A CRITIQUE OF THE RATIONALE FOR REQUIRED CORPORATE
George J. Benston
Professor of Accounting and Finance
Graduate School of Management University of Rochester
May 6, 1974
note: Dr. George J. Benston was educated
at Queens College (B.A.), New York University (M.B.A.) and the University
of Chicago (Ph.D.). He also is a Certified Public Accountant. His
writings span the fields of accounting, finance, economics and banking.
He is a consultant to banks, organizations, and
various Federal agencies, including the Federal Reserve Board, the
Comptroller of the Currency, The President's Commission on Financial
Structure and Regulation, and the National Commission on Consumer
Finance. Presently, he is preparing a study comparing financial
disclosure in the United States and the United Kingdom for the Institute
of Chartered Accountants in England and Wales.
In his lecture, Professor Benston will consider
the rationale for government regulation of financial disclosure
in a free society rather than simply assume that "more is better"
as does the SEC, the AICPA, security analysts, and the financial
press. He then will examine the benefits presumed to flow from mandated
disclosure. This examination will include a review of the extant
evidence on the need for and value of laws requiring financial disclosure.
Some additional insight into the benefits and costs of government
required disclosure will be obtained from a contrast of the regulatory
agency system followed in the United States with the United Kingdom
Professor Benston's analysis is particularly timely
in view of the extensions of disclosure currently sought by the
SEC. He draws some possibly unorthodox conclusions from his analysis
which might give those who believe "more is better" some
reason to question the value to society of government regulation
For many of us who grew up under Franklin Roosevelt, learned our
accounting or worked in the financial markets since the early 1930's,
it's hard to conceive of a world without the SEC. Somehow it seems
as though disclosure always was required, and that until it was required
it didn't really exist. It is very hard, I think, for financial analysts
to conceive of a world in which the 8-Ks, 10-Qs, and 10-Ks aren't
available or in which the Securities Act of 1933 doesn't mandate prospectuses.
Almost without questioning, many accountants, analysts, comptrollers,
the financial press, stockholders and potential investors assume that
were it not for government required financial disclosure and the SEC
the market for securities would be unregulated and hence unfair and
inefficient. So it seems anachronistic to question the desirability
of government required disclosure.
Nevertheless, we should question the desirability and efficacy of
government intervention in the securities markets, or anywhere else
for that matter. This assertion is based on my belief that we are,
or at least want to be, a free society. In such a society, one presumes
that government ought not require anything of its citizens unless
some overriding aspect of the public interest commands it, some benefit
that offsets the cost imposed on private individuals. Thus we must
determine what are those benefits and what are the costs. In the analysis
that follows, I emphasize "what are the benefits" more than "what
are the costs" because, if it turns out that the benefits aren't very
great and if it seems the costs are great, the question essentially
is answered. However, if it turns out that the benefits are considerable,
we must then measure how great the costs are and delineate the distributive
effects of requiring disclosure on different groups in society. Until
we make these calculations and unless the benefits from required disclosure
clearly exceed the costs and the distributive effects either are negligible
or are in accordance with our ethical precepts, I suggest that government
mandated financial disclosure should be repealed or at least not expanded.
WHY SHOULD FINANCIAL DISCLOSURE BE REQUIRED?
The preliminary question is, "why should financial disclosure be
required?" I must emphasize that voluntary disclosure is not questioned,
though this is an interesting question for research and analysis.
However, only disclosure made in accordance with government mandate
(that is, in accordance with the Securities Acts of 1933 and 1934)
requires justification, since one can argue that such a requirement
is contrary to the basic tenets of a free society. In a free society,
contracts among private individuals ought to be freely determined.
If a person wants to buy the stock of a corporation that discloses
financial data, such is that person's right. If a person does not
want to buy the stock of a corporation that does not disclose, that
also is her right. Therefore, we should ask that government mandated
disclosure be justified before it is imposed.
Several reasons can be adduced to justify interference with people's
right to enter freely into security purchase contracts. I'll simply
list five of these and then discuss them: (1) the obligation of government
to provide for the enforcement of contracts; (2) a concept of fairness
or equity in dealings among people; (3) improved efficiency of resource
allocation; (4) administration of government and employees' and the
public's right to know; and (5) political considerations. Each of
the reasons is first explained and then the evidence that supports
or is contrary to the reason is marshalled.
THE OBLIGATION OF GOVERNMENT TO ENFORCE CONTRACTS -- REDUCTION OF
FRAUD AND MISREPRESENTATION
In general the government's obligation to provide for the enforcement
of contracts is a consequence of its police power and the rule of
law. In carrying out its judicial responsibilities, the government
has an interest in reducing the probability of fraud by making its
discovery and prosecution easier. Then the cost to the government
of enforcing contracts would be lower. But, one should ask, why is
disclosure necessarily related to fraud? If a company announced: "We're
not telling you anything, we do not intend to tell you anything, and
if you don't like, it, buy shares in another company", then one need
not invest in this company. If disclosure is not promised, its absence
is not fraud.
Nevertheless, we must recognize the existence of implied contracts
that a given society tends to expect the government to enforce. Specifically,
investors expect corporate management to act in a fiduciary capacity,
to operate the enterprise in the interest of the shareholders rather
than in the interests of management. Of course shareholders can agree
to a contract that allows corporate management to operate the enterprise
any way they wish. But such contracts are rare; it is difficult to
imagine shareholders knowingly agreeing to unrestrained selfserving
behavior by management. Rather when management is found appropriating
corporate resources or writing selfserving contracts, investors seek
redress in the courts or appeal to the legislature for subsidies to
offset their loss. Thus government might find prevention of fraud
and misrepresentation preferable to having to mediate disputes.
The question to be answered is whether or not required disclosure
is necessary or desirable for reducing fraud and misrepresentation.
A preliminary consideration is whether or not fraud and misrepresentation
is a serious problem worthy of government intervention. Unfortunately,
it is difficult to know how much fraud there is at any time, since
unless it is discovered, it is not recorded. Nevertheless, it is useful
to examine evidence on financial statement fraud before 1933 and 1934,
when the federal Securities Acts were enacted. For this purpose, I
reviewed all the evidence I could find including the many volumes
of Senate hearings that preceded the Securities Acts and the published
legal cases. Fortunately, a survey of the cases dealing with the accountants'
liability prior to 1934 was published in 1935 by Wiley Rich.(1)
This study reveals very few cases in which fraud or gross negligence
was alleged against public accountants before the passage of the Securities
Acts. The Senate hearings also reveal virtually no instances of fraud
or misrepresentation with respect to financial statements. However
one reason for the dearth of cases may have been the difficulty of
suing accountants for fraud because of the rule of privity, under
which the courts held that an accountant's statements were prepared
for the corporation and, in general, only the corporation had the
right to sue. So it is possible that there was fraud but it was not
revealed by the court cases (although this does not explain the lack
of claimed fraud at the Senate hearings).
The Securities Acts changed the legal situation considerably. Now
it is relatively easy to sue accountants and others for fraud. The
plaintiff does not even have to prove she ever saw the financial statements,
but simply that the statements contained a material misstatement.
And then the accountant has to prove that she was justified in believing
the figures given to her by management or that the plaintiff experienced
no damages. Nevertheless, almost no cases are reported filed against
accountants until the 1950's. Had fraud in financial statements been
a serious problem before the Securities Acts were enacted, it seems
probable that some accountants would have been sued after 1934, unless
the incidence of fraud suddenly disappeared (which is hard to believe).
So I conclude (although I cannot be absolutely certain) that there
was very little fraud or serious misrepresentation in published financial
statements prior to 1934.
Nevertheless, one should explain why the general public seemed to
believe that financial statements often were fraudulent or misleading
if these were, in fact, not serious problems before the early 1930's.
I suggest that this folklore is due to several sources. One is the
writings of journalists and a crusading economist, William Z. Ripley.
In his work (most notably Main Street and Wall Street)
he excoriated financial accountants and corporations for misrepresentative
financial statements. The cases he refers to, though, date from the
1910's and the early 1920's. Virtually no cases are mentioned in the
middle 1920's and later. Thus the Securities Acts may have been designed,
in part, to solve a problem of an earlier decade that was no longer
a problem by the time the legislation was enacted.
The second source of the belief that financial statements were misleading
is unrealistic expectations about the meaningfulness of accounting
numbers. Indeed, I suggest that complaints about the "quality" of
financial statements heard today, as well as before 1933, are based
on the belief that accountants could produce "meaningful" statements,
but they won't. Thus people complain about misrepresentation because
there are various ways of recording depreciation, sales, purchases
and acquisitions of other corporations, etc., among which management
and accountants can pick and choose. However, the problem basically
is that accountants cannot report the wealth embodied in shares and
the change in that wealth (income) unambiguously because there is
no theoretically correct way to measure these variables. It is not
that accountants are not doing their jobs; rather it is not possible
for anyone to measure value objectively. That is what the securities
markets are all about -- it is there that investors' subjective estimates
of the present value of future expected cash flows are compared and
acted on. Consequently, people expect more from accountants than can
possibly be provided.
In addition, some decisions by accountants that are perfectly reasonable
ex ante may seem deliberately misleading ex post. For
example, ex ante an exchange of land for a long-term note may
properly be recorded as a sale if it appears that the note will be
collected. But if it turns out that the note isn't collected, the
exchange should not have been recorded as a sale because nothing of
value was received. But the accountant doesn't know that until after
the fact. And, should the note be good but the accountant chose not
to record the transaction until it was collected, someone might complain
that she didn't inform shareholders that their investments were more
valuable than they were previous to the transaction.
The problem is made even more difficult by the use of the historical
cost basis for recording assets. As a consequence, reported costs
and profits are a function of the original cost of an asset written
off or sold as much as of contemporary economic events. To reduce
manipulation by management, accountants insist on pre-determined (though
essentially arbitrary) rules to control how the original cost of assets
are expensed (such as straight-line depreciation of fixed assets and
first-in, first-out expensing of inventory). Nevertheless, managers
still can control the amount of profits reported in many situations.
For example, where the enterprise owns a number of economically identical
or similar assets (such as securities or parcels of land) that were
purchased for different amounts, management can choose which one to
sell and thus determine the reported profit.
In any event, is there any evidence that the Securities Acts, as
administered by the SEC, have reduced fraud or misrepresentation or
even improved the "quality" of reported financial data? If one reads
the SEC's dockets, it appears as though most small-time securities
hucksters have been or are being charged with violations or subjected
to stop orders. Nevertheless, investors would be well advised to take
the traditional warning, caveat emptor, seriously. Indeed,
there is reason to believe that in some respects the SEC has made
fraud easier, not more difficult. Their concern with financial statements,
which leads them to promulgate strict rules on what may and may not
be reported, tends to give many investors the impression that approved
practice (which in many instances cannot be other than misleading,
either by commission or ommission) is correct and useful.
More important, though, is the tendency of the SEC to act as though
it were a "blue skies" agency. This was not the intention of the Securities
Acts. As President Roosevelt put it, "This proposal (the Securities
Act of 1933) adds to the ancient rule of caveat emptor the further
doctrine 'let the seller beware.' It puts the burden of telling the
whole truth on the seller." But the SEC has restricted severely what
the seller can say. Appraisal of assets, estimates of future sales
and profits, and other "soft" information may not be included in a
prospectus. Disclosure as such is not allowed. Rather, in 1938, the
SEC declared in ASR (Accounting Series Release) 4: " ... where financial
statements filed ... are prepared in accordance with accounting principles
for which there is no substantial authoritative support, such financial
statements will be presumed to be misleading or inaccurate despite
disclosures in the certificate of the accountant or in footnotes
to the statements provided the matters involved are material" (italics
added). More recently, ASR 115 indicates that companies whose statements
show inadequate working capital after a new issue is floated will
not be accepted for registration. Consequently, despite the "red herring"
warning required on prospectuses that the SEC has not and does not
approve securities, it is not unreasonable for investors to assume
(and for dishonest securities salesmen to claim) that an SEC registered
prospectus has been given a sort of "Good Housekeeping" seal of approval.
(After all, many issues are rejected -- those allowed must be O.K.).
In addition, the "boiler plate" pessimistic descriptions of competition,
prospects, etc., make it easy for unscrupulous salesmen to claim:
"No company could be that bad -- the SEC makes you say that -- but
here's the real, inside story." Thus, investors should not assume
that the "bad old days" have passed. Rather, as SEC Chairman Ray Garrett,
Jr. observed recently: "We have had cases of fraud and mismanagement
and disregard of investor interest that rival anything known to the
men of 1933 who set about to construct a system that would make the
world safe for small investors ... "(2)
Considering such frauds as Equity Funding and Home Stake Production
(among others), the investor would do well not to forget the pre-SEC
warning, caveat emptor.
FAIRNESS (EQUITY) IN DEALINGS AMONG PEOPLE
Fairness (equity) is a difficult concept to define operationally.
For some people, a transaction or contract is "fair" when
they make a "reasonable" profit, whether others gain or lose. Few
would say a transaction is "fair" should they end up with
less than they began or even less than they expected to gain. Therefore,
many people feel cheated when the price of a security in which they
invest drops soon after their purchase. They ask who is responsible?
Perhaps the investors should have done more research, but failed to
exercise prudent care. Perhaps the sellers mislead them by knowingly
giving them false information. Or perhaps no one is responsible; it's
just a consequence of unpredictable events. Regardless of the actuality,
should something appear to have been "unfair," those adversely
affected look to government for a redress of what they think are "wrongs."
As with fraud, it might be less expensive for government to require
disclosure, if the disclosure has the effect of improving perceived
fairness in the market, than to mediate disputes.
Also as with fraud, before mandated financial disclosure is accepted
as a cure for "unfairness" we should determine whether the security
markets were unfair before the Securities Acts were enacted and whether
the Acts, as administered by the SEC, have made the markets more fair.
First "fairness" must be defined operationally. One definition used
by economists is that the market is a "fair game" (Eugene Fama's phrase)
if all available information is reflected completely by the price
of a security. (In the semi-strong form, a fair game market is one
in which all publicly available information is impounded into share
prices). In this event, the market price of shares is unbiased. While
the price of a share may not be "correct" (in the sense that in the
light of subsequent, presently unknown events, one would not offer
or sell the share for the amount paid or received), at the time of
the transaction there is no more likelihood for the market price of
the security to be higher than lower were additional information known.
Thus should one buy a share of General Motors stock for $49, the economic
effect of a plant that burned or an invention completed yesterday
would be impounded into that market price. Therefore, the investor
would not buy at a bargain or sell at a loss, had she but known what
Of course, it is not possible that everything can be known by everyone
at the same time: information cannot come to the market instantaneously,
somebody must know something before somebody else. Therefore it is
not likely that financial statements can be the sole or perhaps even
the primary means by which the market approximates a "fair game."
Nor is it desirable that financial statements be the sole source of
information about corporations. Consider an extreme situation, where
information is known only as it is published in the financial statements.
In this event the market price of shares would be "unfair" on all
except the days when statements are published because between the
dates of the financial statements something happened and whoever buys
or sells at that time wouldn't know about it. So it would be most
unfortunate for investors if financial statements were the sole
means of transmitting information.
But published financial statements might improve the transmission
of information. What is the evidence? There have been quite a few
statistical studies of share price movements and the relationship
between share prices and published financial statements. I will only
summarize the principal findings, since the literature is quite large.(3)
First, there are studies that speak to the question of how well the
stock markets operate. Unfortunately, one cannot ever know whether
the markets are what Eugene Fama calls "the strong form of the efficient
markets hypothesis," that is whether all information is impounded
into share prices because we can't determine when information is known
or even what, exactly, is or is not "information." Nevertheless, we
can say that if the successive price changes or any combinations of
successive price changes are statistically independent, the prices
follow a random walk (the weak form of the efficient markets hypothesis).
Consequently, knowledge of past price change patterns does not enable
one to predict what the next change will be. Nor can a trading rule,
such as those proposed by chartists, allow one to do more than break
even, gross of transactions costs. A market characterized by price
changes that follow a random walk is consistent with a market in which
all information is fully impounded into share prices. (I must emphasize
that this statistical property of share prices does not prove
that all information is reflected by share prices; it is just consistent
with this belief). Quite a few studies have been published on the
manner in which prices change in the securities markets in the United
States and United Kingdom. All of these studies show that the market
follows what can be described as a random walk. These studies include
U.S. data from the New York Stock Exchange before as well as after
passage of the Securities Acts, over-the-counter data, and U.K. data
from the (London) Stock Exchange where considerably less disclosure
is required than in the U.S. None of these studies indicates that
the behavior of share price changes is affected by the degree or extent
of government required disclosure. Specifically, successive share
price changes of U.S. companies followed a random walk before and
after passage of the Securities Exchange Act of 1934, with respect
to companies before and after passage of the '34 Act that did and
did not disclose sales prior to 1934; similar findings are reported
for U.K. companies, who are not required to disclose nearly as much
as are U.S. companies.
Another aspect of "fairness" is the absence of manipulation of share
prices. In a booklet, A 25 Year Summary of the Activities of the
Securities and Exchange Commission, the SEC claims that operators
of the more than 100 pools that operated during 1929 were able to
manipulate share prices because investors were unable to obtain reliable
financial information. First, one should consider that pools are successful
if people believe rumors about impending "great developments." Financial
statements, which refer to the past, should have little effect on
the ability of manipulators to spread rumors. Second, I reviewed the
financial statements of the more than 100 pools operated in 1929 through
1932.(4) All of the
companies whose shares were included in the pools' manipulations published
financial statements. The major item not published was sales. However,
almost the exact proportion of these companies published sales figures
as did the New York Stock Exchange listed companies whose shares were
not included in pool manipulations. Thus the fact and extent of financial
disclosure appears to be unrelated to share price manipulation.
IMPROVED EFFICIENCY OF RESOURCE ALLOCATION
The third reason for required disclosure, improved efficiency of
resource allocation, is a major argument underlying the Securities
Acts in the United States. If resources flow to those uses where the
marginal rate of return is the highest, the nation benefits. Required
disclosure, it is assumed, is necessary to permit investors to make
such resource allocation decisions. It is argued that corporations
would not voluntarily supply all of the information desired by investors
because non-shareholders (including competitors) cannot be charged
for the information and cannot be prevented from obtaining it. Consequently,
management who operate corporations to maximize the wealth of shareholders
would publish less than the optimal amount of information. In addition,
the amount of resources devoted to processing and interpreting financial
data might be reduced were the data available in some standard form,
such as 10-Ks and S-1s.
The quality of the data reported also is expected to be better as
a consequence of government supervision. Some critics believe that
many corporate managers and their hired accomplices, the CPAs, do
not provide data of sufficient quality (or quantity) to allow shareholders
to judge the managers' performance. Government mandated reports of
the managers' stewardship, these critics claim, would result in more
efficiently managed corporations.
The validity of this reason for requiring disclosure rests on the
belief that the data given in financial statements are both useful
and sufficiently timely for investors' decisions. Although the validity
of this belief cannot be determined definitively, there is considerable
reason to doubt that traditional accounting statement do, or even
can, serve this purpose. The meaningfulness of financial statement
data has been questioned in a large number of articles and books by
critics who point out that such practices as the historical cost basis
of recording assets and liabilities, arbitrary rules for allocating
assets to products and expense, non-capitalization of intangible assets
such as advertising, research and development and goodwill, etc.,
reduce the value of financial statements for resource allocation decisions.
At the present time, we have little research evidence that validates
or invalidates these criticisms. We do know, however, that government
intervention with financial disclosure has resulted in few changes
in the type of information disclosed. Indeed, the SEC has prevented
changes by considering financial statements that use current market
or price level adjusted figures, reappraisals of assets, capitalized
goodwill, etc., as unacceptable. While innovations and improvements
in financial reporting may not have occurred had the SEC not enforced
traditional, conservative practices, it is fair to say that the agency
has not used its resources to develop or support research on better
methods of reporting economic events.
The claim has been made, though, that financial statements are more
reliable as a consequence of government regulation of disclosure.
It is true that the SEC has been a positive force in improving auditing
procedures somewhat. The 1936 Interstate Hosiery Mills Inc. and other
cases and the recent moves by the SEC to require peer review of the
audit procedures, working papers, quality control, etc., of some large
CPA firms probably have resulted and will result in more effective
auditing practices. However, there is little evidence that poor audits
and deliberately misleading or grossly unreliable financial statements
were more prevalent before creation of the SEC. Thus, there appears
no more reason to conclude that government regulation of financial
reporting has improved or worsened the quality of financial data available
However, there is little doubt that the quantity of data disclosed
by corporations is greater as a consequence of government regulation.
The amount of data required by Regulation S-X and for various filings
far exceeds that published by most (perhaps all) corporations before
the SEC was established. But is this detailed data useful to investors
or are they simply more costly for corporations to produce? While
this question cannot be answered definitively, some statistical studies
on the relationship between published financial reports and share
prices are enlightening.
Tests of changes in share prices at or about the time financial
statements are published can provide us with insights into the use
to which investors (on the whole) put the statements. If the numbers
revealed in the statements provide investors with information we would
expect share prices to change as this information caused investors
to alter their previously formed expectations. The validity of the
tests is, of course, dependent on the validity of the method used
to estimate what information was expected before the statements were
published, on the expectation that the effect of the information published
is captured in fairly rapid changes in share prices, in an adequate
specification of other factors, etc.. Since one cannot be certain
that all of these requirements were met, the tests cannot be said
to be conclusive. Nevertheless, they are instructive.
I published the first of these statistical tests in which the "market
model" was used to account for the effect on share prices of changes
in the market as a whole. This paper is available elsewhere and the
method is now fairly commonly used and described in a large number
of papers.(5) Other
researchers have used several versions of the model and all have reached
the same conclusion.(6)
The share price change associated with the publication of financial
statements is inconsistent with the belief that investors find that
financial statements provide them with previously unknown or with
useful information, at least with respect to the reporting corporations.(7)
In addition, several studies have been made on the relationship between
share prices and changes and differences in reporting practices (such
as shifts from accelerated to straight-line depreciation and deferred
vs. flow-through treatment of deferred tax liabilities).(8)
These studies also are consistent with the hypothesis that market
participants either are not fooled by the effect of different accounting
practices on net income or disregard the financial statements.
Before a conclusion is drawn from these studies, I should mention
the possibility that no relationship between share prices and financial
statement data, when published, was found because investors learned
the contents of the statements before publication from other sources.
It can be argued that these other sources would not have existed or
would not be trusted were it not for eventual government-required
financial reporting. I tested this hypothesis, to the extent possible,
by examining the effect of the SEC's requirement that sales figures
be reported. Prior to passage of the Securities Exchange Act of 1934,
38 percent of the companies whose shares were listed on the New York
Stock Exchange did not report sales. If the SEC's requirements provided
investors with new information, one would expect a greater change
in the share prices (up or down) or in the perceived riskiness of
these companies, relative to the companies that previously reported
sales arid relative to general market changes (which, at that time,
were particularly important). But, the data reveal virtually no differences
in the market prices or riskiness of the shares of the companies that
newly revealed their sales compared to those who were unaffected by
the SEC's disclosure requirement.(9)
Finally, does the availability of corporate financial data in fairly
standardized form reduce the cost to investors of analyzing and using
information about corporations and increase the efficiency with which
corporations are managed? No studies that speak directly to this question
have been published. However, several observations can be made that
might provide a partial answer. First, there is little data to support
the assumption underlying this question that published financial data
in the form presently required by the SEC provide investors with "information."
Second, the principal services that tabulate financial data, Moody's
and Standard and Poor's, present little of the detailed data
required by the SEC. Indeed, with the exception of sales (amount and
by lines of business), the contents of these services are not much
different now from what they were before 1933. Third, at least one
measure of market efficiency, the random character of successive share
price changes, does not indicate a more rapid impounding of information
into share prices after than before passage of the Securities Acts.
Nevertheless, one cannot be certain that some unmeasured improvements
in the efficiency with which the security markets are operated and
corporations are managed has not resulted from required financial
disclosure. Thus far, though, no evidence has been presented that
identifies and measures these benefits.
The evidence and reasoning, then, do not support the contention that
government-required disclosure is used by investors in setting share
prices. Does this mean that disclosure as such is not useful? The
available studies do not answer this question directly. However, the
studies do indicate that the current emphasis placed by the SEC and
others on the value of published financial data for the efficient
allocation of resources in the economy is, at the least, overstated
and not demonstrated.
GOVERNMENT ADMINISTRATION AND EMPLOYEES' AND THE PUBLIC'S RIGHT
It has been argued that the economic affairs of a corporation are
not solely a matter of concern between the owners and their hired
managers, creditors, and the taxation authorities. Since privately-owned
corporations control most of an industrialized nation's wealth, employ
most of its citizens and produce most of the goods and services consumed,
their efficiency and the way in which they use the resources "entrusted"
to them is of concern to the general public. Employees also have a
basic right to information about their employer. Without required
disclosure, it is argued, shareholders would not demand as much information
as the public would wish, because shareholders cannot capture the
externality (a "better run economy") the information provides. In
addition, government administration (such as breaking up monopolies
and price control) may be more efficient were corporate financial
data generally available.
Considering the last reason for required disclosure first -- more
efficient government administration -- it is not clear that continuous
financial reporting by all except the smallest corporations is the
most efficient way to provide the government with needed information.
Stratified random sampling not only would be less costly for companies,
as a group, but would enable the government to collect the data needed
to answer specific questions. The data disclosed in financial reports
might not speak to the questions on which information is wanted.
In addition, the costs of and benefits expected from the data which
corporations must provide should be calculated. For example, the Federal
Trade Commission recently wants to require considerable information
about sales and profits by individual products and product groups
from large corporations. The benefits expected to be derived from
these data are not at all clear. Presumably they are useful for determining
whether monopoly practices exist in specific markets. However, to
my knowledge, no studies have been reported which indicate whether
these accounting data measure monopoly profits or how the data can
provide the government with other than useless and potentially misleading
information. Nor do I know of an analysis that shows that the estimated
costs of providing and using the data exceed the expected benefits
The contention that employees have a right to know their employer's
financial data is based on the belief that corporations have as much
responsibility to their employees as to their shareholders. The validity
of this belief could be debated. As a minimum, the implications of
dual responsibility by management toward employees and shareholders
for the latter's willingness to invest in corporations and for the
quality and quantity of products and services provided to consumers
should be examined. But even if one decides that employees have "a
right to know," it is doubtful that government required disclosure
of financial data can provide them with relevant information. As is
discussed above, accounting statements provide rather crude measures
of the economic condition and progress of an enterprise. For example,
employees cannot learn from financial statements whether or not specific
plants will offer continued employment at any given level. Neither
can the ability of a company to pay a wage increase be determined
from its earnings report, except where the company clearly has absorbed
losses. Therefore, whether or not the costs to shareholders and the
economy would offset benefits to employees from receiving financial
data, employees appear to obtain few benefits from required disclosure.
Finally, with respect to the public's right to know, it is difficult
to determine whether the expected externality of a more efficient
economy would be a consequence of required disclosure. If public reporting
of financial figures would tend to make managements more efficient
and allocate resources to their most productive uses, it is unclear
whether the cost to shareholders (corporations) of providing the information
exceeds the benefits derived therefrom. In any event, if this type
of information is useful to the general public, it is most likely
limited to rather broad data such as net income, sales, changes in
capital investments and inventories, etc.. It is difficult to imagine
how detailed information on allowances for depreciation, accrued expenses,
subcategories of revenue, etc., can be assimilated, much less used,
for general analysis of the economy. While at specific times specific
information may be desired by some people, it would seem that the
cost of requiring all corporations to report data that are, at best,
meaningful to a few, would result in costs that exceed any benefit.
A claimed public benefit from required financial disclosure, revelation
of the "excessive" profits of some corporations (such as many oil
producers), should be considered. Some advocates of required disclosure
believe that oligopolistic pricing practices and government subsidies
(direct and indirect as a consequence of tariffs, etc.) can be reduced
or eliminated as a consequence of public knowledge ("sunlight is the
best disinfectant") and outrage. They should consider that the traditional
accounting practices currently enforced by the SEC provide poor measures
of economic profits, particularly during and after periods of unstable
general prices. They also should consider that the conceptional impossibility
of "correct" allocations of joint costs over products and over time
allows management considerable latitude in reporting the amount of
profits on individual products. Consequently, it is doubtful that
the reported information would provide the public with other than
misleading conclusions about markets and the consequences of corporate
Though many shareholders and managers might not want government to
interfere in their private contracts by requiring disclosure, they
should consider the political aspects of disclosure laws. These laws
may substitute for much more restrictive, and hence much more costly,
statutory and administrative arrangements, such as commissions who
must pass on the merits of securities before they may be purchased
by the public.
Required disclosure may be a benefit were it to prevent scandals
or the appearance of scandals that seem to necessitate governmentally-imposed
remedies. Those who oppose required disclosure should consider that
legislators and political leaders are expected to solve problems.
However, because of the relatively short range of political horizons,
the appearance of solving a problem generally is an acceptable and
often a preferable substitute for its actual solution. Study of the
causes of problems and the effectiveness of proposed solutions usually
is too time consuming to be pursued. In addition, studies often result
in the uncovering of additional, even more intractable problems and
conclude with ambiguous suggestions. Therefore, given a crisis or
scandal, government tends to take immediate action, any action.
The specific action taken depends on the ability of the lawmakers
and their staffs to understand the situation and draft legislation
and on the skill of special interest groups in getting their views
represented in the bills. These special interest groups are in an
advantageous position for having laws drafted in their favor since
they generally have well-defined views whose implementation via legislation
is well thought out. Therefore it is not surprising that laws passed
to alleviate a crisis or correct abuses discovered as a result of
a scandal often turn out to serve quite different purposes. Specifically,
versions of the U.S. securities laws had been proposed for years before
their passage in 1933 and 1934. Investment bankers wanted a law that
would bring their smaller competitors "up to their standards," reformers
appalled by some reporting practices wanted legislation to enforce
acceptance of their ideas, other reformers wanted a government agency
to pass on the merits of securities, corporations that disclosed more
wanted similar practices required of their competitors, some accountants
wanted a means to require clients to follow practices the accountants
thought correct, etc.. Virtually no group had prepared studies to
support their beliefs. The legislation passed was a function of the
special interest groups' demands, newspaper reporting of a few scandalous
happenings in the late 1920s, and the demand that the government "do
Therefore, if some disclosure of financial information and the required
auditing of financial statements by independent accountants prevents
or reduces the incidence of scandal, the benefit from avoiding circumstances
in which punitive or special interest legislation can be passed may
exceed the cost to private corporations. In this regard, it would
seem that more emphasis should be placed on auditing to prevent or
reduce misappropriation of resources and similar frauds, events which
are considered "scandals" when they occur. Prevention of
general, precipitous declines in share prices also would be advisable.
Unfortunately, there is no reasoning or evidence that supports the
contention that required financial disclosure would play this role.
SUMMARY AND CONCLUSIONS
A number of reasons that would justify government interference with
the rights of individuals to write contracts -- specifically contracts
that would or would not require publication of financial information
by corporations to which people entrusted their resources -- were
The first reason concerns the implied contract that corporate officers
act in a fiduciary capacity towards shareholders (and to a lesser
extent, towards creditors). Self-serving actions by management and
false or misleading financial reports given to prospective investors
would constitute frauds which would involve mediation and redress
by government. Consequently, government might find it more efficient
to require financial disclosure should this requirement prevent or
reduce fraud and misrepresentation. An examination of the extent of
fraudulent financial statements before disclosure was required by
the Securities Acts revealed very little evidence of this practice.
Rather, it appears that the fraud that existed occurred primarily
before the early 1920's, aside from a few highly publicized cases.
This is not to say that the late 1920's were free from fraudulent
financial statements. But then neither are the post-SEC, required
disclosure years, as a reading of recent newspaper stories shows.
Indeed, there is some reason to believe that SEC regulation has tended
to make investors too accepting of financial statement data even though
these data cannot provide unambiguous measures of a company's economic
position and activity. Further, by insisting on conservative practices
(such as historical costs) and rigid rules the SEC has allowed managers
to manipulate net income while appearing to follow generally accepted
The second reason analyzed is "fairness" in dealings among
people. One operational definition of this concept is that one should
be able to purchase or sell a company's securities with the knowledge
that the economic effect of information about the company is impounded
in the market price of its shares. In this event, the market is a
"fair game." A related definition is that the ordinary shareholders
should expect that share prices are not manipulated. The evidence
on the random character of share price changes in the years before
and after passage of the Securities Acts and in the United Kingdom
(where much less disclosure is required than in the U.S.) is consistent
with the belief that the market is a "fair game" with and without
required disclosure. The operations of the manipulative practices
of "pools" also appear to have no relationship to the extent
of financial disclosure.
Improved efficiency of resource allocation is the third reason postulated.
Despite the belief of those who enacted and administer the Securities
Acts, there is little (if any) evidence to support the contention
that financial statements provide useful information to investors.
Statistical studies reveal little relationship between share price
changes and the publication of financial statements. This lack of
relationship may be due to (1) misspecification of the statistical
model or mismeasurement of the variables, (2) to the poor measures
of the economic condition and progress of companies reported in financial
statements, or (3) to investors having knowledge from other sources
of the data published before the statements were released. The first
explanation is always possible, but has not been shown to be a serious
problem by critics. With respect to the second explanation, the SEC
has done little to improve accounting measurements and reports. The
third explanation was tested and supported by a number of studies,
particularly one which found that the 1934 requirement that sales
be published had no discernible effect on the share prices or measured
riskiness of companies who published the amount of their sales for
the first time. Thus it appears that the third explanation, which
does not support required disclosure, is the most likely one.
More efficient administration by government and employees' and the
public's right to know is the fourth reason considered. Though no
studies on this subject have been made, casual observation indicates
that government would obtain required data more efficiently with stratified
random sampling than by requiring continuous reporting by all corporations.
Though it is debatable whether corporations have the same obligation
towards employees as toward shareholders, it nevertheless is doubtful
whether published financial statements would provide employees with
useful information. In general, the data required to determine future
employment possibilities, the maximum amount that could be won in
wage bargaining, etc., is more specific and requires better estimates
than are provided by the data presented in the statements. Finally,
the public may have a general right to know how corporations are faring,
but the information that can be provided by financial statements is,
at best, crude and possibly misleading.
Political considerations is the last reason given for requiring financial
disclosure. Legislatures and government administrators generally are
motivated to take some action when a scandal is sufficiently serious
to concern the general public. Therefore, those who oppose government
interference in private affairs should consider that some sort of
required disclosure is preferable to alternative laws (such as "merit"
legislation), if disclosure prevents or mitigates the degree of scandals,
or at least, gives the impression that no other general governmental
action is desirable.
Thus, few of the often proposed reasons for government required disclosure
are supported by the analysis. The only benefits that are supported
are audits that prevent or reduce fraud and scandals and possibly
the right of the public to some information about the operations of
corporations. However there is no evidence that required disclosure
as compared with voluntary disclosure provides the public with information
that is more useful for the efficient allocation of resources in the
Against these benefits one must consider the costs of required disclosure.
Corporations, and hence shareholders, must incur the expense of producing,
printing and distributing a large amount of data that provide few,
if any, benefits to investors but may benefit competitors. The burdens
also are relatively heavier on small than on large corporations.(10)
A CONCLUDING COMMENT
Assuming that the costs of required disclosure exceed the benefits
to the public as a whole, would repeal of the disclosure statutes
mean the absence of meaningful financial statements? The evidence
indicates otherwise. Before passage of the Securities Exchange Act
of 1934, all companies listed on the New York Stock Exchange (NYSE)
published balance sheets and income statements. Ninety-four percent
were audited by CPAs. Sixty-two percent disclosed sales, fifty-four
percent cost of goods sold, and ninety-three percent depreciation.
(At this time 70 percent of securities transactions took place on
the NYSE). Though the extent of financial disclosure was increasing
over time, the quantity of data made public proba-bly would not have
been as great as is now required by the SEC. The cost of producing
and distributing the data also would be lower. But there is no evidence
or reason to believe the benefits derived from the smaller amount
of data disclosed would be less than are garnered under our present
laws and regulations.
*I gratefully acknowledge
the helpful criticism of my colleagues, especially Jerold Zimmerman.
(1). See George J.
Benston, "The Effectiveness and Effects of the SEC's Accounting
Disclosure Requirements," in Economic Policy and the Regulation
of Corporate Securities, Henry G. Manne, Editor, American Enterprise
Institute for Public Policy Research, Washington, D.C., 1969, pp.
515-32 for references and a more complete discussion.
(2). Ray Garrett,
Jr., "New Directions in Professional Responsibility," The
Business Lawyer, Vol. 29 (March 1974) p. 9.
(3). The literature
on which this very brief summary is based is reviewed in George J.
Benston, "Corporate Financial Disclosure in the U.K. and the U.S.:
A Comparison and Analysis," unpublished manuscript, Chapter 4.2.5.
References to some 33 articles are given there.
(4). See George J.
Benston, "Required Disclosure and the Stock Market: An Evaluation
of the Securities Exchange Act of 1934," American Economic Review,
Vol. LXIII (March 1973), pp. 132-55.
(5). See Benston,
ibid., among others.
(6). See Benston,
"Corporate Financial Disclosure in the U.K ... .," section 4.2.6 for
a review of this literature and for references thereto.
may be obtained that is used for other than decision to buy, sell
or hold the shares of the reporting corporation: unfortunately no
studies have been published that test this hypothesis.
(8). See ibid., section
4.2.3 for a review of this literature and for references thereto.
(9). See Benston,
"Required Disclosure and the Stock Market ... " for details.
(10). See Benston,
"Corporate Financial Disclosure in the U.S. and the U.K ... ,"
op. cit., Chapter 3 for a more extended analysis.