe that the "value" of the entity
has increased substantially in a given period (subjective income),
this belief may not be the opinion of willing buyers and sellers in
the market place who establish market price. Because market price,
as a realizable price, reflects the opinion of several, it is considered
more objective than the subjective opinion of one evaluator. If this
difference between realizable value (market value) and subjective
value is defined as subjective goodwill, the relation between sub¬
jective income and realizable income is that realizable income (/,„) is
equal to subjective income (Is) plus or minus the changes in subjec¬
tive goodwill (CGS) during the period of time:
J z= 1 ~h (Z
Lm Ls — c^s-

But market price constantly changes, and the use of it to measure

the amount of income that has been created in a period of time and
will ultimately be realized is always somewhat suspect as a measure
of periodic income. While it may be contended that the realizable
income concept allocates income more accurately to the period in
which it is created than realized income does, it must also be rec¬
ognized that income created by mere fluctuations in prices would
be more volatile over time periods (as a review of changes in stock
market prices reveals) than realized income would.
Some will contend that these very volatile changes in value
should be reflected in income to reveal or suggest the risk asso¬
ciated with the activities and thus achieve a full and complete dis¬
closure with a minimum allowance for the right to offset a gain
 Relationships among Income Measurements 147

against a loss. In answer one can only suggest that the frictions
(commissions, purchasing and selling costs, freight, idle employ¬
ees, and so on) that would result if all market value changes were
realized by the entity through a sale would far exceed the income
gain that could be realized if management were to confine its func¬
tion to the trading of its assets in the market rather than creating
new values by using the assets in the production and distribution
process.
From this it follows that realized income will tend to be more
stable and will serve to motivate economic activity toward the pro¬
duction and distribution of goods and services rather than toward
the making of trading gains and losses. Realized income does suffer
from a lesser ability to associate periodic income with the time
period in which the income is created. Hence the relation between
realizable income (Im) and realized income (IR) is that the latter
includes income created and realizable in previous periods and
excludes realizable income created in the current period (clm):
hi hri chn j>hn ■

Overall the relation between subjective income and realized income

is
IR = Is ± CGS - C(IS ± CGS) + p(Is ± CGS).

The Elements of Income

The relation between various kinds of income—acquisition, hold¬
ing, production, and distribution (IA, IH, IP, In)—is that of a parti¬
tioned set of periodic income (any combination of psychic, real,
accounting, or money, and subjective, realizable, or realized). In
this sense, the emphasis is on the conceptual independence of these
income measures and the lack of a relation between them. They do
have the common feature of being elements of total periodic income
(lT) and are thus additive
Ia + + Ip + ID = It-

The operational concept on which the partitioning of total periodic

income is based is the notion of business functions. All income-
producing activities of a business entity are attributed to the four
operations of buying (acquisition gain or loss), holding resources
(holding gain or loss), producing a good or service (production gain
or loss), and distributing a good or service (distribution gain or
loss). These income measures are based on the operations of a busi¬
ness entity and differ from income measures based on the account-
148 Norton M. Bedford

ing operations of assigning numbers to the business operations.

These two notions of operational income must be distinguished, for
only the accounting operations can provide unambiguous measures
of income.

The Recipients of Income

The problem of whose income to report for a given period arises

from the existence of multiple interests in an enterprise. To an
economic entity, independent of its suppliers of labor and capital,
income is the total increase in resources produced by the entity (1E).
Because creditors and employees claim part of the entity income,
the residual equity holders' income (ISh) is smaller than the total
periodic entity income. Further, to the extent that senior share¬
holders (holders of preferred stock) have a prior claim on the equity
holders' income, the junior shareholders' income (I}) is still less a
reflection of the economic contribution made by the economic en¬
tity in terms of value added. If all claims to entity income having
priority over shareholders' income (/,.)—wages, income taxes, inter¬
est charges—are treated as distributions of income, and preferred
dividend rights (lPR) as nonincome items to junior shareholders, the
relation among these income measures is:

Ie — tc = ISH

lj = t.SH ~~ tpR

IE ~ Ic ~ I PR = tj.

These foregoing categories of income recipients are, of course, a

gross oversimplification because the priority of rights to entity
income could be specified in some detail, if reason to do so ex¬
isted, to permit disclosure of the income of multiple types of
groups. Also excluded from the discussion is the transfer of rights
of one type of income recipient to another type. For example, the
"gain" arising from the retirement of bonds payable below carry¬
ing value would not represent an increase in entity income. Rather
it would represent a gain accruing to shareholders because of the
loss by the bondholders. It does seem fair to suggest that the
classification of income recipients and the measurement of income
according to income recipients have received far too little attention
in accounting literature. Typically the problem is discarded with
the suggestion that "different income measures are needed by dif¬
ferent users."
 Relationships among Income Measurements 149

The Usage of Income Measures

The relation between operating income, variable income, and dis¬

posable income is based on different constitutive concepts of in¬
come. Whether or not operating income should be categorized into
“regular" and “extraordinary" items is not of concern. Rather, the
distinction between operating and variable income relates to the
costs or expenses to be deducted from revenue in measuring in¬
come, while disposable income relates both to the liquidity of the
revenue items acquired and the cash flow requirements of the costs
or expenses. Defining sunk, extraordinary, and nonrecurring items
as £,, and distinguishing between revenue and expense recognized
on the accrual basis and revenue and expense recognized on a near
cash basis, the relation between these income concepts is:
Operating income (/„) less the sunk, extraordinary, and nonrecurring
items (£/) is equal to variable income (Iv) or I0 - E/ = Iv; and
If we then deduct from variable income the accrual and nondisposable
elements in income (IM), we have disposable income (ID) and lv - lM = In.
From this the relation between operating income and disposable income
is as follows: I0 — £/ — hi = In-

Assuming that the foregoing relation between the five-segment

framework of income concepts is valid, the complete structure of
the income concepts will be provided if the relation between the
five constitutive income concepts implied by the framework is set
forth. In general, as Figure 2 shows, this relation lies in the fact that
the nature of income (Framework I) is the overriding notion of
income, so that periodic income (Framework II) refers to the portion
of income, as predetermined in Framework I, that should be allo¬
cated to specific periods. Framework III deals with the components
or elements of income as prescribed in Framework II, while Frame¬
work IV separates the periodic income prescribed in Framework II
according to claimants to that income. It should be noted that a
relation seldom exists between Framework III and Framework IV,
though there are instances where the rights of income recipients or
claimants are dependent upon the existence of specific kinds or
elements of periodic income. Finally, the need to advise users of
periodic income as to its availability for use requires consideration
of Framework V.
Framework I represented by (a, b, c, d) is the set of income com¬
ponents (nature of income). Framework II represented by (e, /, c, g)
is the subset of income components assigned to one period of time
(periodic income). The four subsets (e, s, x, v), (v, x, u, g), (s, /, t, x).
150 Norton M. Bedford

Nature
of
Income

Fig. 2. Relation among frameworks of Income characteristics.

and (x, f, c, u) are respectively acquisition, holding, production,

and distribution income (the third framework of income elements).
The fourth framework refers to the subsets enclosed in the circles q
and r, which identify entity, equity, and junior shareholder income
(income recipients). Framework V refers to the partitioning of peri¬
odic income (e, /, c, g) by items outside (h, i, j, k), items between (h,
i, j, k) and (/, m, n, o), and items within the set [(1), (2), (3), (4)]. It
indicates operating, variable, and disposable income measures for
different uses (income usage).
Turning from the relation between constitutive concepts of in¬
come, and dealing with relations between measured income con¬
cepts in the area of practical affairs, the following relations are pro¬
posed between the historical cost, price-level adjusted historical
cost, current replacement cost, current exit value, and present value
of future cash flows income measurement models.
Common to all five of these operational income measures is the
view that the periodic income of the shareholder group is the clos¬
est notion of constitutive income associated with all five of the
income measurement models. But from this somewhat ambiguous
base divergences exist.
 Relationships among Income Measurements 151

The historical cost model is based on the matching concept and is

oriented toward the accountability role of accounting. According to
this model the income statement and the balance sheet articulate—
the balance sheet being an instantaneous status picture between
two income statements. The income statement aims to reflect the
effectiveness and efficiency of the entity, whether managerially in¬
duced or due to the fortunes of change. As a measure of effective¬
ness and efficiency, the results of this model do not purport to
depict the change in the economic wealth of an economic entity in
any one period. Rather, the results reflect the ability of an entity to
acquire resources and dispose of them in the same or a transformed
form. Income reflects the periodic success of the entity, and over a
longer period this success indicator would be reflected in an in¬
crease in wealth. Periodic income, according to this model, is a
surrogate for the wealth-creating ability of the entity. Its relation to
income measures produced by the other models is similar to that of
a relativity index or surrogate for the long-range average increase in
economic wealth. Conceptually, the historical cost model would re¬
flect an assumption that there is no reality to notions of economic
wealth and that the only reality is the relativity-based operational
measure of input-output comparisons. But this assumption is sel¬
dom proposed in accounting literature, and the more pedestrian
view prevails that the notion of increases in economic wealth is real
and that measures of income based on historical cost are indicative
of the extent of that increase. Some feel that variations in the pur¬
chasing power of money diminish the functional relation between
historical cost income measures and increases in wealth, and they
propose that the historical cost model be adjusted.
Theoretically the price-level adjusted income measurement model
may be used in conjunction with all the other income measurement
models. It is, however, typically associated with the historical cost
model and represents an adaptation of it. Effectively, it adjusts his¬
torical cost data for changes in the general purchasing power of
money, and thus removes from income as a measure of efficiency
and effectiveness those historical cost "gains” or "losses” due solely
to changes in the monetary measuring unit. Because the general
purchasing power of money changes, particularly in times of infla¬
tion, the unadjusted monetary unit is not a stable measuring unit. By
applying an index of changes in the purchasing power of money to
historical cost data, the accounting measuring unit can be stabilized.
While it might seem that the relation between income based on
historical cost and that based on historical cost adjusted to reflect
152 Norton M. Bedford

purchasing power is merely the variation in the purchasing power

of money, actually the relationship is much more complex. An addi¬
tional element of income may arise, for gains and losses may occur
if the entity has monetary debts or receivables that can be paid in
cheaper money. Currently our monetary-based legal contracts are
expressed in terms of the unadjusted monetary unit. Until this lag
in legal custom is corrected, the gains and losses due to the mone¬
tary position held by a firm will go unrecognized unless price-level
adjusted accounting is applied. Whether or not the accounting ad¬
justment is made, the “real" gain or loss arising from the holding of
a debit or credit net monetary position will continue to misdirect
managerial efforts until the legal profession adjusts the wording of
monetary-based contracts to a uniform monetary unit. The old basis
of expressing monetary obligation in terms of gold is no longer
available, and there is a social need for an alternative method of
expressing such obligations. Until this is accomplished, accounting
can help solve the problem by revealing the extent of the monetary
gain or loss in the hope that disclosure will ultimately result in
corrective action.
The price-level adjusted income measurement model is related to
the historical cost model in the following manner: "gains or losses"
due to changes in the purchasing power of money are separated
into "fictional" and "real" component parts. The price-level ad¬
justed income measurement model does not require a change in the
revenue recognition point from that used for the historical cost mod¬
el. Typically it too is based on the matching of revenues and ex¬
penses and serves to measure the efficiency and effectiveness of
managerially controlled operations.
The current replacement cost income measurement model, which
must be distinguished sharply from the price-level adjusted model,
may be viewed as a moving forward of the revenue recognition
point from the point of sale date to the balance sheet date, while the
price-level adjusted model may or may not make the shift. The
excess or deficiency of current cost over historical cost on unsold or
unused assets is included as income (holding gain) along with other
income items arising from the production and sale of goods and
services. When current replacement cost refers not to the cost of
replacing previously acquired economic resources, but rather to the
current replacement cost of the service equivalent, the gain or loss
involved may arise not only from a change in the revenue recogni¬
tion point but also from a change in the values in the balance sheet.
Income under this latter concept is the result of the difference in
 Relationships among Income Measurements 153

two balance sheets. Revenue is the increases in assets or decreases

in liabilities, and expenses are the decreases in assets or increases
in liabilities, both assets and liabilities being valued in terms of
what it would cost to replace the services or obligations at a balance
sheet date. It may be well to go over this distinction again, for it is
not commonly discussed in the accounting literature.
In those situations where the only change from historical cost is
that all assets are restated to current replacement cost, any differ¬
ence between historical cost and current replacement cost would be
merely a lag between the time a value increase had occurred and its
recognition in the accounting records, because both would be reval¬
ued to the same amount at the date of sale.
As a consequence, the primary difference between historical cost
income and current replacement cost income is the time the gain or
loss is recognized, though increases or decreases in current replace¬
ment cost that are later offset by decreases or increases in current
replacement cost before the point of sale would be reflected under
the current replacement cost income model, but not under the his¬
torical cost income model. In the long run, total income under the
historical cost model and the current replacement cost model would
be the same. Both would include all "gains or losses" due to
changes in the general price level as income.
When current replacement cost refers to the current replacement
cost of service equivalent, and service equivalent is defined in
terms of units of output, there is the possibility that technological
change will introduce a significant decrease in the auxiliary costs
of the productive capacity for each dollar cost investment. The
resulting increase or decrease in current replacement cost may not
result in a change in sales prices of outputs, and the operating
efficiency of the management will not be reflected by matching
revenues with the current replacement cost. That is, if income is
defined as sales price (R) less depreciation (D) and less auxiliary
costs of labor and overhead (A), an increase in D due to an in¬
crease in current replacement cost, caused by the opportunity to
reduce auxiliary costs, could be offset by a decrease in A to such
an extent that the sales price R would remain unchanged. But the
decrease in A occurs only for physically replaced productive assets.
For assets continued in use at replacement cost, the advantages of
the lower auxiliary cost would not be available to the continuing
management and a matching of the higher depreciation D against
the unchanged selling price R would not reveal managerial effi¬
ciency. Only if replacement did in fact occur would income under
154 Norton M. Bedford

the matching concept reflect managerial efficiency. The option is to

abandon matching under the effort-accomplishment (input-output
index) concept and adopt a noneffectiveness notion of income. In
this way the comparative balance sheets approach for measuring
income becomes appropriate.
The current exit value model reflects a somewhat more conserva¬
tive concept of income. It tends to be associated with the constitu¬
tive concept of disposable income. Depending upon how it is mea¬
sured (immediate liquidation, orderly liquidation in a systematic
manner, and expected regular sales price), it indicates the cash that
could be available for reinvestment (immediately, near term, or in
the ordinary course of business) if the entity wished to reinvest. To
the extent that the three alternative measurement methods yield the
same measure of total assets-liabilities net, the entity possesses the
capacity to adapt and change the type of economic activities in
which it engages. In the instance of immediate cash liquidation, the
measured income is the amount that the entity can spend and have
the same amount available for investment as existed at the begin¬
ning of the period. The current exit model may well be appropriate
for a society under a high degree of uncertainty and rapid change
where the second-best alternative use value of a resource is signifi¬
cant as an indicator of the entity's ability to avoid the risk asso¬
ciated with the great uncertainty and rapid change. In a situation
where it is not expected that resources will be sold for their second-
best use, but will be used for the productive purpose for which they
were acquired—the best use—realism would indicate that the cur¬
rent replacement cost income model would be indicative of the
entity's success.
Thus the relevance of the current exit value income model to most
notions of constitutive income is limited, not only because it is
ambiguous but also because it is not in accord with the expected
future. In a pure and perfectly competitive market, with no market
frictions, the difference between the current exit value model and
the current replacement cost model would be negligible. The rela¬
tionship between the two is therefore founded on the extent of the
frictions, risk, and imperfect competition of the market.
While the current exit value model assumes the sale of assets in
their current form, the present value of the future cash flows income
measurement model assumes the opposite: that the assets will be
used according to management plans. The assumption is that a
discount rate can adjust expected future cash flows for risk and for
waiting and that the firm itself, knowing how its resources will be
 Relationships among Income Measurements 155

used, is in the best position to estimate future cash flows, whose

present value is the most realistic value of the firm's resources and
obligations. The resulting valuation of the entity at a point in time
is rather subjective; and, while in a world of certainty the present
value model would provide an ideal measure of income, the world
is uncertain, and periodic income measures based on it might prove
highly unstable over time as management expectations changed.
Given the foregoing basic relationships among income measure¬
ment models, opportunities arise to develop specific relations be¬
tween them. Such is beyond the scope of this paper, but the follow¬
ing list of income measurement issues suggests the types of specific
relationships in need of development.

1. What is the relation between income measured as the difference be¬

tween the value of the assets and liabilities at the beginning and end
of the accounting period and income as the difference between the
matching of revenue and expense of the period?
2. What is the relation between articulated and unarticulated income
statements and balance sheets (in the sense that all activities reflected
in the income statement are also reflected in the balance sheet)?
3. What is the relation between assets defined as cash or cash equivalent
and assets defined as means to future economic benefits?
4. What is the relation between liabilities defined as legally binding
obligations and liabilities defined as obligations of all types?

This paper closes with the proposition that it is far more impor¬
tant for accountants to investigate the relation between income
measures than to determine which income measure is "right/' for
there is no universal right. Disclosure of these relationships will
enhance the understanding and usefulness of accounting income
measures.

January 1977
Cost Accounting Principles for
External Reporting
A Conceptual Framework

Gordon Shillinglaw

Unlike most of the other contributors to this volume I have never

met William Paton. I haven't emerged untouched by the Paton pres¬
ence, however. Most accounting texts when I started through were
mainly descriptive, but when Frank Smith took a small group of us
in hand at the University of Rochester in 1947 he gave us Paton's
Advanced Accounting (Macmillan, 1947). Now that was a book of a
different color! It challenged us, tested us, made us think. Even in
print, Paton made us feel the force of his intellect.
My second exposure to Professor Paton came two years later,
when I taught a section of the only accounting course offered to the
undergraduates of Harvard College. The coordinator of the course
was Edwin Frickey, a senior professor in the Harvard economics
department. He had persuaded the Macmillan Company to keep
enough copies of Paton's Accounting, originally published in 1923,
in stock to service the Harvard course. No other text would do. And
he was right. The writing was just as fresh in 1950 as it had been a
generation earlier.
Paton's 1923 text was written long before we had official bodies to
codify accounting standards. I'd like to think my contribution to
this volume comes from the same tradition. Cost accounting, like
financial accounting, has developed primarily because practical
people have needed answers to questions arising in the course of
their daily work. A body of concepts, principles, and practices has
evolved over the years, gradually becoming what might be referred

157
158 Gordon Shillinglaw

to as generally accepted cost accounting principles. These principles

are reflected in textbook descriptions and in the systems designed
and used by practitioners, but they have never been codified by
any authoritative body. The purpose of this paper is to outline a
conceptual framework to serve as a foundation for such a code.

The Reasons for a Cost Accounting Code

A codification of generally accepted cost accounting principles

certainly isn't indispensable for managerial accounting. Each man¬
agement is free to perceive its needs in its own way and to adopt a
set of measurement principles that seem to meet those needs. Text¬
book descriptions can provide the necessary conceptual foundation
for systems of this sort.1
Pressures for codification, if any, arise from the needs of people
outside the organization, those who must deal with information
emerging from many different organizations at the same time.
These users of information are investors, tax collectors, and pur¬
chasers of goods and services, through cost-based contracts, as well
as regulatory bodies and even private individuals who interest
themselves in the social aspects of business organizations. All of
these groups have one element in common: they are interested in
that part of cost accounting that determines how much cost is to be
assigned to each of the goods and services the organization has
produced or sold. This is the field of product costing, and it is here
that the framework developed in the present paper is to apply.
Even in this field codification has had little appeal in the past.
Authoritative financial accounting bodies have limited themselves
to broad statements such as, "[T]he exclusion of all overheads from
inventory costs does not constitute an accepted accounting proce¬
dure.''2 The Internal Revenue Service hasn't been much more spe¬
cific, and even the Armed Services Procurement Regulations have
provided only the broadest of guidelines to govern the use of cost
accounting for contract-costing purposes.
The establishment of the Cost Accounting Standards Board in 1972
created a new situation. Although the pronouncements of the board

1. One of the better pronouncements on cost accounting was issued some twenty
years ago by the Cost Committee, American Accounting Association, "Tentative
Statement of Cost Concepts Underlying Reports for Management Purposes," Ac¬
counting Review (April 1956), pp. 182-93.
2. Professional Accounting Standards, vol. I (New York: Commerce Clearing House,
Inc., 1977), sec. 5121.05.
 Cost Accounting Principles 159

have only a limited direct applicability, many observers are con¬

vinced they will have a far greater indirect influence. A conceptual
framework therefore becomes essential, as a basis for evaluating
whether the board's standards should be extended to other areas. It
may also influence the board's own standard-setting process.

The Conceptual Framework

The conceptual framework outlined in this paper will emerge full¬

blown as an integrated set of relationships among six groups of
elements: concepts, principles, constraints, criteria, standards, and
methods:
1. A concept is an understood abstraction of a class of relationships, a
trait, or a characteristic (e.g., cost is a concept describing a relation
between a sacrifice of a resource and a result of that sacrifice).
2. A principle is a normative statement of how a concept or concepts
should be applied.
3. A constraint is a boundary of the zone within which principles are
allowed to govern.
4. A criterion is a test to be applied to alternative principles or alterna¬
tive methods to help an interested party choose among these alterna¬
tives. Criteria are designed to reflect the objectives of the people
applying them, or of the organizations these people represent, or of
the purposes to be served. (E.g., in choosing between a full cost
principle and a partial cost principle, the cost accountant should con¬
sider the purpose for which the measurement is intended.)
5. A standard is a prescription specifying the principles to be followed
and the ways in which they are to be applied.
6. A method is a set of instructions and procedures by which the organi¬
zation's personnel will assemble cost data to meet the needs of users of
these data. Methods should be consistent with principles appropriate
to these needs and with the standards embodying these principles.

The relationships among these elements can be summarized in a

simple diagram:
Criteria

Concepts

Methods
160 Gordon Shillinglaw

This paper will examine the relationships among concepts, con¬

straints, criteria, and principles, with only passing reference to the
codification process. Generous use will be made of materials pub¬
lished by the Cost Accounting Standards Board, but no attempt will
be made to evaluate the standards issued so far by the board.

Basic Concepts

Concepts are essentially definitions of useful terms. As such, they

form no structure in themselves; only when reflected in criteria and
principles do they have an impact on the system. The following
twelve concepts are the most important and must be described:

1. Cost objective
2. Cost
3. Causality
4. Traceability
5. Variability
6. Service generation
7. Capacity occupancy
8. Interchangeability
9. Interdependence
10. Homogeneity
11. Attributability
12. jointness of cost

A number of other concepts, such as accuracy, verifiability, and

materiality are more generally familiar and will be introduced only
in connection with the discussion of the principles in which they
are embodied.

Cost Objective
A cost objective in an organization is a result of the use of re¬
sources in that organization. Some results are productive—that is,
useful products or services emerge from the use of resources. Other
results are unproductive, meaning that the resources have been
wasted. Costing systems for some purposes may recognize both
productive and nonproductive results; systems for other purposes
may recognize only productive results.
The terminology is slightly unfortunate here. Only productive
results can be truly regarded as objectives of the use of resources.
 Cost Accounting Principles 161

Waste is never an objective. (Lighting a cigar with a $100 bill is

wasteful, but the objective is to demonstrate affluence, not to create
waste.) The term is widely used, however, and creating a new term
would only compound the confusion.
Cost objectives are of two kinds: final cost objectives and inter¬
mediate cost objectives. A final cost objective is a specific end result
or output of the application of resources in a given period, contrib-
uting to no other final cost objective during that period. An interme¬
diate cost objective is any result of the use of resources in a given
period that contributes to or supports two or more other cost objec¬
tives in that period.3
An intermediate cost objective may be the activity of an individ¬
ual responsibility unit, or even a cost center within a responsibility
unit. It may also be a group of similar activities carried out by many
different responsibility units. Because intermediate cost objective is
an awkward term, the terms cost center and cost pool may be used
instead. A cost center is defined for this purpose as a unit in the
organization structure serving as an intermediate cost objective; a
cost pool can be any kind of intermediate cost objective, but gener¬
ally the term designates one that is not also a cost center.
The structure of intermediate cost objectives may be very simple
or very complex. In the simplest situation each intermediate cost
objective serves or supports only final cost objectives. In complex
situations long chains of intermediate cost objectives may be neces¬
sary to link the use of resources to the final cost objectives.

Cost
It was necessary to define cost objective before defining the more
basic term, cost, because cost is always the cost of something. The
structure of cost objectives must be built before costing can take
place. Given that, cost can be defined as the amount of physical
resources used to achieve a cost objective.
Since statements of physical resource consumption are difficult to
use, physical quantities are usually multiplied by unit prices. Cost,
therefore, can be defined further as a monetary measure of the
sacrifice accompanying the use of resources to achieve a specified
cost objective. (Notice that this definition can encompass historical

3. These definitions and a few others were developed jointly with Robert N. An¬
thony in connection with a joint submission to the staff of the Cost Accounting
Standards Board. Some words have been changed; in any case Professor Anthony
bears no responsibility for the ways these terms are used or interpreted in this paper.
162 Gordon Shillinglaw

cost, opportunity cost, or even estimated future cost, depending on

the purpose of the system and the constraints within which it must
operate.)

Causality
Causality is one of the most important concepts underlying cost
accounting, and may be the least understood. According to one
dictionary, causality is "the relationship between cause and ef¬
fect."4 In the present context, causality is a relation between a cost
and a cost objective whose nature is such that the cost must be or
must have been incurred if the cost objective is to be achieved. If
such a relation exists, it can be said that the cost objective "causes”
the cost.

Traceability

Traceability is an attribute of a cost that permits the resources

represented by the cost to be identified in their entirety with a speci¬
fied cost objective on the basis of verifiable transactions records.

Variability

Variability is the responsiveness of total cost in a period to varia¬

tions in total operating volume during that period.

Service Generation

Service generation is not an attribute of a cost or even of cost

behavior. Instead, it identifies a relation between the activities of a
specific cost center or group of cost centers (the "supplying cost
centers") and two or more other cost objectives. It signifies the
existence of a quantifiable measure of the amount of work done by
the supplying cost center to serve each of the cost objectives receiv¬
ing the service.
Service for this purpose may be either tangible (physical goods)
or intangible (personal advice or assistance). Consumption of ser¬
vices can be measured either by the quantity of resources used by
the supplying cost center to provide the services (input basis) or by
the number of units of service provided (output basis).

4. William Morris (ed.), The American Heritage Dictionary of the English Language
(New York: American Heritage Publishing Company, 1969), p. 214.
 Cost Accounting Principles 163

Capacity Occupancy

Capacity is an attribute of a group of assets or an organizational

entity, measured by the demand for service the asset group or en¬
tity is capable of meeting in a given period. Capacity occupancy is a
measure of the demand for capacity imposed by the achievement of
a particular cost objective.
The costs of resources that are consumed to give an organization
the capacity to carry out its activities are referred to as capacity
costs. A capacity cost is incurred, in other words, to provide the
support necessary to achieve the cost objectives carried out within
the limits of that capacity.

Interchangeability

Interchangeability is an attribute of any two or more resources

that can be substituted for each other without affecting the costs of
the other resources that are required to carry out the activities to
which the interchangeable resources are devoted. Quantities of
electric power obtained from two different generating plants and
deliverable to the same consumption point are examples of inter¬
changeable resources.

Interdependence
Interdependence is a relation between cost objectives which oc¬
curs when the use of resources to achieve one cost objective will
affect the amount of resources required to achieve other cost objec¬
tives of the same level. For example, a manufacturer opening new
production facilities may transfer experienced workers from exist¬
ing operations to provide a nucleus for an efficient work force at the
new location. The existence of interdependence is demonstrated by
the consequent increase in the average cost of the final cost objec¬
tives achieved at the old location.

Homogeneity
Homogeneity is an attribute of any group of costs which are all
governed by the same set of determinants and in an identical
manner.

Attribut ability
Attributability is the responsiveness of cost to decisions to pro¬
vide and use the capacity to support individual cost objectives. The
164 Gordon Shillinglaw

attributable cost of a final cost objective is the sum of the variable

costs of meeting that objective, plus any other costs that would not
have been incurred if the organization had not provided the needed
capacity for achieving that cost objective, all other things being
equal.

Jointness of Cost
Some resources are used to achieve two or more cost objectives
simultaneously. The costs of such resources are referred to as joint
costs. The various cost objectives achieved immediately with the
aid of these joint costs are known as joint products. Allocation of
joint costs among the joint products is the most difficult conceptual
problem in cost accounting.

System Constraints

Every system is constrained by requirements implicit in the con¬

text in which it operates. Cost measurement for external financial
reporting appears to be subject to at least seven constraints:

1. Objectivity
2. Accuracy
3. Consistency
4. Verifiability
5. Measurability
6. Materiality
7. Articulation

Objectivity

Objectivity is an inescapable requirement of professional work of

any kind. The objectivity constraint is simple: the costs assigned to
any cost objective should not be influenced by biases arising from
the personal or pecuniary interests of those measuring or relying on
the costs to be assigned to that cost objective.
The meaning of this principle was set forth succinctly some years
ago:

One way of achieving objectivity is to eliminate the use of judgment

entirely; quantities measurable without the exercise of the measurer's
personal judgment are always objective. This solution may be incompat¬
ible with other [principles], however, and in such cases the choice of a
 Cost Accounting Principles 165

cost standard will have to be a compromise, accepting some application

of judgment to achieve a greater degree of something else.5

Accuracy

Accuracy is an attribute of the costs assigned to a cost objective.

The accuracy of a cost total is measured by its degree of fidelity to
the cost concept that the cost total is intended to reflect. The users
of cost information have the right to assume that the information
provided them is accurate, subject to the levels imposed by the
measurability and materiality constraints.

Consistency

It is unlikely that cost accounting standards will ever be so spe¬

cific that only one measurement method will be possible in any
given situation. This means that the amount of cost assigned to a
cost objective will be affected to some extent by the methods chosen
to implement the measurement principles. Accordingly, the user of
cost information has a right to expect that costing methods will not
be changed without good reason and due notification.
This notion is embodied in the consistency constraint, which
others have phrased in the following terms:

Cost measurement should be consistent from one period to another and

even within an accounting period, unless there are significant reasons for
changes. Changes in measurement methods should be made (a) to in¬
crease their accuracy materially . . . , (b) to attain an adequate degree of
accuracy at a lower record-keeping cost, or (c) to comply with changes in
standards.6

Verifiability

Accounting figures for different purposes are subject to differing

verification requirements. A good definition of verifiability
emerged from a committee of the American Accounting Associa¬
tion: “Verifiability is that attribute of information which allows
qualified individuals working independently of one another to

5. Subcommittee on Basic Cost Concepts, National Association of Accountants,

"Tentative Concepts for Cost-Type Contracts," Management Accounting (May 1971).
6. Management Accounting Practices Committee, National Association of Accoun¬
tants, Concepts for Contract Costing: Statement on Management Accounting Practices No.
1 (New York: National Association of Accountants, 1971), p. 5.
166 Gordon Shillinglaw

develop essentially similar measures or conclusions from an exami¬

nation of the same evidence, data or records."7
Verification is always important in external financial reporting
because of the interests of outsiders in the results. In cost measure¬
ment, the question to be verified is whether the appropriate mea¬
surement principles have been followed. The verifiability constraint
is that the amount of cost assigned to each cost objective should be
verifiable by independent professional observers through examina¬
tion of the assumptions, records, and procedures which led to the
assignment of costs to that cost objective.

Measurability
Some relations between resources and cost objectives are easy to
identify; others are very difficult. Measurability is an attribute of a
relation between a cost and a cost objective which enables the
accountant to measure that relationship at a reasonable cost. Cost
measurement is not an end in itself, and the purposes of assem¬
bling cost data may be achieved in other ways than by expending
the costs and effort required to implement the measurement prin¬
ciples without modification. The fifth constraint, therefore, is the
measurability constraint: any measurement principle may be
modified if the modification will reduce the cost of implementa¬
tion sufficiently to justify the resulting decrease in the accuracy of
cost measurement.
Acceptance of the measurability constraint is only the first step,
of course, and probably the easiest one. It requires that the appro¬
priate standard-setting bodies undertake to develop practical tests
of measurability. This task is far from easy, but it is better to estab¬
lish measurability as a requirement than to leave it to individual ad
hoc decisions at the point of measurement.

Materiality
No conceptual framework in accounting would be complete with¬
out some reference to the concept of materiality. This constraint
differs from the others in that it does not refer to the measure of
cost itself or to a relation between cost and a cost objective, but
rather to errors embodied in the measure of cost. Materiality refers
to any error large enough that a user of cost information would be

7. American Accounting Association, A Statement of Basic Accounting Theory (Sara¬

sota, Fla.: American Accounting Association, 1966), p. 7.
 Cost Accounting Principles 167

led to a different decision from the one which accurate information

would lead to.
The materiality constraint is simply stated: departures from the
measurement principles will be accepted as long as they do not lead
to material changes in the amounts of cost assigned to individual
cost objectives.

Articulation
The final constraint is a requirement that the financial statements
articulate—that is, for any given measurement basis, such as histori¬
cal cost, the total amount assignable to cost objectives in any given
period is fixed. If a change in a cost assignment principle increases
the amount of cost assigned to one cost objective, it must decrease
the amount of cost assigned to one or more other cost objectives.
Articulation is not an essential feature of the structure outlined in
this paper. The articulation requirement is so ingrained in profes¬
sional accounting practice, however, that it must be taken as given.
According to generally accepted financial accounting principles, for
example, finished goods inventories and the cost of goods sold
must not be measured in such a way that the total amount assigned
to these two quantities exceeds the total historical cost of the goods
available for sale during the period. Any product costing system
today must operate within this constraint.

Criteria for System Choices

The central element in any conceptual framework is the criterion

(or set of criteria) to be applied to system choices. Four possible
criteria need to be examined:

1. Relevance
2. Causality
3. Benefit
4. Equity (fairness)

Relevance
It was considered a great coup some years ago when a commit¬
tee of the American Accounting Association adopted the relevance
criterion as a key element in its structure of accounting theory.8

8. American Accounting Association, A Statement of Basic Accounting Theory.

168 Gordon Shillinglaw

Relevance in this context means the capacity to improve the deci¬

sions made by users of the system.
While this is a valid attribute of any system, it is not a very useful
criterion. Usefulness to the user depends on what the user wants
the cost figures to mean. User surveys haven't addressed this issue,
and it is doubtful whether they would throw much light on it in
any case. The answer must come from an analysis of the implied
meaning of cost information, a process that requires examination of
other criteria.

Causality

The choice of an operational criterion to be used in selecting

costing principles for any purpose depends on the quantity which
the cost figure purports to measure. The main clue therefore rests in
the definitions of cost and cost objective, the two key concepts
embodied in any measure of cost. These definitions, it will be re¬
called, require measures of the relations between resource con¬
sumption and the results of resource consumption. Another way to
express the same relations is to say that the costs assigned to any
cost objective (the result) should be the costs that have led to the
achievement of this cost objective.
This is an expression of a causal relation, in that cost objectives
are achieved only as a result of a purposive application of resources.
The primary costing criterion arising from this definition of cost
therefore is the causality criterion: costs should be assigned to cost
objectives on the basis of the underlying causal relations between
the costs and the cost objectives.
Since this criterion emerges from an analysis of the definition of
cost, it would seem to be incontrovertible, not to say tautological. It
is challenged, however, as we shall see in a moment, and support
from other sources will be helpful. The principal support derives
from the interests which outsiders have in the cost figures. The
interests of those associated with contract costing are the easiest to
deal with. Cost is to be used to determine the amount of the direct
benefit to the organization from achieving the final cost objectives
associated with specific contracts. Since the size of the benefit is
directly related to the size of the cost, the cost can be regarded as
the cause of the beneficial result.
Fiscal authorities are required to take the same view. Income
taxation is based on the notion of a surplus of resource inflows over
resource outflows. The relevant outflows (costs) are those that were
 Cost Accounting Principles 169

necessary to generate the inflows (revenues). Again a causal relation

is presumed; if costs have been incurred to achieve future revenues,
they are to be deferred to the tax returns of future periods.
Finally, investors are also likely to be concerned with causal rela¬
tions, although the case for this position is not as clearly demon¬
strable as it is in taxation and contract costing. The rewards to
investors arise mainly from spreads between resource inflows and
the resource outflows necessary to achieve them. The wider the
spread, the greater the incentive to invest, other things being equal.
The causal relation may also be looked at from the other side. The
achievement of the final cost objective may be viewed as the cause
and the use of resources as the result. Either way, the relation
between cost and cost objective is inescapably a relation between
cause and effect. The causality criterion is the only one reflecting
this relationship.

Benefit

Causality is not universally accepted as the only criterion in se¬

lecting product costing principles. Another term that is widely used
is benefit. Most cost accounting textbooks, for example, apply it in
phrases such as, "Costs should be allocated to the cost centers bene¬
fiting from them." The Cost Accounting Standards Board uses the
phrases, "benefiting cost objective" and "beneficial or causal rela¬
tionship" repeatedly in its public pronouncements.9 A benefit crite¬
rion would require that costs be assigned to cost objectives in pro¬
portion to the benefits derived from the costs.
Vatter dismissed this criterion in a single sentence a generation
ago in a paper still worth studying: " 'Benefit/ unless measured in
terms of the use or provision of facilities, is too vague a term to
result in objective division of joint cost among departments or other
costing units."10
The truth of this statement is apparent from the meaning of the
term, benefit. One definition is: "Anything that promotes or en¬
hances well-being; advantage."11 The same dictionary defines the

9. Cost Accounting Standards Board, Statement of Operating Policies and Proce¬

dures, March, 1977, pp. 16-18. This statement is now being revised, but preliminary
drafts of the revised version retain this same language.
10. William J. Vatter, "Limitations of Overhead Allocation," Accounting Review
(April 1945), pp. 163-176, at p. 171. Vatter used the term joint cost to refer to all
nontraceable costs, including the unit prices of interchangeable resources.
11. American Heritage Dictionary, p. 123.
170 Gordon Shillinglaw

related term, beneficial, in the following way: “Promoting a favor¬

able result; enhancing well-being; advantageous. .Unfortunately,
cost objectives are inanimate and therefore don't have well-being.
Furthermore, the “advantage" or “favorable result" that a cost ob¬
jective receives from the use of resources is the present value of the
future cash flows that result from achieving the cost objective.
These cash flows are the joint result of all the resources used.
Adopting a benefit criterion therefore would require costs to be
assigned in proportion to present value, and all cost objectives
would have equal cost-benefit ratios. This is a clearly nonsensical
result.
How, then, is the extensive use of the term, benefit, in cost ac¬
counting to be explained? Part of the explanation is that it is simply
a synonym for a causal relation that is relatively imprecise. A cost
center “benefits" from the cost of providing it with floor space
because that space, and therefore the cost, is indispensable if the
center is to achieve its objectives. The amount of the benefit is
neither measured nor even considered. Since the presence of a de¬
pendency relationship is the basis for a presumption of benefit
received, inserting the term benefit adds nothing to the causality
criterion.

Equity
The final criterion to be considered is equity or fairness. This has
been suggested by Anthony12 as a companion to causality and by
Bedingfield13 as an overriding criterion. Bedingfield reaches his
conclusion by examining the rulings of the Armed Services Board of
Contract Appeals in cases involving the interpretation of benefit.
He concludes, following the decisions in a number of cases, that
benefit is too nebulous a term for practical application. Since the
judges on the board have fallen back on the equity criterion to
resolve issues presented as questions of relative benefit, Beding¬
field concludes that equity is the more basic criterion.
Equity unfortunately suffers from as much vagueness as the ben¬
efit criterion. Equity is defined as “the state, ideal, or quality of

12. Robert N. Anthony, "What Should 'Cost' Mean?" Harvard Business Review
(May-June 1970), pp. 121-131, at p. 127.
13. James P. Bedingfield, "Equity: The Implicit Basis of Accounting," Federal Ac¬
countant (Sept. 1972), pp. 101-9; also Howard W. Wright and James P. Bedingfield,
"Benefit as a Criterion for Indirect Cost Allocation," Federal Accountant (Sept. 1973),
pp. 67-76.
 Cost Accounting Principles 171

being just, impartial, and fair.''14 The problem is that equity is a

highly subjective concept. What appears just and fair to one party
may seem grossly unjust and unfair to another. Courts of law exist
to resolve differences of this sort, and the rulings of these courts
provide practical definitions of what is just and fair in specific
cases. This is exactly what the judges of the Armed Services Board
of Contract Appeals were called upon to do in the cases cited by
Bedingfield.
It is uneconomical, however, to resort to judicial proceedings
every time a costing principle is established or a cost is assigned.
Costing principles are designed to help answer recurring questions.
These questions deal with economic relationships, relationships be¬
tween inputs and outputs. Subjecting these to tests of equity makes
no more sense than using equity tests to choose between two expla¬
nations of the origin of the solar system.
Anthony's argument is different. He reasons that the application
of logical, cause-and-effect tests to cost data will leave a residual of
unallocated costs, because no clear causal relation can be found. The
point is well taken. No costing principle based on the causality
criterion is up to the task of dividing the president's salary among
the final cost objectives of the period. If this cost is to be divided,
some noncausal formula must be used.
The ideal solution to this dilemma is to define it away. No
method of allocating the president's salary will make the cost total
assigned to a cost objective more meaningful or more useful than
any other method. This being so, it is better to regard such alloca¬
tions as outside the realm of cost accounting than to try to develop
criteria for bringing them in.
These allocations may be desirable as part of the pricing process,
or for public justification of actions that management has decided
on (or outsiders wish to impose), but they don't reflect economic
relationships. Cost accounting is intended to measure relationships
between inputs and outputs, and its meaning becomes confused if
it includes costs for which not even the haziest relationship can be
identified. The next section will examine how far this conclusion
can be pushed.

Measurement Principles
Once causality has been accepted as the primary criterion for
product costing, the remaining question is how causality is to be

14. American Heritage Dictionary, p. 443.

172 Gordon Shillinglaw

reflected in the body of cost accounting principles. A principle was

defined earlier as a normative statement of how a concept or con¬
cepts should be applied. The following basic principles should
guide the accountant in deriving the costs of individual final cost
objectives for use in external financial measurement and reporting:

1. Attributability

2. Traceability

3. Interchangeability

4. Interdependence

5. Variability

6. Capacity provided

7. Homogeneity

8. Reasonableness

9. Imputability

The Attributability Principle

The greatest single advance in cost accounting theory in this cen¬
tury was the recognition that cost data for management need not
always be based on estimates of the full cost of individual goods
and services. The phrase “different costs for different purposes"
could be emblazoned on the cover of almost any cost accounting
text now in print.
Some authorities have tried to take this development one step
farther, suggesting that costs for external reporting might be based
on concepts such as variable cost or “relevant" cost. Much ink has
flowed in an effort to resolve the issues these authorities have
raised, mainly the issue of whether costs for external financial re¬
porting should be based on causality in the short run (costs caused
by the use of capacity) or on causality in the long run (costs caused
by the provision and use of capacity). This is not the place to add to
this flow. It is very clear that the use of variable costs to measure
the costs of final cost objectives will not be accepted for external
financial reporting in the foreseeable future. Enough important
issues remain to be discussed without dwelling on this one, which
appears to be moot.
The assumption here, therefore, is that the first costing principle
is the attributability principle: the costs of individual final cost
objectives should reflect the full costs resulting from the provision
and use of capacity to achieve those cost objectives. This calls for
 Cost Accounting Principles 173

the use of the concept of attributability because the causality crite¬

rion will take us no farther.
Attributability is not a widely accepted principle, nor even one
that is widely understood.15 Most authorities would substitute "full
cost" for "attributable cost" in the preceding paragraph. Interest¬
ingly enough, they probably would change nothing else except for
stylistic reasons. The explanation of this apparent contradiction is
that full cost is generally presumed to answer the attributability
question: "What costs would not have been incurred if the orga¬
nization had not provided and used the capacity to achieve this
specific cost objective, other things being equal?"
Even so, some authorities have been willing to move toward attrib¬
utability without admitting to a departure from full cost. Full cost
for costing inventories of manufactured products, for example, is
supposed to include a share of all manufacturing costs necessary to
place the goods in a form and condition suitable for sale. Although
general and administrative costs presumably support both manu¬
facturing and nonmanufacturing activities, they are still to be ex¬
cluded entirely from inventory costs "except for the portion of such
expenses that may be clearly related to production."16 In other
words, the fact that no clear relation exists between characteristics
of individual manufacturing cost objectives is reason enough to
exclude these costs from measures purporting to represent full cost.
This means that if full cost figures are deemed appropriate for
specific purposes, the costing principle must be based on some
criterion or criteria other than causality. The final principle in this
set, imputability, will deal with just that situation.

The Traceability Principle

The second costing principle is traceability: costs traceable to a
cost objective should be included in the costs of that objective.
The argument for this principle is that, in most cases, traceability
is prima facie evidence that a causal relation exists between the cost
objective and the cost. If the cost objective had not been created,
the costs traceable to that objective would not have been incurred.
Because this is such an important principle, three examples will be
useful:

15. For more on the concept of attributable cost, see Gordon Shillinglaw, “The
Concept of Attributable Cost/'Journal of Accounting Research (Spring 1963), pp. 73-85.
16. Professional Accounting Standards, sec. 5121.05 (italics added).
174 Gordon Shillinglaw

Example 1: Materials costs of $2,000 are traceable to a batch of

manufactured products. These costs can clearly be said to have been
caused by the manufacture of that batch of products.
Example 2: Property taxes of $80,000 levied on a factory building
are traceable to that factory's activities, taken as a whole; the taxes
can be avoided by disposing of the building. Ownership of the
building is the immediate cause of the company's obligation to pay
this tax; the factory's activities are the immediate cause of the com¬
pany's ownership of the building; the tax is therefore caused by the
factory's activities.
Example 3: A company issues an electric typewriter to the secre¬
tary of one of its product managers. Depreciation on that type¬
writer, amounting to $200 a year, is traceable to the manager's prod¬
uct line. In the absence of this product line the company would
have needed one less typewriter; the product line therefore caused
the company to incur the depreciation cost.
Two comments should be made on the traceability principle.
First, a cost can be said to be traceable to a cost objective even if it
represents only a part of a single resource. Depreciation on the
typewriter was traceable to the product line even though the
amount of the resource used each year was estimated rather than
measured.
This application of the traceability principle is likely to be the most
controversial. The question is where the line is to be drawn. If a
piece of equipment is used for one day on a particular project, can it
be said that one day's depreciation is traceable to that project? The
answer must be yes, as long as depreciation is assumed to be propor¬
tional to elapsed time. If depreciation is identified with an annual
time period on the basis of that year's proportionate share of the
asset's total lifetime, there is no way of denying its traceability to
segments of this annual period. The only difference is that the inter¬
val on which the calculation of depreciation is based is much shorter
than a year. This argument extends even to the assignment of depre¬
ciation charges to cost objectives on the basis of those objectives'
percentages of the total hourly use of the depreciating assets.
A second point on traceability is that causality is present whether
an activity uses resources already in the organization's possession
or resources purchased in the spot market for carrying out a specific
activity. The materials in example 1 may have been bought and
paid for in previous periods; the typewriter may have been used in
another department before being transferred to the product man¬
ager's office.
 Cost Accounting Principles 175

Management's earlier decision to acquire the resources caused the

resources to be present and usable. This decision was based on the
assumption that a specific use or uses would be found later on, and
on the recognition that spot purchases of resources after a specific
use had been identified would be either unfeasible or uneconomi¬
cal. The cause, in other words, lies in the anticipated activity and
can be identified with the activity that actually materializes. To
deny this relation is to deny causative status to any resource ac¬
quired prior to the initiation of the activity being costed.

The Interdependence Principle

Traceability sometimes has a real shortcoming as a costing princi¬

ple. When two cost objectives are interdependent, the costs traceable
to each may not be a good measure of the underlying causal rela¬
tions. In the example cited to explain this concept earlier, a manufac¬
turer opening new production facilities transferred experienced
workers from existing operations to provide the nucleus of an effi¬
cient work force at the new location. The costs traceable to the activi¬
ties at the new location did not include the added costs of maintain¬
ing the production volume formerly achieved at the old location.
The solution to this problem depends on the level of verification
required. In making decisions, management cannot ignore the indi¬
rect effects of operations at the new location; the revenues must
cover the indirect costs as well as those technically traceable to the
new activity. For external financial reporting, however, a high level
of verification will be necessary. If convincing statistical evidence
can be presented to document the effect on the costs at the old
location, the traceability principle should not be allowed to block
the transfer of costs from the cost objectives at the old location to
those in the new facilities. The interdependence principle should be
followed: when two cost objectives are interdependent, the costs
traceable to one but attributable to the other should be transferred
to the latter if they can be appropriately verified. If adequate verifi¬
cation can't be found, however, the traceability criterion will have
to govern.

The Interchangeability Criterion

Although traceability is prima facie evidence of causality, it is not
conclusive. Interdependence is one cause of exceptions; another is
the presence of interchangeability. A problem will arise if different
interchangeable resources have different costs.
176 Gordon Shillinglaw

For example, building space is often at least partially interchange¬

able. One space may have been leased recently at an annual rental
approximating current market rates; another space may be in a
building the company has owned for several decades. If the two
spaces are interchangeable, then the causality criterion would sug¬
gest measuring each at the same cost.
A strict interpretation of historical causality would require each
space to be measured at the cost of the one most recently acquired.
After all, the relevant question is, what costs would not have been
incurred if the company hadn't provided the capacity to achieve the
cost objective? The implication is that the newly leased space was
acquired because management anticipated the need to meet not
only the cost objectives made possible by using the older space but
additional cost objectives as well. Since each of a group of cost
objectives being achieved simultaneously can be regarded as the
"additional” cost objective, each can be assigned the cost of the
space it takes to achieve additional cost objectives. (The amount of
capacity required in any given period is a function of the total
production load, regardless of the sequence in which orders were
received.)
Although this argument is virtually irrefutable, it fails to satisfy
the articulation constraint, since this approach will either overdis¬
tribute or underdistribute the costs clearly attributable to the need
to provide space for achieving all of the cost objectives combined.
As a result, neither the Financial Accounting Standards Board nor
the Cost Accounting Standards Board, nor the Securities and Ex¬
change Commission, nor the Internal Revenue Service, nor any
other authoritative body is likely to give this approach more than a
disdainful glance.
That being the case, the interchangeability principle has to be
stated in a modified form: if similar interchangeable resources are
available for achieving a cost objective, and if they have different
unit prices, the unit price used in calculating the cost to be assigned
to individual cost objectives should be the average unit price of the
entire group of interchangeable resources. (Notice once again that
nothing in this statement precludes the use of average replacement
price or average anticipated resource price, if the appropriate au¬
thoritative body agrees to depart from historical cost.)17

17. Stating the principle in this way rules out the use of methods such as LIFO and
FIFO, unless the differences are immaterial. Otherwise the sequence in which cost
objectives are achieved will influence their cost.
 Cost Accounting Principles 177

The implementation of this principle will depend on how clear

the interchangeability is and how much verification is required. In
a relatively easy case, two different workers who have identical
skills and identical job classifications but different seniority proba¬
bly should have identical charging rates. Little disagreement about
seniority is likely to arise, and the recorded job ratings would prob¬
ably be acceptable as verification of the interchangeability.
At the other extreme, the interchangeability of 100,000 square feet
of space in a downtown office building with the same amount of
space in a manufacturing plant in the country is not obvious on the
surface. If the purpose of costing is to help management decide
whether to discontinue an activity in one of these locations, a judg¬
mental estimate with little verification is likely to be acceptable. If
the purpose is to measure cost for contract costing, however, judg¬
mental estimates of this sort are likely to be rejected for lack of
adequate verification.

The Variability Principle

Not all of the resources consumed in achieving a particular cost
objective are traceable to it, even in a superficial way. The con¬
sumption of some of these nontraceable resources in individual
periods is likely to be roughly proportional to some characteristic of
individual cost objectives in these periods. For example, the
amount of labor used in moving materials from stockrooms to pro¬
duction locations is likely to be closely related to the amounts of
material required by production.
In cases of this sort, a causal relation can be assumed between the
cost objective and the cost. The third costing principle therefore is
the variability principle: when costs not traceable to a cost objective
vary in total with variations in some characteristic of that cost ob¬
jective, they should be assigned to the cost objective on the basis of
the estimated rate of variability.
The only conceptual problem likely to be encountered in applying
this criterion arises whenever variability is demonstrably nonlinear.
The clearest example of this is the variability of overtime premiums.
Budgeted overtime premiums may be virtually zero through much
of the customary volume range but rise steadily after volume ex¬
ceeds a certain threshold level. A strict application of the variability
principle would require the use of a zero variability rate for over¬
time premiums when operations are below the threshold level; at
higher volumes the rate should be the rate of increase represented
by the slope of the solid line in the diagram below.
178 Gordon Shillinglaw

This solution creates the same problem that arose from the appli¬
cation of the interchangeability criterion. Use of the rate of increase
will overdistribute overtime premiums because it represents a mar¬
ginal cost rather than an average cost. As long as the financial
statements must articulate—that is, a measurement change that af¬
fects a balance sheet figure must have a compensating income state¬
ment effect—overdistribution will be an unacceptable feature of a
cost accounting system for external financial reporting.
The only acceptable alternative in such cases is to use an average
rate of variability spanning the range of volume from zero to the
outer limit of the range encountered—the cash line on the diagram.
The argument to support this approach is that during an extended
period overtime premiums will vary in this proportion rather than
in the nonlinear fashion included in the short-term budget.
Notice that the rate of variability is to be expressed as a rate per
unit of some characteristic of the cost objective to which the costs
are to be assigned. This characteristic can be either an input mea¬
sure (the total number of input units used to achieve the cost objec¬
tive) or an output measure, indicating the amount of production
arising from the achievement of the cost objective.
Notice also that this rule applies whether the cost objective is a
final or an intermediate cost objective. If a production center re¬
ceives service from a service center, for example, the "characteristic
of the cost objective" can be a measure of either the amount of
service provided or the amount of production achieved by the pro¬
duction center during the period. Input, in turn, can be measured
by the number of units of service received from the service center or
by the amounts of resources used by the service center to provide
these services.
 Cost Accounting Principles 179

The Cost Accounting Standards Board, in Standard 403, recog¬

nized all three of these possibilities in establishing a hierarchy of
bases for allocating the costs of home office service functions to
individual segments of the organization.18 These bases were pre¬
scribed for all costs, not just the variable costs, and the terminology
was different, but all three were there, nevertheless, and are worth
describing.
The board's first choice is to use "a measure of the activity of the
organization performing the function." By this it means the number
of labor hours or the amount of materials used by the service center
to provide services to individual segments. This is like using a
system of job order costing to account for the costs of the service
center.19
The second basis in the board's hierarchy is "a measure of the
output of the supporting function," meaning the number of units of
service performed. Each unit of service is presumed to be like every
other, and no effort is made to find out whether it has cost more to
provide one than to provide any other. Here the analogy is to a
process costing system of assigning the costs of the service center to
the segments receiving the services.
The third basis in the board's hierarchy is a "surrogate" of some
sort; generally these are "measures of the activity of the segments
receiving the service." An example would be the number of labor
hours in the segments themselves. If segment labor hours are used
as a surrogate, the presumption is that variations in labor hours are
highly correlated with the needs of the segment (cost objective) for
services.

The Capacity-provided Principle

Many of the costs necessary to achieve a cost objective, but not

traceable to it, are capacity costs. The causal relation between the
cost objective and such costs is more indirect than in the case of
the variable costs. The key phrase, however, is that they are neces¬
sary to the achievement of the cost objective. If they were not
incurred, the cost objective could not be achieved. The various

18. Cost Accounting Standards Board, Standards, Rules and Regulations (Washing¬
ton, D.C.: U.S. Government Printing Office, looseleaf service), pt. 403, para.
403.50(a), p. 132.
19. I am indebted to Nelson Shapiro of the board's staff for pointing out this
analogy.
180 Gordon Shillinglaw

cost objectives supported by the costs of providing capacity are

therefore the causes of those costs. The only question is what
proportion of total capacity cost can be attributed to each of the
supported cost objectives.
The answer to this question is embodied in the capacity-provided
principle: the costs of providing capacity should be assigned to the
cost objectives relying on that capacity in proportion to their rela¬
tive occupancy of the capacity when it is used fully. In other words,
if a cost objective uses capacity only at off-peak periods, it would
not be charged a portion of capacity costs because it plays no role in
causing the organization to incur any of its capacity costs.
This principle may not be strictly applied in practice, for any or
all of three reasons. First, peak capacity occupancy may be too diffi¬
cult to measure. The measurability constraint requires the use of
average capacity occupancy in such cases.
Second, the principle of interchangeability may intervene, as it
does in the case of some apparently traceable costs. If the cost objec¬
tives that are achieved at peak periods are interchangeable with
those met at other periods, then causality has to be assigned equally
to them all.
Third, the capacity-provided criterion may not be applied strictly
because those doing the measuring believe it would lead to divid¬
ing costs "unfairly" among the various cost objectives. Instead, ca¬
pacity costs are spread on the basis of the average use of capacity,
rather than usage in the peak period.
This is not a legitimate reason for modifying the costing principle
in such a way. The argument has already been made that fairness is
either an ethical or a legal concept, and is not applicable to ques¬
tions of economic consequences when adequate measures of causal¬
ity are available. Equity may even be a political concept and thus
even farther from the arena of objective economic measurement.
When electrical power rates in the Northeast first began to rise
sharply, people whose houses and apartments were heated by elec¬
tricity demonstrated so loudly and persistently that the appointed
regulatory officials assigned lower power rates to these users. The
result, of course, was that other users had to pay a larger share of
the costs of providing power capacity than their own capacity re¬
quirements justified. Because the users in this latter group were
more numerous, the effects of the cost redistribution were spread
over a much wider base and the cause-and-effect relation was not
readily apparent. The political reaction of these users was therefore
directed not at those who benefited from subsidized heating costs
 Cost Accounting Principles 181

but at the average rates charged by the power-generating compa¬

nies and at the regulatory commissions which approved those rates.
The main defect of the capacity-provided principle is its pre¬
sumption that the long-run cost function is linear. If expansion or
contraction of capacity will lead to a change in total capacity cost
that is either greater or less than the proportionate change in capac¬
ity, then the average cost of a unit of capacity will be an inaccurate
measure of the causal relation between capacity costs and the cost
objective.
One sees here another example of the problems in applying the
interchangeability and variability principles. Articulated accounting
systems can't deal with nonlinear relationships. As long as articula¬
tion is a requirement, cost figures that are imperfect reflections of
the causality criterion must be accepted.

The Homogeneity Principle

Cost objectives in any organization can be arranged logically in
tiers, according to their distance from the final cost objectives and
their degree of dependency on each other. The task of cost account¬
ing may be to measure the cost attributable to cost objectives at any
level. For external financial reporting, however, it is the costs of
final cost objectives that matter.
Many costs are not traceable to final cost objectives, of course, but
are needed to achieve those objectives. To assign these costs to final
cost objectives, the accountant must use costing rates based on the
variability and capacity-provided principles. Each costing rate ap¬
plies to a particular group or pool of costs, and each pool consists of
the costs of achieving an intermediate cost objective.
In structuring these pools, the cost accountant should follow the
homogeneity principle: costs that are not traceable to individual
cost objectives at a given level but are necessary to achieve those
cost objectives should be grouped in pools, each of which is homo¬
geneous. If the costs in a pool are homogeneous, a single variability
rate and a single capacity rate will be appropriate. The application
of the homogeneity principle usually means that costs will be as¬
signed in stages, with many costs passing through two or more
intermediate cost objectives before reaching the final cost objective.

The Reasonableness Principle

One final implication of the causality criterion is that the ac¬
countant can search for causes not only in what an organization
182 Gordon Shillinglaw

accomplishes but also in what it doesn't accomplish. One set of

causes of costs consists of events that lead to wast.e. Waste can be
regarded as a form of cost objective. (Remember that the term cost
objective was defined in terms of the results of the use of resources,
and waste is a result in many cases.) From this conclusion one may
arrive at a reasonableness principle: the costs of wasted resources
should be excluded from the costs of productive final cost objectives.
Application of this principle requires judgment. It also requires a
probabilistic approach to costing; a provision for "normal" waste
because of factors inherent in a process should be included in the
costs of productive final cost objectives because these costs are "rea¬
sonably necessary" to achieve those objectives. Abnormal and sub¬
normal wastage on any particular final cost objective should be
excluded from the calculation of the cost of that cost objective.

The Imputability Principle

joint production is the only situation in which the causality crite¬
rion fails to provide a clear conceptual guide to cost assignment.
Joint production is undertaken to obtain all of the joint products as
a group; causality applies to the group as a whole, not to any one
joint product.
The obvious solution is to recognize only one cost objective—the
entire group of joint products. This will not be enough, however, if
the joint products are inventoried in substantially different propor¬
tions or if different joint products are elements of different final cost
objectives, at least one of which is a cost-based pricing contract.
The solution in the first of these two situations is to apply the
imputability principle: the cost of each joint cost objective should
be measured by the imputed opportunity cost of achieving this
objective independently by other means, usually by open market
purchases.
This principle has three weaknesses. First, it can't be applied in
contract costing because the values of joint products used to meet
final cost objectives embodied in cost-based pricing contracts are
unknown. Second, it fails the articulation test in that its application
will either underdistribute or overdistribute the total joint cost.
Third, data on opportunity costs may be unavailable because no
outside markets exist.
The second and third of these weaknesses can be overcome quite
simply, either by reducing all of the opportunity cost figures pro¬
portionally to bring the total down to the total of the joint costs, or
 Cost Accounting Principles 183

by using some variant of the net realization method. Both of these

solutions finally bring the benefit criterion into the system, benefit
being measured by the value of the joint products. As such, they
are expedients and the resulting cost figures will lack economic
meaning.
The first weakness of the imputability criterion is even more in¬
tractable. If cost is a determinant of value, value can't be used as a
determinant of cost. Here at last may be a place to apply the equity
criterion—that is, the principle should be that joint cost will be
assigned in contract costing by a method negotiated by the parties
in advance and therefore accepted by both as “just, impartial, and
fair."
The Cost Accounting Standards Board or other authoritative body
could shortcut this process by establishing a standard to govern
situations of this sort. Any such standard would be arbitrary, sub¬
stituting the board's definition of equity for the specific definitions
that would emerge from bargaining between the contracting par¬
ties. This may not be too high a price to pay, however, to avoid
lengthy, contentious bargaining on an issue with no “true" answer.
The exceptions to this final principle should serve to demonstrate
that cost allocations of this kind, although probably an unavoidable
part of the costing structure in some situations, have no firm roots
in the theoretical foundations of cost accounting. They must of ne¬
cessity rest on concepts drawn from other disciplines or, preferably,
be left to the affected parties to resolve for themselves. Only when
this is impractical, as in taxation or financial reporting to investors,
should a standard solution be developed.

May 1977
Social Performance Accounting
An Unheeded Challenge to the Accounting
Profession

R. Lee Brummet

Never has there been such a demand for individual and organiza¬
tional accountability. Never has there been such a felt need by such
large numbers of people for so much information. Never has there
been such emphasis on integrity of information and credibility of
sources. Never has the accounting profession enjoyed such visibil¬
ity and notoriety, and never have the services of the accounting
profession been in such demand by a society so willing to pay for
them.
The challenge and the potential for the accounting profession are
unmistakably greater than at any previous time in the history of the
profession. Yet the response is uncertain, sometimes confused, usu¬
ally defensive, and in my opinion far short of the mark to assure a
bright future for the profession.
My purposes in this paper are twofold. The first is to consider the
challenge to the accounting profession provided by demands for
information on social performance and the profession's reluctance
to be responsive to that challenge. The second is to encourage mem¬
bers of the profession to think more broadly of their role in the
present environment—an environment which reflects a montage of
social and economic cross-currents. More specifically, my purpose
is to encourage members of the accounting profession to welcome
the challenge by becoming active if not aggressive participants in
developing and improving a technology for the measurement, com¬
munication, and attestation of social performance of organizations.

185
186 R. Lee Brummet

The Growing Need for a New Technology

The period from 1920 to 1960 in the United States was marked by
rapid business formation, capital accumulation, and industrial de¬
velopment along with relative blindness to social costs. During this
same period we saw significant advances in the state of the art of
accounting and its important theoretical underpinnings. In this set¬
ting, economic fulfillment and financial strength were perceived as
primary goals of individuals, organizations, and society in general.
Great faith was shown in the "invisible hand" described by Adam
Smith. Profit maximization with minimal social constraints was the
way of the business world. In this setting, further, the accounting
profession provided the concepts and the technology for profit mea¬
surement and the assessment of the financial strength of private
sector enterprises. Economists, with only minor assists from ac¬
countants, provided the technology for compiling data on gross
national product and related macrostatistical data for viewing the
"progress of society." Financial statements prepared in accord with
generally accepted accounting principles served the private sector
well; and, although perhaps not so clearly, the macroeconomic sta¬
tistics or "social accounting" figures of the late forties and fifties
served the public sector well in shaping economic policy emanating
largely from our federal government.
While we may find indications of a concern for the social impact
of organizations throughout history, the past two decades have
multiplied and amplified such concerns. Well-organized activist
groups, government agencies, and individual citizens are now ex¬
pressing their concerns loudly and clearly.
Private sector enterprises were once viewed as producers and
distributors of economic goods and services to their customers and
as producers of financial rewards for their owners. Now such enter¬
prises are viewed as a part of a social infrastructure with broad
constituencies that include not only customers and owners, but
creditors, employees, and inhabitants in any area of the world
where the enterprise may exert an impact through its operations or
through the use of its goods or services.
In 1971 the Committee on Economic Development published its
much-read and much-discussed report of its Research and Policy
Committee, entitled "Social Performance of Business Corpora¬
tions." This committee comments,

The mounting public demands for better social performance necessitate

goal setting and performance measurement . . . There is little in the
 Social Performance Accounting 187

present accounting and reporting systems of corporations that enables

anyone to determine whether corporations have well-formulated sets of
goals for social performance or to measure the extent of progress toward
realization of these goals.1

Several years earlier we find the well-known social scientist, Abra¬

ham Maslow, saying:

I am impressed again with the necessity, however difficult the job may
be, of working out some kind of moral or ethical accounting scheme. . . .
I'm going to dictate sometime soon my thoughts on how stupid our
present accounting systems are, because they leave out practically all the
important personal, psychological, political, educational intangibles. . . .
Partly it must be put up to the accountants to try to figure out some way
of turning into balance-sheet terms the intangible personal values that
come from improving the personality level of the workers, making them
more cooperative, better workers, less destructive, etc. It does cost
money to hire this kind of personnel; it costs money to train and teach
them and to build them into a good team, and there are all sorts of other
costs involved in making the enterprise attractive to this kind of worker
and this kind of engineer, etc. All these real expenditures of money and
effort ought somehow to be translated into accounting terms, so that the
greater value of the enterprise that contributes to the improvement of the
whole society can somehow be put on the balance sheets. We all know
that such a company, for instance, is a better credit risk, and lending
banks will take this into account. So will investors. The only ones who
don't take these things into account are the accountants.2

Further, and again from outside the immediate ranks of accoun¬

tants, we hear from Carl Madden, Chief Economist of the Chamber
of Commerce of the United States, and a most perceptive scholar and
observer of economic, cultural, and social dynamics. Some of the
changes in management style that he has listed as likely include:

1. Profit [continuing to be] dominant but modified by the assumption

that a business manager has other social responsibilities
2. The application of both an economic and social measure of perfor¬
mance
3. Change from a business system which has little concern for social
costs of production to a business system with increasing concern for
internalizing social costs of production, both here and around the
world

1. Committee for Economic Development, Research and Policy Committee, Social

Responsibilities of Business Corporations (New York: Committee for Economic Develop¬
ment, 1971), p. 48.
2. Abraham Maslow, Eupsychian Management (Homewood, Ill.: Richard D. Irwin,
1965), pp. 49, 60, and 80.
188 R. Lee Brummet

4. The substitution for financial accounting of both financial and human

resources and social accounting3

This perception of a changing role for the business enterprise in

our society has not been limited to those outside the real world of
business. In 1973, C. Peter McColough, president of Xerox Corpora¬
tion, was quoted in News Front as saying, "We do not see Xerox as
solely a profit-making institution. I don't want to minimize prof¬
its .. . but I must emphasize that we regard Xerox as a social insti¬
tution as well as a business institution."4 This is indeed not at all
an unusual statement.
Many private sector enterprises are responding, and in some in¬
stances they are taking leadership roles. While financial performance
remains critical, many companies are planning, executing, and as¬
sessing their operations within a broader framework of responsibil¬
ity to society. Some are well down the road to an emphasis on contri¬
bution to human welfare rather than simply profit maximization.
As our business leaders increasingly recognize this changed role
of private sector enterprises and consider social performance as a
clear objective, the need for information surely becomes obvious.
Accountants could assist in articulating objectives, analyzing cost-
benefit relationships, and providing information on performance to
help monitor progress toward social objectives. Yet the accounting
profession has not responded to any significant extent.

A Minimal Response from the Accounting Profession

While I believe that I am justified in charging the accounting
profession with a lack of responsiveness, it would be unfair not to
recognize some commendable responses of certain individuals and
groups within the profession. In 1972, when Walter Oliphant was
its president and Leonard Savoie was its executive vice-president,
the American Institute of Certified Public Accountants (AICPA)
sponsored a conference which brought together businessmen, po¬
litical scientists, economists, sociologists, government officials, and
accountants to discuss the broad areas of social measurement. In his
opening remarks at this conference, Oliphant referred to the insti¬
tute's description of the professional practice of accounting as "a
discipline which provides financial and other information essential

3. Carl H. Madden, “Changing Roles of Government and Industry/' Evaluating

Government Performance: Changes and Challenges for GAO (Washington, D.C.: Gen¬
eral Accounting Office, 1975), p. 136.
4. News Front, Jan.-Feb., 1973.
 Social Performance Accounting 189

to the efficient conduct and evaluation of the activities of any or¬

ganization."-"' Also at this meeting, Arthur Toan pointed out that

the accountant's interest in social performance measurement evolves

quite logically from the knowledge and experience he has accumulated
over centuries of producing relevant financial information in an organi¬
zational setting for a variety of uses and users. . . . As soon as it became
apparent that financial data would be understandable and more useful
when coupled with non-financial information, the accountant was asked
to obtain that information also and to integrate it into financial re¬
ports. . . . The trend toward blending both kinds of data continues; in
fact, it continues at an accelerated pace.5 6

Within a few months after this conference President Oliphant asked

Arthur Toan to serve as chairman of a task force on measurement of
corporate social performance for the AICPA. Other accountants who
participated in the conference agreed to serve on this task force. After
some three years of study and deliberation, this group authored a
book. The Measurement of Corporate Social Performance, which has
now been published by the Institute. This work, while admittedly
only a preliminary coverage, represents a demonstrable indication of
the profession's interest in the area. In addition, a subcommittee of
the Management Advisory Services division of the AICPA is active
in the study of accounting professional services to be provided in
connection with ecology and the physical environment.
In 1973, Robert Beyer, then the incoming president of the Na¬
tional Association of Accountants, published his paper, "Pilots of
Social Progress," stating that "because of our [accountants'] unique
relationship with business and the public, we must become the
pilots of social progress, charting the course for business and gov¬
ernment."7 He then named a Committee on Corporate Social Per¬
formance Measurement for the NAA with I. Wayne Keller as its
chairman. During the following years this committee published an
outline for study and research in the area and sponsored and super¬
vised several research projects whose results were reported in NAA
publications.
During the past five or six years the American Accounting Asso¬
ciation has had several committees working on nonfinancial mea¬
sures, environmental pollution, and social cost measures; and most

5. Walter Oliphant in Social Measurement (New York: American Institute of Certi¬

fied Public Accountants, 1972), p. 2.
6. Arthur Toan in Social Measurement, p. 41.
7. Robert Beyer, "Pilots of Social Progress," Management Accounting, LIV (July,
1972), 11.
190 R. Lee Brummet

recently this association sponsored a project and published a report

prepared by Professor Robert Jensen with the intriguing title of
"Phantasmagoric Accounting."
Several accounting academicians have published materials on so¬
cial measurement and accounting. Three of these noteworthy au¬
thors include Ralph Estes and Lee Seidler of Wichita State Univer¬
sity and New York University, respectively, and Trevor Gambling
of the University of Birmingham in England.
One of the more highly regarded publications with relevance to
social accounting is the report of the Committee on Objectives of
Financial Statements, chaired by the late Robert Trueblood. This
committee included the following statement in its list of twelve
items:

An objective of financial statements is to report on those activities of the

enterprise affecting society which can be determined and described or
measured and which are important to the role of the enterprise in its
social environment.8

Thus there are numerous indications of individuals and groups of

individuals who favor the profession's participation in activities
that measure social performance. Yet real involvement of the profes¬
sion continues to be limited mainly to cost-benefit or cost-effective¬
ness analysis for public sector organizations and for some nonprofit
entities, such as hospitals and educational institutions; and these
services are provided by only a very small portion of the accounting
profession.9
Even though the accounting profession is not greatly involved,
many organizations are doing some social accounting. The most
usual approach is to inventory those activities of the corporation
that are socially motivated. This is at least a reasonable start, utiliz¬
ing a process not totally unfamiliar to accountants.
Each year more corporations reflect their sensitivity to public in¬
terests by increasing the disclosure of their social performance in

8. Robert M. Trueblood in Objectives of Financial Statements, Report of the Study

Group on the Objectives of Financial Statements (New York: American Institute of
Certified Public Accountants, 1973), p. 66.
9. The General Accounting Office (GAO), which serves the Congress of the United
States, has undoubtedly done more than any other group toward developing technol¬
ogy for measuring social performance, since such performance is usually the goal of
government programs sponsored by our congressional bodies. At the conference on
social measurement in 1972, referred to earlier in this article, Stuart McElyea of the
GAO said, "It seems to me somewhat unrealistic to be discussing here whether
social programs are to be measured. I think [such] measurement is inevitable."
 Social Performance Accounting 191

annual reports. Dennis Beresford reports that 425 of the Fortune 500
companies made social responsibility disclosures in their 1975 an¬
nual reports, whereas only 239 did so in 1971.10

An Unjustified Reluctance of the Profession

A most compelling question in the minds of many accountants, as

they consider their role in social measurement, relates to their belief
in the free enterprise system and their uncertainty about the role of
the private sector in our society. Should corporations in a "free
society" accept conscious and substantive responsibility for social
welfare, or should they rely on the "invisible hand" and strive
within the limits of the law to maximize profits? If the answer to
this question is the latter, then the reluctance of the profession is to
some extent understandable. If the answer is the former, then the
profession may well be derelict in its hesitancy.

Monolithic Accounting and the Objective of Business

Neatness of structure and orderliness have been ingrained in the
business establishment and in accountants. This has caused a com¬
mon bias among business managers and accountants in favor of the
great simplicity represented by the single objective of profitability.
The accountant feels a kinship with the business manager who
strives to bring about satisfying profits, and the business manager
feels a like kinship with the accountant who is expert in applying
the measuring rod of success or failure in terms of this single di¬
mension. Each contributes to the other's myopia concerning long-
range perpetuation of the organization in a changing society by
resisting a change to a more complicated system of multiple objec¬
tives and multiple areas for performance measurement.
Breaking out of these intersupportive attitudes is a difficult chal¬
lenge—particularly when such views represent a school of thought
led by Milton Friedman, the noted economist and recipient of a
Nobel prize in 1976, who wrote in the New York Times in 1970,
"There is only one social responsibility of business—to use re¬
sources and engage in activities designed to increase its profits."11
Other similar statements of Professor Friedman could be noted. Yet

10. Dennis Beresford, Social Responsibility Disclosures—1976 Survey of Fortune 500

Annual Reports (Cleveland, Ohio: Ernst & Ernst, 1976), p. 4.
11. Milton Friedman, “The Social Responsibility of Business Is to Increase Prof¬
its," New York Times Magazine, Sept. 13, 1970.
192 R. Lee Brummet

Time magazine reported in 1975 that seven other Nobel prizewin¬

ners, including Gunnar Myrdal and Kenneth J. Arrow, signed a
declaration condemning Western capitalism for bringing on a crisis
by producing "primarily for corporate profits."12 So we see the
paradox of conflicting views on this most important issue of the role
of business.
Yet I believe the direction of change in the value structures within
society, the broadening of the public's expectations of how enter¬
prises perform, and the increasing demands reflected in governmen¬
tal legislation are clear, and that although the trend may vary in
intensity from time to time, we may expect it to be unrelenting. Far
from sounding the death knell of free enterprise—a possibility that
some have bemoaned—the response of business enterprises to the
many and varied pressures from the many different public interests
is itself critical to the survival of that free enterprise system. Alfred C.
Neal, president of the Committee on Economic Development, writes:

The larger companies of necessity must be more concerned about their

integration with the best of society, especially with those groups in society
that constitute their most important constituencies. The view that busi¬
ness firms should concentrate only on their own profitability—behaving
like mindless atoms reacting to the economic environment but unrespon¬
sive to the social environment—is, I hope, relegated to the archives, and to
a few warlock economists who conjure up a world that never was. A few
hundred large companies which account for the bulk of the nation's out¬
put largely determine the level of the economy and greatly influence the
quality of our lives. Such companies cannot divorce themselves from their
environment—be it physical, social, economic, or international.13

All of this does not mean that the profit objective is passe. It does
mean that other objectives are also critical to an enterprise's success
and survival. It does mean that a visible hand can and should sup¬
plement the invisible one. To the accountant it should mean that
corporations need assistance in fulfilling a complex of objectives
rather than only financial accounting assistance in monitoring profit
performance.

Stewardship and Accountability

Stewardship and accountability relationships between people and

organizations are among the most critical aspects of our society.

12. "Can Capitalism Survive?" Time, July 14, 1975, p. 50.

13. Alfred C. Neal, foreword to J. J. Corson and G. A. Steiner, Measuring Business's
Social Performance: The Corporate Social Audit (New York: Committee for Economic
Development, 1974).
 Social Performance Accounting 193

There is strong support for the contention that the fundamental role
of accounting is to facilitate this accountability function. Yuji Ijiri
states. Accounting . . . starts with the recording and reporting of
activities and their consequences and ends with the discharging of
accountability. . . . We may . . . say that accountability is what dis-
tinguishes accounting from other information systems in an organi¬
zation or in a society."14 This view is reflected significantly in early
writings of William A. Paton and particularly in the 1940 Paton and
Littleton monograph, which emphasizes the "need for developing
standards of accounting to secure proper protection for the equities
of corporate investors."15
Our economic, social, and political setting has changed signifi¬
cantly during the past forty years. There has clearly been a shift of
emphasis to "performance" accountability in addition to asset stew¬
ardship, and this emphasis has extended to a number of objectives,
including employee and community welfare, quality of products
and services, and other social performance goals as well as profits
for equity holders. Further, the Paton and Littleton monograph re¬
flected primary concern for the relation between managers and
owners. This continues today to be an important relationship, but
we now must also consider relations between the organization and
many other interested and involved constituencies in our society.
Rosita Chen concludes succinctly from her study of the stewardship
function of accounting that "management's performance should be
evaluated in terms of both profit and the accomplishment of social
objectives. The latter aspect has long been neglected by the account¬
ing profession. It is the responsibility of the accountant, therefore,
to measure, report, and audit management's social performance."16
With regard to possible dereliction, the profession may some¬
times be naive about the impact of what it does or does not do.
While recognizing that the accounting profession has made a great
contribution to social and economic welfare, Norton Bedford states:

Accounting is closely related to [some] social problems and opportunities

and to an extent may be partly responsible for them. That is, performing
a quantitative information function, accountants have measured and dis¬
closed the private costs of consumer-selected choices and have ignored or

14. Yuji Ijiri, Theory of Accounting Measurement, Studies in Accounting Research

#10 (Sarasota, Fla.: American Accounting Association, 1975), p. 32.
15. William A. Paton and A. C. Littleton, An Introduction to Corporate Accounting
Standards (Chicago: American Accounting Association, 1940), p. 2.
16. Rosita Chen, "Social and Financial Stewardship," Accounting Review, L (July,
1975), 542.
194 R. Lee Brummet

discounted certain overall elements of social costs and benefits. As a

result, accountants must bear some responsibility both>for opportunities
not realized and for any loss of dignity of human life that is due to
overall misdirected economic activity or to the over-crowding, pollution,
transportation failures, crime, and inadequate housing and education
that may have resulted. It is becoming increasingly apparent that with
changes in society, the conventional accounting disclosures cover a
smaller proportion of those activities that affect the quality of human
life. ... An appropriate adjustment in the scope of accounting disclo¬
sures may now be the need of the land.17

Accountants are probably well advised to take part in economic,

political, and social movements only as good and intelligent citizens
and not as accounting professionals. Yet their work plays an impor¬
tant role in success or failure in these areas. We, as accountants,
must view—with all the objectivity and intelligent perception we
can muster—the changes taking place in our economic and social
environment, and we must respond in the way that will contribute
the most.

A Matter of Definition
Some of the reluctance of the profession to expand into new areas
of social measurement and disclosure may relate to our self-im¬
posed definition of accounting. We have repeatedly defined and
explained our profession and its aims. For example:

Accounting is the art of recording, classifying, and summarizing in a

significant manner and in terms of money, transactions and events
which are, at least in part, of a financial character, and interpreting the
results thereof. (Terminology bulletins of the 1930s.)

The purpose of accounting is to furnish financial data concerning a

business enterprise, compiled and presented to meet the needs of man¬
agement, investors, and the public. (Paton and Littleton, 1940.)

Accounting is the process of identifying, measuring, and communicat¬

ing economic information to permit informed judgments and decisions
by users of the information. (A Statement of Basic Accounting Theory,
American Accounting Association, 1966.)

Accounting is a service activity. Its function is to provide quantitative

information, primarily financial in nature, about economic entities that is
intended to be useful in the making of reasoned choices among alterna¬
tive courses of action. (Accounting Principles Board Opinion No. 4,
1970.)

17. Norton Bedford, Extensions in Accounting Disclosure (Englewood Cliffs, N.J.:

Prentice-Hall, Inc., 1973), pp. 74, 75.
 Social Performance Accounting 195

It is noteworthy that all these efforts to articulate the fundamental

nature of accounting use such constraining words as economic, fi¬
nancial, and quantifiable; they reflect, it seems to me, a tendency to
be conditioned by the boundaries of current practice.
Now definitions are of some importance. In this instance they
enable accountants to perceive their role and to concentrate on de¬
veloping a technical expertise within an identifiable body of
knowledge. They provide some feelings of security. Further, they
may help the users of accountants' services to know what to expect
and thus to avail themselves of the profession's well-known stock in
trade. Yet definitions of the societal role of the profession should
not be only a reflection of the basic techniques—techniques that
may for some period be fundamental to the services performed, but
may outlive much of their usefulness. If what we do conditions our
definition and our definition conditions what we do, our ability to
adjust—to be imaginative and creative—is seriously limited.
Accounting should be defined in terms of its role in society. If the
needs of society change, calling for new techniques to enable ac¬
counting to continue to fulfill its role, then the profession and its
activities can and should change within the framework of the defi¬
nition. I suggest that for the good of the accounting profession and
for the good of us all, we adopt the following definition:
Accounting is a systematic symbolic process which enables individuals,
organizations, and society to articulate objectives, to provide projections
and feedback information that assists in monitoring efforts to accomplish
such objectives, and to provide reliable information regarding perfor¬
mance and condition accountability to legitimate constituencies.

This definition captures a most critical and eminent domain for the
profession. It does not include constraints implied by adjectives
such as economic, financial, or quantifiable, which may encourage
or at least allow the use of tenacious caveats by accountants—ca¬
veats that may cause the profession to ignore the changing needs of
society and relegate itself to performing only the more menial kinds
of bookkeeping and reporting of less and less important data. It
challenges members and prospective members of the profession to
be creative in developing concepts and techniques which best serve
this proclaimed role. The accounting profession, as currently prac¬
ticed and perpetuated by our educational establishment, carries out
this role in a very minimal way, and it will continue to serve only
minimally or even come to serve less than minimally if the self-
imposed constraints of the more usual definitions of accounting are
not abandoned.
196 R. Lee Brum met

The proposed definition is general enough also to include na¬

tional economic and social accounts and economic and social indi¬
cators—areas of concern that have not enjoyed adequate attention
from accountants. The techniques and problems of measuring the
success or performance of a country are of course somewhat differ¬
ent from those related to corporate organizations, but the role of
objective articulation, performance monitoring, and accountability
to constituencies is common to these settings, and accountants' con¬
cern for relevance, objectivity, and freedom from bias applies
equally well.

Exchange Transactions and the Bookkeeping Equation

Another reason why the accounting profession is reluctant to re¬
spond to today's challenge relates to the assumption that its tech¬
niques must be limited to exchange transactions and "events” that
are amenable to the use of the bookkeeping equation and entry into
financial statements which the bookkeeping equation supports. Ac¬
countants are quick to point out the differences between bookkeep¬
ing and accounting, yet they allow the duality of the impact of
exchange transactions and the neat fit of the entity-oriented book¬
keeping equation to establish the limits of "financial reports," and
virtually the limits of the role of accounting in society. This self-
imposed constraint of the profession was not seriously objection¬
able a half-century ago, when there was not the perceived need to
consider social costs, externalities, and enterprises' responsibility to
broad constituencies. It is objectionable now.

Over forty years ago J.M. Clark asked:

Does anyone really accept the scale of the competitive market? We shall
have found such a man, when we have found one who honestly ap¬
proves of existing conditions of poverty: not one who is resigned to them
as an inevitable evil, but one who would not lessen them if he could. So
long as this does not represent the prevailing sentiment, so long it will
be impossible to say that market value measures "social value" in the
sense of "value to society."18

Clark was writing as an economist four decades ago, but econo¬

mists of today recognize the wisdom of his words. Such words also
have special relevance to the social accountant of today. Executives
of modern corporations must think in terms of generalized social

18. J.M. Clark, A Preface to Social Economics; Seidler and Seidler, Social Accounting,
(Los Angeles: Melville Publishing Co., 1975), p. 33.
 Social Performance Accounting 197

values as well as economic values to limited constituencies. Ac¬

countants of these modern corporations and accountants in the
public arena could provide important assistance.
Accountants do, of course, deal with value concepts, but their
techniques apply almost exclusively to economic value concepts as
they relate to individual entities and as they are validated in the
market place. Although the profession appears to sense great
changes that it may champion in connection with "fair value" ac¬
counting, its self-imposed constraints continue to establish bounda¬
ries that seem almost inviolate. Referring to the possibility of ac¬
countants' stretching their domain of activity, Daniel Gray ex¬
presses first pessimism and then optimism as he writes:
Because conventional accounting has been so useful to business, it is
only natural to try to stretch it to cover social costs and benefits. . . . We
have concluded that traditional corporation accounting is a derivative of,
and is coterminous with, the theory of market economy. As such, it is
concerned with transactions. Yet the social problems that disturb us most
lie outside the realm of transactions. To some extent, they have become
problems because they had to be excluded for our theory of a market
economy to achieve closure. Accounting has the same kind of closure. It
owes its rigor to what it excludes. To try to stretch it to measure all social
costs and benefits is to violate its very foundations. Isn't it ironic that a
theory that has encouraged us to neglect "externalities" is now embar¬
rassed by its inability to measure "external" distress? It looks to us as
though it is not "stretching" but innovation that is required if we are to
have social accounting. Accounting has renovated itself in the past—
most recently in the areas of national-income accounting and manage¬
ment accounting—and must now do so again.19

I suggest that our profession can maintain its rigor and enjoy the
closure provided by our conventional approaches relating to the
economics of our market system. But we can and should do more.
We can develop new approaches to the measurement and commu¬
nication of social welfare efforts and impacts. We can deal in a
broader arena of performance assessment while continuing to im¬
prove our ability to perform our conventional role.

Credentials of the Accounting Profession

Since I am concerned both for the need for better information on

social costs benefits and for the future of the accounting profession,

19. Daniel Gray, "Methodology: One Approach to the Corporate Social Audit," in
Dow Votaw and S. Prakash Sethi, The Corporate Dilemma: Traditional Values Versus
Contemporary Problems (Englewood Cliffs, N.J.: Prentice-Hall, Inc., 1973), pp. 93-94.
198 R. Lee Brummet

I am naturally led to a conclusion which favors significant involve¬

ment of accountants. Some may ask and some have asked, "Why
accountants?"
The conventional roles of accountants and the thrust of their
training have emphasized (1) the design and installation of infor¬
mation detection and measurement systems, (2) the administration
and operation of such systems in an organization, (3) the communi¬
cation and interpretation of information systems outputs, and (4)
the verification and perhaps attestation of information provided by
such systems. The design of information detection and measure¬
ment systems in the social accounting area must rely substantially
on the expertise of social scientists other than accountants. Surely,
psychologists and sociologists must be involved, but involvement
or dependence on outside expertise is of course not at all new for
accountants.
When it comes to the administration and operation of the detec¬
tion and measurement system and communication of results to ap¬
propriate decision points in organizations or in society, accountants
have much to offer. In most instances accountants have "the ear of
management," and increasingly they are becoming oriented toward
a wide-ranging set of users. They have the ability to speak the
language of users and thus affect decisions—the obvious goal of
social accounting.
In the area of attestation of information regarding social perfor¬
mance, accountants should play the primary role, although the de¬
velopment of measurement technology must precede the develop¬
ment of verification, audit, or attestation technology. The inherent
uncertainty related to social performance information may necessar¬
ily prevent the development of audit techniques to justify our usual
unqualified opinion, but some degree of assurance may be pro¬
vided short of attestation. The Committee on Social Measurement
has suggested the concept of a "Redsa," an acronym for "review to
develop a suitable appraisal."20
Still another reason for accountants to be more involved in mea¬
suring social performance relates to the profession's primary com¬
mitment to independence and objectivity and the old saw that "if
accountants don't do it some other group will." It is clear that some
other groups are already involved. For the most part, these groups

20. American Institute of Certified Public Accountants, Committee on Social Mea¬

surement, The Measurement of Corporate Social Performance (New York: American
Institute of Certified Public Accountants, 1977), p. 255.
 Social Performance Accounting 199

such as the Council on Economic Priorities and the consumer advo¬

cates come into the field with a motivational bias that assumes or
implies varying degrees of social irresponsibility in corporations.
Clearly such a bias is contrary to the primary roles of objectivity
and independence for important experts. Social activism and social
accounting expertise do not mix well. The professional hallmarks of
objectivity and freedom from bias in the information function
clearly favor the involvement of accountants. Accountants, more
than any other obvious group, can enter the field because their
expertise is needed, and not because of any perception that corpo¬
rations are more or less socially responsible than they should be.
We should recognize some of the potential pitfalls of the profes¬
sion's involvement in social measurement, especially since this new
area for accountants relates so directly to human welfare. Mildred
Francis has reflected her concerns for our profession's activities in
assessing performance of social programs—one aspect of social ac¬
counting—charging their lack of expertise, particularly in experi¬
mental design and statistical sampling theory. She points out:

(1) Since the starting point in the development of accounting theory was
a set of axioms or postulates, the systems he [the accountant] developed
were classification systems which rearranged the facts already contained
in the axioms; and (2) whatever system of checks and balances he devel¬
oped to validate these rearrangements had, as its rationale for existence,
the protection of moneys against fraudulent use. . . . Maybe the time has
come for society to consider establishing an independent body to put its
stamp of approval on statistics and statistical series of public importance.
But nothing can be gained by delegating this responsibility to inexperi¬
enced people [accountants]—people who are not familiar with statistical
theory and its problems, the data needs, and the data collection problems
in the functional area of concern.21

These concerns are appropriate and perhaps understandable for a

statistician with a very narrow view of accounting. And indeed the
narrow view is not far from the mark if we consider the history of
our profession up to the most recent decade. However the profes¬
sion itself and the expertise of its members are changing. Axioms
and postulates have been useful in implementing the double entry
accounting model and "rearranging facts." But these axioms and
postulates do not and certainly need not provide boundaries for the
accountant's role. Accountants can work effectively—and some are
doing so—using both inductive and deductive approaches to

21. M. E. Francis, "Accounting and the Evaluation of Social Programs: A Critical

Comment," Accounting Review, XLVIII (Apr., 1973), 250—51.
200 R. Lee Brum met

understand and represent economic and social phenomena. "Protec¬

tion of moneys against fraudulent use" is an important role of the
accountant but certainly not the sole "rationale for existence." The
suggestion of a new "independent body . . ." is also understandable
but probably not a likely response. My view that the accounting
profession should respond is made with full recognition of the mind-
boggling complexity of this new area that we call social accounting. It
is made, nevertheless, on the solid foundation of (1) the substance of
the needs of society for social information, and (2) my faith in the
capability of present and future members of the profession. While
accountants do not all have currently the needed understanding of
concepts and the tools of measurement needed for this new area, the
profession is rich with perceptive men and women who are commit¬
ted to serve the information needs of their constituencies—men and
women who are capable and willing to develop and apply new para¬
digms in their activities, new concepts of organizational perfor¬
mance, and new techniques of measurement and communication.
Accounting has relied greatly in the past on measurement con¬
cepts of engineering and economics, and of course on algebraic and
arithmetic notions. The field must now consider the concepts and
possible measurement constructs that may be provided by statisti¬
cians, sociologists, psychologists, and perhaps anthropologists. A
joining of the expertise of these groups with that of accountants
would provide good justification for an optimistic forecast of prog¬
ress in the new field of social accounting. The accounting profes¬
sion should, with reasonable modesty, take a leadership role. Ac¬
countants should be aggressive participants in the development of
social accounting without any thought of preempting any other
groups. There are not, I think, too many groups clamoring for prim¬
acy in this area. I think, rather, there are probably too few trying to
"light the candle" or "get on the learning curve."

Responsibility of Accounting Educators

Accounting education is the principal agent for change in our pro¬

fession. Accounting educators have a critical responsibility if social
accounting is to become a significant part of the activities of the
profession. First, we, as educators, must instill in our students a view
of the broad and honored role of the profession that they aspire to
enter. Second, we must encourage our students to develop view¬
points and concepts which enable them to grasp the nature and
significance of social performance and social well-being. Third, we
 Social Performance Accounting 201

must see that our students develop a firm grasp on the measurement
techniques that are most likely to find application in this new area of
interest. And fourth and finally, we must encourage our students to
develop a kind of personal modesty regarding their capability in this
field, recognizing the tentative nature of our knowledge in these
early days of the development of the area and the dangers as well as
the great potential for our profession's involvement.
The educational setting should be one in which optimality is not
necessarily synonymous with profit maximization. Concepts of so¬
cial good and welfare economics should be considered as critical as
classical economics with its emphasis on marginalism, price theory,
and rather elegant model building which generally ignores individ¬
ual differences. Psychometrics should be considered as important as
econometrics. The setting should be amenable to dealing with con¬
cepts of attitudes, judgments, and perceptions and thus with less
elegant models that do consider individual differences. Accountants
in such an environment should in time become almost as comfort¬
able in dealing with “sensing” technology as they are today with
“counting” activities.
These may seem to be almost revolutionary ideas. They should
not be interpreted as a weakening of the accounting profession's
role in the management of private sector, profit-making organiza¬
tions or of its service in the maintenance of orderly and active
capital markets. They are, rather, expansionistic, based on a convic¬
tion that our profession can in time expand its role and become
stronger by serving society in a broader spectrum of ways.

Extending the Role of an “Embattled" Profession

Some accountants find entry into the field of social performance

accounting particularly unacceptable to them at a time when the
profession is experiencing more and more critical pressures from
many directions. I suggest that some prodding is in order at this
time, especially because of this special juncture in the history of the
profession. The profession must resist understandable tendencies to
find comfort in safe, continually more limited and less relevant
areas. The future of the profession depends far more on its success
in fulfilling the legitimate and deeply felt needs of society in the
next decades than it does on the resolution of all of the controversial
technical issues that it considers day after day, or on the success of
its major firms in winning their day in court in connection with
pending litigation.
202 R. Lee Brummet

John Lawler made the point very well when speaking to the
Council of AICPA in 1976:
The mission of accounting is being dramatically redefined by public
needs—summoning the profession not only to experiment with creative
approaches to its traditional responsibilities, but to venture forth into
new areas of service even though that entails the risks inherent in all
pioneering efforts.

The ogre of legal liability will remain a frightening presence; but the
most grievous injury that this threat could inflict upon the profession is
to instill a timidity which inhibits CPAs from achieving their full poten¬
tial in service to society.22

We must not claim to be able to do that which the state of the art
does not make possible, but we must not shrink from the effort and
opportunity to extend the state of our art. Measuring the unmeasur¬
able is not possible today, but measuring that which previously
was unmeasurable is always a possibility. As Leo Herbert stated
several years ago:
... to retain its present commanding role in the society and to play a
more significant one in the future, the accounting profession must legiti¬
mately extend its boundaries and capabilities. It then will not only have
to accept what changes the environment forces upon it but will be able to
direct needed changes in social, economic, governmental, and other en¬
vironmental conditions.23

A great profession constantly does its research in an effort to pre¬

pare for the future. It reviews and anticipates its potential for service
to society. It continually retrains itself to serve in the fullest and most
relevant ways. At the same time, it is expert in highly specialized
areas of service and eclectic in its view of its role in society.
The accounting profession cannot claim such greatness. Accoun¬
tants continue to cling tenaciously to practices that are crudely
based on the questionable assumption of 'The economic man,” and
to the double-entry balancing techniques so ingeniously devised in
Italy over five centuries ago.24 The accounting profession needs to

22. John Lawler, Remarks before the AICPA Council, Boca Raton, Fla., May 3, 1976.
23. Leo Herbert, "A Perspective on Accounting," Accounting Review, XLVI (July
1971), 440.
24. This point comes through clearly in the work thus far on the Conceptual
Framework Project of the Financial Accounting Standards Board, where the role of
accounting is visualized as providing "investors" and "lenders" with information
useful in "assessing the risks and returns associated with an investment." Then,
further, it sets out to determine definitions of assets, liabilities, revenues, and ex¬
penses and to decide whether the first two or the second two elements should
dominate in keeping the books in balance.
 Social Performance Accounting 203

define its role in society—not just to determine the purposes of

financial statements. It needs to assess performance of organiza¬
tions along many dimensions—not just to determine standards for
financial accounting. An aggressive move to become involved in
measurement and communication of social performance of corpora¬
tions could contribute to the image and stature of the profession
and go far to assure it a bright future.
Bibliography

Comprehended within this bibliography are virtually all of Pa-

ton's major published writings. Every effort has been to make the
list complete, although it is improbable that the compilation is
wholly without errors or omissions. We have elected not to include
translations into other languages of works originally published in
English. Also excluded are reprintings, in anthologies and other
collections, of articles already listed. Although we have endeavored
to include book reviews and submissions to the federal govern¬
ment, we have made no attempt to cite all of the many occasions
when Paton gave testimony before regulatory commissions in the
several states.

Books

Principles of Accounting. (Russell Alger Stevenson, co-author.) New York:

The Macmillan Co., 1918.

The Economic Position of the United Kingdom: 1912-1918. Economic Studies of

Countries during the War. Bureau of Foreign and Domestic Commerce,
Misc. Series No. 96. Washington: Government Printing Office, 1919.

Studies prepared for the Bureau of Research and Statistics, War Trade
Board, and published in 1919 by the Government Printing Office:
japanese Steel and Shipping Negotiations.
Rationing Agreement with Norway.
Rationing and Shipping Negotiations with Sweden.
Requisitioning of Dutch Shipping.
Trade and Shipping Negotiations with Denmark.
Trade and Shipping Negotiations with Spain.

205
206 Bibliography

Accounting Theory, With Special Reference to the Corporate Enterprise. New

York: The Ronald Press Co., 1922. Reprinted, Chicago: A.S.P. Account¬
ing Studies Press, Ltd., 1962; Lawrence, Kan.: Scholars Book Co., 1973.

Accounting. New York: The Macmillan Co., 1924.

Editor, Accountants' Handbook. 2d ed. New York: The Ronald Press Co.,
1932; 3d ed., 1943.

Corporate Profits as Shown by Audit Reports. New York: National Bureau of

Economic Research, Inc., 1935.

Essentials of Accounting. New York: The Macmillan Co., 1938; 2d ed. re¬
vised, 1949.

An Introduction to Corporate Accounting Standards. (A. C. Littleton, co-au¬

thor.) Chicago: American Accounting Association, 1940.

Recent and Prospective Developments in Accounting Theory. Boston: Bureau of

Business Research, Graduate School of Business Administration, Harvard
Univ., 1940. (Lectures delivered under the Arthur Lowes Dickinson Fund
at the Harvard Graduate School of Business Administration in the spring
of 1940.)

Advanced Accounting. New York: The Macmillan Co., 1941.

Statement of W. A. Paton Regarding Overstatement and Misinterpretation of
Corporate Earnings under Current Conditions. New York: Republic Steel
Corporation, 1949.

Asset Accounting. (With the assistance of William A. Paton, Jr.) New York:
The Macmillan Co., 1952.

Shirtsleeve Economics: A Commonsense Survey. New York: Appleton-Cen-

tury-Crofts, Inc., 1952.

Statement of W. A. Paton Regarding the Interpretation of Corporate Earnings

and the Position of Stockholders under Current Conditions. New York: Re¬
public Steel Corporation, 1952.

Corporation Accounts and Statements: An Advanced Course. (With the assis¬

tance of William A. Paton, Jr.) New York: The Macmillan Co., 1955.

Essentials of Accounting. (Robert L. Dixon, co-author.) New York: The Mac¬

millan Co., 1958.

Make-or-Buy Decisions in Tooling for Mass Production. (Robert L. Dixon,

co-author.) Michigan Business Reports, No. 35. Ann Arbor, Mich.: Bu¬
reau of Business Research, School of Business Administration, The Univ.
of Michigan, 1961.

Corporate Profits: Measurement, Reporting, Distribution, Taxation; A Survey

for Eaymen and Accountants. Homewood, Ill.: Richard D. Irwin, Inc.,
1965.

Assets—Accounting and Administration. (William A. Paton, Jr., co-author.)

Warren, Mich.: Roberts & Roehl, Inc., 1971.
 Bibliography 207

Articles
Severity of Trade Cycle in America/' Current Economic Problems. Edited by
Walton H. Hamilton. Chicago: University of Chicago Press, 1915, pp.
265-67.

“Theory of the Double-Entry System," Journal of Accountancy, XXIII (Janu¬

ary, 1917), 7-26.

"Significance and Treatment of Appreciation in the Accounts," Twentieth

Annual Report, 1918. Edited for the Michigan Academy of Science by
G.H. Coons. Ann Arbor: Michigan Academy of Science, 1918, pp. 35-49.
"Some Phases of Capital Stock," Journal of Accountancy, XXVII (May, 1919),
321-35.

"Transactions between Partner and Firm," Journal of Accountancy, XXVIII

(July, 1919), 33-38.
"Proprietors' Salaries," Journal of Political Economy, XXVIII (March, 1920),
240-56.

"Some Current Valuation Accounts," Journal of Accountancy, XXIX (May,

1920), .335-50.
"Depreciation, Appreciation and Productive Capacity," Journal of Accoun¬
tancy, XXX (July, 1920), 1-11.
"The Drug Store's Overhead," American Druggist, LXVIII (September, 1920),
17-24.
"Interest during Construction," Journal of Political Economy, XXVIII (Oc¬
tober, 1920), 680-95.
"Methods of Measuring Business Income," Administration, I (April, 1921),
509-26.
"Assumptions of the Accountant," Administration, I (June, 1921), 786-802.
"The Pro and Con of a Sales Tax," Administration, II (September, 1921), 367-
72.
"The Valuation of Inventories," Papers and Proceedings of the Fifth Annual
Meeting, American Association of University Instructors in Accounting.
Chicago, 1921, pp. 66-77.
"Inventory Valuation," Administration, III (March, 1922), 291-302.
"The Relation of Accounting Training to Economics," Commerce and Admin¬
istration, II (March, 1922), 110-12.
"Dividends in Securities," Administration, IV (October, 1922), 394-402.
"Valuation of Inventories," Journal of Accountancy, XXXIV (December,
1922), 432-50.
"The Educational Value of Training in Accounting," (President's Address)
Papers and Proceedings of the Seventh Annual Meeting, American Associa¬
tion of University Instructors in Accounting. Chicago, 1922, pp. 66-73.
208 Bibliography

"Simplification of Federal Income Taxation," Certified Public Accountant, II

(June, 1923), 141-43.
"Tendencies in Accounting Literature," Papers and Proceedings of the Ninth
Annual Meeting, American Association of University Instructors in Ac¬
counting. Chicago, 1924, pp. 64-69.
"Shall Whole Matter of International Debt Settlements Be Re-opened?"
American Accountant, XII (February, 1927), 17, 62.
"Accrued Depreciation on Seasoned Properties," Certified Public Accoun¬
tant, VII (July, 1927), 206-10.
"Distribution Costs and Inventory Values," Accounting Review, II (Septem¬
ber, 1927), 246-53.
"The Program in Accounting at Michigan," Michigan Alumnus, XXXIV (Feb¬
ruary, 1928), 167-68.
"Limitations of Financial and Operating Ratios," Accounting Review, III
(September, 1928), 252-60.
"Special Applications of Discounting," Journal of Accountancy, XLVI (Oc¬
tober, 1928), 270-82.
"How Michigan Trains Her Young Men for Accountancy Careers," Ameri¬
can Accountant, XIV (September, 1929), 495-96.
"The Dividend Code," Accounting Review, IV (December, 1929), 218-20.
"Economic Theory in Relation to Accounting Valuation," Accounting Re¬
view, VI (June, 1931), 89-96.
"Working Capital in Public Utility Regulation," Journal of Accountancy, LIV
(October, 1932), 287-99.
"Accounting Problems of the Depression," Accounting Review, VII (Decem¬
ber, 1932), 258-67.
" 'Shoestring' Banking," Certified Public Accountant, XIII (June, 1933), 333-
38.
"Public Ownership Unnecessary," Public Service Magazine, LVI (February,
1934), 41-42.
"Shortcomings of Present-Day Financial Statements," Journal of Accoun¬
tancy, LVII (February, 1934), 108-32.
"Aspects of Asset Valuations," Accounting Review, IX (June, 1934), 122-29.
"Costs and Profits in Present-Day Accounting," National Association of Cost
Accountants Bulletin, XVI (October 1, 1934), 123-39.
"Valuation of the Business Enterprise," Accounting Review, XI (March
1936) , 26-32.

"Presentation of Bond Discount," Accounting Review, XII (September

1937) , 285-90.

"Comments on A Statement of Accounting Principles," Journal of Accoun¬

tancy, LXV (March, 1938), 196-207.
 Bibliography 209

“Is It Desirable to Distinguish between the Various Kinds of Surplus,”

Journal of Accountancy, LXV (April, 1938), 285-89.
' The 'Gross Profit' Convention,” Accounting Forum, IX (October, 1938), 19-

Principles Related to Deferred Charges' and 'Prepaid Expenses,' ” Papers

on Accounting Principles and Procedures. New York: American Institute of
Accountants, 1938, pp. 26-30.

"Recording Revenue on Other Bases Than Sales,” Accounting Ledger I

(April, 1939), 22-24.

"Income and the Inventory Problem,” Proceedings of the Dedication of Vin¬

cent Hall. Minneapolis: Univ. of Minnesota, 1939.
"Objectives of Accounting Research," Papers on Auditing Procedure and
Other Accounting Subjects. New York: American Institute of Accountants
1939, pp. 229-33.

"Last-In, First-Out,” Journal of Accountancy, LXIX (May, 1940), 354-60.

"Analysis of Financial Statements," Cost and Management, XXII (September,

1940), 303-19.

"Basic Concepts in Property Accounting," Edison Electric Institute Bulletin,

IX (March, 1941), 93—96. (Address at Fourth National Accounting Confer¬
ence, Detroit, Michigan, December, 1940.)

"Accounting Reports," Proceedings of the Third Annual Institute of Account¬

ing. Columbus: Ohio State Univ., 1940, pp. 5-12.
"The Cost Approach to Inventories," Journal of Accountancy, LXXII (Oc¬
tober, 1941), 300-307.

"Comments on The Cost Principle by Walter A. Staub," Accounting Review,

XVII (January, 1942), 10-19.

"Accounting," Encyclopedia Americana, Vol. I. New York: Americana Cor¬

poration, 1943—.

"Adaptation of the Income Statement to Present Conditions," Journal of

Accountancy, LXXV (January, 1943), 8-15.
"Classification and Sequences in Financial Statements," Accounting Prob¬
lems in War Contract Termination, Taxes, and Postwar Planning. New York:
American Institute of Accountants, 1943, pp. 57-67.

"Simplification of Federal Tax Administration," Accounting Review, XIX

(January, 1944), 11-19.

"Accounting Policies of the Federal Power Commission—A Critique," Jour¬

nal of Accountancy, LXXVII (June, 1944), 432-60.
"Responsibilities of the Postwar Accountant," Ohio Certified Public Accoun¬
tant, III (June, 1944), 1.
"Modern Reports to Stockholders," Current Considerations for Controllers
(Annual Meeting Proceedings, No. 2). New York: Controllers Institute of
America, 1944, pp. 116-24.
210 Bibliography

“Responsibilities and Training of the Postwar Accountant," Proceedings of

the Seventh Annual Institute on Accounting. Columbus:'Ohio State Univ.,
1944, pp. 6-18.
"Transactions between Affiliates," Accounting Review, XX (July, 1945), 255-
66.
"Current Economic Fallacies," Commercial arid Financial Chronicle, CLXII
(August 23, 1945), 825, 843.
"Balance Sheet," Chapter II of Contemporary Accounting. Edited by Thomas
W. Leland. New York: American Institute of Accountants, 1945.
"Cost and Value in Accounting," Journal of Accountancy, LXXXI (March,
1946), 192-99.
"Accounting Policies and Management," Proceedings of the Accounting Con¬
ference. Chicago: Illinois Society of Certified Public Accountants, 1946,
pp. 11-16.
"Restoration of Fixed Assets to the Balance Sheet—Second Affirmative."
Accounting Review, XXII (April, 1947), 198-200.
"The Queue Society—Do We Want It?" Michigan Alumnus Quarterly Re¬
view, LIII (May, 1947), 185-91.
"The Accountant and Private Enterprise," Journal of Accountancy, LXXXV
(January, 1948), 44-58.
"What Is Actual Cost in Depreciation Accounting?" Journal of Accountancy,
LXXXV (March, 1948), 206-7.
"Accounting Procedures and Private Enterprise," Journal of Accountancy,
LXXXV (April, 1948), 278-91.
"Depreciation and the Price Level—Second Affirmative," Accounting Re¬
view, XXIII (April, 1948), 118-23.
"Some Comments on Group and Sectional Interests," Michigan Alumnus
Quarterly Review, LIV (May, 1948), 203-8.
"Analysis of Financial Statements," Cost and Management, XXII (September,
1948) , 303-19.
"Why Corporate Profits Are Overstated," Commercial and Financial Chroni¬
cle, CLXVIII (December 16, 1948), 2, 33-35.
"Basis for Adequate Financial Reporting," Organization Controls and Execu¬
tive Compensation. New York: American Management Association, 1948,
pp. 34-44.
"1948 Revision of the American Accounting Association's Statement of
Principles: Comments on Item 5, under 'Expense,' " Accounting Review,
XXFV (January, 1949), 49-53.
"Profits Are Not Too Fligh," Credit and Financial Management, LI (January
1949) , 15-17, 28-29.

"Financing Problems as Seen by the Accountant," Credit and Financial Man¬

agement, LI (September, 1949), 18-20.
 Bibliography 211

"Lack of Venture Capital," National Savings and Loan Journal, IV (Septem¬

ber, 1949), 23-24.
"Semantics of Annual Reporting," Annual Report Forum. Ann Arbor: De¬
troit Trust Company in Co-operation with the School of Business Admin¬
istration, Univ. of Michigan, 1949, pp. 4-11.
"Accounting Problems Relating to the Reporting of Profits," Proceedings of
a Public Forum, with J. K. Lasser and Herbert E. Hetu. New Wilmington,
Pa.: Economic and Business Foundation, 1949, pp. 8-23.
"Measuring Profits under Inflation Conditions: A Serious Problem for Ac¬
countants," Journal of Accountancy, LXXXIX (January, 1950), 16-27.
"Economic and Accounting Aspects of Pension Plans," Michigan Certified
Public Accountant, II (September, 1950), 5-6.
"Use of Accounting Data in Economic Analysis," Accounts and Taxes. Ken¬
tucky Institute on Accounting and Tax Practitioners Forum Bulletin, No.
21. Lexington, Ky.: Bureau of Business Research, College of Commerce,
Univ. of Kentucky, 1950, pp. 66-83.
"How Have Stockholders Been Faring?" Michigan Business Review, IV (May,
1952) , 8-13.
"Should the SEC Continue to 'Study' Utility System Operation?" Public
Utilities Fortnightly, L (October 9, 1952), 473-80.
"Common Sense—The Curb to Public Spending," Tax Digest, XXXI (March,
1953) , 81-86.
"Utility Rates Must Recognize Dollar Depreciation," Howard C. Greer, co¬
author, Public Utilities Fortnightly, LI (March 12, 1953), 333-56.
"Unless We Have Courage—Spending Won't Be Cut Until 'Handout' De¬
mands Stop," Tax Outlook (April, 1953), 9-11.
"The Curb to Public Spending—There's No Such Thing as a Free Lunch,"
Public Service Magazine, XCV (July, 1953), 4-8.
"Premature Revenue Recognition," Journal of Accountancy, XCVI (October,
1953), 432-37.
"The Depreciation Deduction—A Neglected Aspect," Michigan Business Re¬
view, V (November, 1953), 23-26.
"Effect of Inflation on Measurement of Cost and Income," Underwear News
(August, 1959), pp. 4, 16-20.
"Depreciation—Concept and Measurement," Journal of Accountancy, CVIII
(October, 1959), 38-43.
"Insufficient Depreciation: What It Does To The Economy," Current Busi¬
ness Studies, Nos. 32-33. New York: Society of Business Advisory Profes¬
sions, Inc., 1959, pp. 26-40.
"Depreciation Deduction—LIFO Principle Should Be Extended to Cover
Depreciable Plant," Tax Revision Compendium, II (1959), 877-90.
212 Bibliography

"Measuring the Depreciation Deduction/' Proceedings of the Conference of

the Great Lakes Utility Commissioners. Columbus, Ohio* Great Lakes Rail¬
road and Utility Commissioners, 1960, pp. 59-65.
"More Squeeze on Earning Power Could Spell Trouble," Credit and Finan¬
cial Management, LXIII (January, 1961), 11-12.
" 'Deferred Income'—A Misnomer," Journal of Accountancy, CXII (Septem¬
ber, 1961), 38-40.
"Taxpayer Support for Higher Education in Michigan—Some Sidelights,
The Detroiter, LII (September 25, 1961), 6-7.
"The Tale of the Little Red Hen," The Freeman, XXI (November, 1961), 18-
19.
"On Going Underground," Michigan Quarterly Review, I (January, 1962),
19-26.
"The 'Cash Flow' Illusion," Accounting Review, XXXVIII (April, 1963), 243-
51.
"Accounting and Utilization of Resources," Journal of Accounting Research, I
(Spring, 1963), 44-72.
"Incredible Schlude Decision," The Michigan CPA, XV (July-August, 1963),
17-18.
"Comments on Stock Options," Michigan Business Review, XV (November,
1963), 21-26.
"The Beetle and the Centipede," The Freeman, XIII (October, 1963), 27-29.
"Middle-of-the-Roadism—A Questionable Stance," Michigan Quarterly Re¬
view, II (Winter, 1963), 22-26.
"Let's First Mend Tommy's Trousers," The Freeman, XV (January, 1965), 12-
17.
"U.S.A.—Disaster Area," The Freeman, XVI (February, 1966), 16-20.
"Controlling HCL," The Freeman, XVII (January, 1967), 25-31.
"Some Reflections on Education and Professoring," Accounting Review,
XLII (January, 1967), 7-23.
"Why Splits?" Michigan Business Review, XX (January, 1968), 16-19.
"Observations on Inflation from an Accounting Stance," Journal of Account¬
ing Research, VI (Spring, 1968), 72-85.
"Comments" in Accounting for Goodwill, Accounting Research Study No.
10, by George R. Catlett and Norman O. Olson. New York: American
Institute of Certified Public Accountants, 1968, 143-51.
"An Inquiry Concerning Inequality," The Freeman, XIX (January, 1969), 53-
60.
"Postscript on 'Treasury Shares,' " Accounting Review, XLIV (April, 1969),
276-83.
"Funds, Flows, and Fancies," U.S. Army Audit Agency Bulletin, (Winter,
1969), 49-59.
 Bibliography 213

"The E's Have It," The Freeman, XX (February, 1970), 108-21.

"The Protesters," The Freeman, XXI (January, 1971), 51-59.
"Earmarks of a Profession—And the APB," Journal of Accountancy, CXXXI
(January, 1971), 37-45.

"Do Casualties Accrue?" Michigan Business Review, XXIII (November 1971)

20-24.

"Accounting's Educational Eclipse," Journal of Accountancy, CXXXII (De¬

cember, 1971), 35-37.

"Let's Recognize Depreciation in Government Operations," GAO Review,

(Winter, 1971), pp. 38-44.

"Recollections Re a Kindred Spirit," in Toward Liberty, a collection of essays

honoring the 90th birthday of Ludwig von Mises. Menlo Park, Calif.:
Institute of Humane Studies, 1971, vol. II, pp. 249-67.
"Can We Sustain Prosperity?" The Freeman, XXII (January, 1972), 33-41.
"Who Is the Marginal Producer?" The Freeman, XXII (March, 1972), 160-63.
"5° Above Zero," The Freeman, XXII (October, 1972), 612-15.
"Inflation—Measurement, Impact, Culprits," Financial Executive, XL (No¬
vember, 1972), 48-60; XL (December, 1972), 44 -54.
"Back to Basics—Fable of the Berry Pickers," The Freeman, XXII (December,
1972), 707-14.
"Comments on the Role of Professional Accountants," Michigan CPA, XXIV
(January-February, 1973), 11-13.
"What P-R-O-D-U-C-T-I-O-N Means," Michigan Business Review, XXV
(March, 1973), 7-8.
"Financial Misrepresentation and Measurement," Financial Executive, XLI
(April, 1973), 14-21.
"Competition—What and When?" The Freeman, XXIII (June, 1973), 353-65.
"The Gullible Society," The Freeman, XXIV (January, 1974), 3-9.
"Accounting Today—A Bird's-Eye View," Michigan Business Review, XXVI
(January, 1974), 22-25.
"The 'Social Security' Mirage—Current Production Paramount," The Free¬
man, XXIV (March, 1974), 167- 71.
"Facts and Fallacies in Business Finance," Financial Executive, XLII (June,
1974) , 76-82.
"Opportunity for the Financial Accounting Standards Board," Michigan
CPA, XXV (May-June, 1974), 27-30.
"Some Amendments to Interest and Profit Theory," Michigan Business Re¬
view, XXVI (September, 1974), 16-24, 33-35.
"Significance of Services—Hiring and Firing," The Freeman, XXV (January,
1975) , 48-57.
214 Bibliography

“Sensitivity to Dollar Devaluation—Bane or Blessing?" Michigan CPA, XXVI

(January-February, 1975), 9-11.
“An Ancient Fable Retold," The Freeman, XXVI (March, 1976), 150—54.
“Notes on Handicapping," The Accounting Historian, III (Spring, 1976), 4.

Book Reviews

Review of Income in the United States: Its Amount and Distribution, 1909-1919
(New York: National Bureau of Economic Research, 1922), in Journal of
the American Statistical Association, XVIII (June, 1923), 804-8.
Review of Anthracite Accounting and Finance, Statement to the United States
Coal Commission by the Anthracite Representatives of the United Mine
Workers of America, 1923, in Journal of the American Statistical Associa¬
tion, XVIII (December, 1923), 1071-73.
Review of Studies in the Economics of Overhead Costs, by J. Maurice Clark
(Chicago: University of Chicago Press, 1923), in Journal of the American
Statistical Association, XIX (June, 1924), 257-59.
Review of Interest as a Cost, by Clinton H. Scovell (New York: Ronald Press
Co., 1924): in American Economic Review, XV (March, 1925), 321-26.
Review of Managerial Accounting, I, by J.O. McKinsey (Chicago: University
of Chicago Press, 1924): in Journal of Political Economy, XXXIII (August,
1925), 470-72.
Review of Theory of Accounts, by D. R. Scott (New York: Henry Holt, 1925),
in Journal of the American Statistical Association, XXI (March, 1926), 102-5.
Review of Retail Accounting and Control, by Albert C. Hodge (Chicago:
University of Chicago Press, 1925), in Journal of Political Economy, XXXIV
(December, 1926), 782-83.
Review of Income in the Various States: Its Sources and Distribution, 1919,
1920, and 1921, by Maurice Leven and Assistants (New York: National
Bureau of Economic Research, 1925), in Journal of the American Statistical
Association, XX (March, 1927), 116-18.
Review of Survey Course in Accounting, by Wesley James McCarty and L.
Cleveland Amidon (New York: Prentice-Hall, Inc., 1926), in Journal of
Political Economy, XXXVI (February, 1928), 174-76.
Review of Accounting Method, by C. Rufus Rorem (Chicago: University of
Chicago Press, 1928), in Accounting Review, III (December, 1928), 411-13.
Review of Consolidated Balance Sheets, by George Hillis Newlove (New
York: Ronald Press, 1926), in Journal of the American Statistical Associa¬
tion, XXIV (March, 1929), 96-97.
Review of Economic Rhythm, A Theory of Business Cycles, by Ernst Wage-
mann (New York: McGraw-Hill, 1930), in Accounting Review, VI (March,
1931) , 75-76.
Review of A Source-Book for the Study of Industrial Profits, by R.C. Epstein
and F.M.A. Clark (Washington, D.C.: Government Printing Office,
1932) , in American Economic Review, XXIV (March, 1934), 172-73.
 Bibliography 215

Review of The Depreciation of Capital, Analytically Considered, by R. F.

Fowler (London: P. S. King, 1934), in American Economic Review, XXIV
(December, 1934), 726-27.
Review of Accounting, by Charles H. Porter and Wyman P. Fiske (New
York: Henry Holt, 1935), in American Economic Review, XXVI (March,
1936), 122-24.
Review of Engineering Valuation, by Anson Marston and Thomas R. Agg
(New York: McGraw-Hill, 1936), in American Economic Review, XXVII
(March, 1937), 155-57.
Review of Studies in Income and Wealth, I, Conference on Research in Na¬
tional Income and Wealth (New York: National Bureau of Economic Re¬
search, 1937), in Journal of Accountancy, LXVII (January, 1939), 63.
Review of Depreciation: Principles and Application, by Earl A. Saliers (New
York: Ronald Press, 1939), in Journal of Political Economy, XLVII (October,
1939), 735-38.
Review of Capital Consumption and Adjustment, by Solomon Fabricant (New
York: National Bureau of Economic Research, 1938), in Journal of Accoun¬
tancy, LXVIII (November, 1939), 358-59.
Review of Truth in Accounting, by Kenneth MacNeal (Philadelphia: Univer¬
sity of Pennsylvania Press, 1939), in Journal of Political Economy, XLVIII
(April, 1940), 296-98.
Review of The Taxation of Corporate Income, by Charles John Gaa (Urbana,
Ill.: University of Illinois Press, 1944), in American Economic Review,
XXXV (December, 1945), 976-78.
Review of Accounting: Administrative /Financial, by George R. Husband
(Philadelphia: Chilton Co., 1960), in Accounting Review, XXXV (July,
1960), 572-73.
Review of Principles of Public Utility Rates, by James C. Bonbright (New
York: Columbia University Press, 1961), in Accounting Review, XXXVII
(January, 1962), 166-69.

Correspondence

“Some Phases of Capital Stock," Journal of Accountancy, XXVIII (August,

1919), 158-60.
"Some Phases of Capital Stock," Journal of Accountancy, XXVIII (November,
1919), 474-77. "A Statement of Accounting Principles," Journal of Accoun¬
tancy, LXV (April, 1938), 328.
"A Question of Usage," Journal of Accountancy, Cl (June, 1956), 24.
Reproduction of correspondence with Howard C. Greer, Accounting Review,
XXI (January, 1946), 85-88.
"An Application of Replacement Value Theory," Journal of Accountancy, CX
(September, 1960), 31-33.
"Tax Reform for Goodwill," Tax Adviser, II (May, 1971), 262.
216 Bibliography

"Pollution Cost/' Journal of Accountancy, CXXXIII (June, 1972), 28, 30.

"Calendars and Time Zones," Journal of Accountancy, CXXXIII (June, 1972),
28.
The Accounting Historian, III (Winter, 1976), 4.

Submissions to the Federal Government

Revenue revision. Hearings before the Committee on Ways and Means,

House of Representatives, 72d Congress, 1st Session, January 13, 1932
(Statement, pp. 366-71).
Corporate profits. Hearings before the Joint Committee on the Economic
Report, U.S. Congress, 80th Congress, 2d Session, December 7, 1948
(Statement, pp. 53-77).
General revenue revision. Hearings before the Committee on Ways and
Means, House of Representatives, 83d Congress, 1st Session, Part II, July
22, 1953 (Statement, pp. 699-704).
Financial policy for the Post Office Department. A statement by the Post
Office Department, 1954 (letter on "nonapplication of 'incremental cost'
principle to postal cost accounting," dated January 22, 1954, pp. 221-25).
Federal tax policy for economic growth and stability. Papers submitted by
panelists appearing before the Subcommittee on Tax Policy, Joint Com¬
mittee on the Economic Report, U.S. Congress, 84th Congress, 1st Ses¬
sion, November 9, 1955 ("Significance of Depreciation Accounting with
Special Reference to Plant Replacement," pp. 528-38).
Federal tax policy for economic growth and stability. Hearings before the Sub¬
committee on Tax Policy, Joint Committee on the Economic Report, U.S.
Congress, 84th Congress, 1st Session, December 13, 1955 (Panel discus¬
sion, pp. 417-56).
Postal rate revision. Hearings before the Committee on Post Office and Civil
Service, House of Representatives, 84th Congress, 2d Session (Letter
dated March 14, 1956, pp. 195-96).
Postal rate revision. Hearings before the Committee on Post Office and Civil
Service, House of Representatives, 85th Congress, 1st Session (Letter
dated March 15, 1957, p. 706).
General revenue revision. Hearings before the Committee on Ways and
Means, House of Representatives, 85th Congress, 2d Session, Part II,
January 30, 1958 (Statement, pp. 2402-25).
Tax revision compendium. Compendium of papers on broadening the tax
base, submitted to the Committee on Ways and Means, House of Repre¬
sentatives, 86th Congress, 1st Session, Volume II ("The Depreciation
Deduction—Lifo Principle Should be Extended to Cover Depreciable
Plant," pp. 877-90).
Income tax revision. Panel discussion before the Committee on Ways and
Means, House of Representatives, 86th Congress, 1st Session, November
30, 1959 (Statement, pp. 416-19).
 Bibliography 217

Economic concentration. Hearings before the Subcommittee on Antitrust and

Monopoly, Committee on the Judiciary, Senate, 89th Congress, 1st Ses¬
sion. Part IV, August-September, 1965 (Statement on make-or-buy deci¬
sions, pp. 1855-61.)
Statement regarding the overstatement and misinterpretation of corporate
earnings under current conditions. In the matter of United Steelworkers
of America—CIO and Republic Steel Corporation, et al., before the Presi¬
dential Steel Board, August 19, 1949.
Statement regarding the interpretation of corporate earnings and the posi¬
tion of stockholders under current conditions. In the matter of United
Steelworkers of America—CIO and various steel and iron ore companies,
before the steel panel. Wage Stabilization Board, February 14, 1952.

Miscellaneous Writings

Editorial Note, in Maurice Moonitz, The Entity Theory of Consolidated State¬

ments. Bloomington, Ind.: American Accounting Association, 1944, pp.
52-53.
Foreword to the reissue of Henry W. Sweeney, Stabilized Accounting. New
York: Holt, Rinehart and Winston, Inc., 1964, pp. xiii-xv.
Introduction, in Williard E. Stone (ed.). Foundations of Accounting Theory.
Gainesville, Fla.: University of Florida Press, 1971, pp. ix-xi.
Foreword to the reissue of Charles E. Sprague, The Philosophy of Accounts.
Lawrence, Kans.: Scholars Book Co., 1972, pp. iii-vii.
Response, in Economic Calculation Under Inflation. Indianapolis, Ind.: Li¬
berty Fund, Inc., 1976, pp. 65-72.
 Date Due

MAY 1
 HF 5629 .E68
010101 000
Essays in honor of William A.

163 0107810 5
TRENT UNIVERSITY

HF5629 .E68
Essays in honor of William A.
Paton : pioneer accounting
theorist / 6

DATE
ISSUED TO