2(b). Reliability and Neutrality, including Conservatism and Volatility
As part of its oversight activities, the Oversight Committee of the Financial Accounting Foundation interviewed and requested written comments (collectively, "the interviews") from thought leaders among the FASB's constituencies. There were 107 interviews in total, including 12 with representatives of financial statement users and 17 with regulators (a special class of financial statement users). [FASOversight, p. 1]
While the interviews were not designed to elicit criticisms of financial reporting, in general, or to identify the needs of users of financial information, interviewees did comment on those matters. [FASOversight, p. 1]
Following is a summary of the principal comments received [on the subject] from users and regulators relating to criticisms of financial reporting. . . . [FASOversight, p. 1]
The expected financial reporting result of a proposed transaction often influences management behavior; financial reporting should report the results of economic decisions, not drive such decisions. [FASOversight, p. 1]
__________
The standard-setting process should be even-handed. The process (and the pronouncements resulting from it) should be free from bias and designed to withstand pressures, political and otherwise, from outside parties. [RMA90, p. 2]
Understandability is an important characteristic of accounting data. The [following item] listed below [is] vital to understandability. [RMA90, p. 3]
Conceptual support: Financial accounting standards should emanate from a sound conceptual framework that ensures consistency, comparability and neutrality. [Also included in 2(c)] [RMA90, p. 3]
The APC [Accounting Policy Committee] has considered and expresses below its opinions on a number of specific issues affecting financial accounting standards and financial reports. The APC believes that the following items should be included in the single body of accounting concepts, standards, principles and methods: [RMA90, p. 5]
Accruals and deferrals are necessary for proper matching to occur, and by their nature deferrals, which require allocations, and accruals, which require estimates of future expenditures, are arbitrary and imprecise. Therefore, care must be taken to see that their use not be extended to permit "normalization" of earnings in any accounting period, annual or interim. Normalization, like forecasts and projections, is the province of the financial statement user and should not be incorporated into financial reporting. [Also included in 5(a)] [RMA90, p. 5-6]
Conservatism is a doctrine that serves users of financial statements well and should be observed consistently by financial statement preparers. It is difficult to define, but its spirit is contained in the following two statements: (1) "Recognize all losses when they occur, but do not recognize gains until they are realized."; (2) "When in doubt, err on the side of undervaluing assets and overvaluing liabilities." Conservatism is, of course, antithetical to the notion that accounting should be even-handed and free from bias. [RMA90, p. 6]
__________
[C]redibility is . . .the most serious problem for the annual report, in that investors often distrust what companies tell them in their annual reports. [HILL KNOWLTON, p. 6]
More than nine out of ten of the professional investors polled -- 92 percent -- said that annual reports all too often fail to candidly discuss bad news, and problems, and what management is doing to solve them. [HILL KNOWLTON, p. 6]
[Seventy-three] percent of the individual investor [polled] agreed that annual reports often play down bad news or hide it in the back of the report. [HILL KNOWLTON, p. 6]
[Thirty-two] percent [of individual investors] agreed with the statement, "I don't trust what I read in annual reports." [HILL KNOWLTON, p. 6]
An Atlanta investor commented: "In general, bad news is disguised, and good news is overplayed." Added a Tampa investor: "I don't trust what I read in annual reports . . .because I believe that the company is putting themselves in a favorable light." [HILL KNOWLTON, p. 6]
A Baltimore security . . . analyst said: "There's not enough honesty in annual reports. Too often, the blame for mistakes is placed outside the company, and management won't take responsibility." [HILL KNOWLTON, p. 6]
"I look at the (Form) 10-K to make sure the annual report isn't full of lies. . . . (St. Louis brokerage firm analyst) [HILL KNOWLTON, p. 7]
"You know the company is in trouble when they show 20 or 30 photos of children at play and no footnotes [in their annual report]." (Hartford investment counseling firm analyst) [HILL KNOWLTON, p. 7]
"Write [annual reports] like they're being written for someone who owns the company. Make 'em frank and honest." (Chicago mutual fund analyst) [HILL KNOWLTON, p. 7]
__________
To all types of investors, the credibility of an annual report, or any other information source, depends on the degree to which it is correct, complete, and objective. [SRI, p. 59]
The credibility question is focused primarily on the front half of the annual report, the narrative part, which is subject to less rigorous scrutiny by regulators and auditors offers wide latitude to management on inclusion (or exclusion) and presentation of information. The back half, the financial statements and footnotes, is perceived to offer less latitude for presentation of information and is considered to be much more credible than the front half. [SRI, p. 59-60]
As shown in [the] table [below], investors believe that many company managements are not forthright when reporting problems and poor company performance, that much of the information they disseminate is too "promotional," and that troubled companies take great pains to convey the impression that they are not seriously troubled. The investors do not believe that management tells outright lies in its information reporting except in unusual cases. They do believe, however, that executives strive to present their situations in the best possible light, delay reporting negative information in the expectation (hope?) that the situation will soon be corrected, or are simply so involved with the company that they report biased information without realizing it. Very few investors, however, doubt the integrity of corporate management, yet, for the reasons outlined above, most believe that corporate reporting is not objective. It is this perception that makes "objectivity" one of the components of the value of investment information. [SRI, p. 60]
Individuals believe that the credibility of annual reports is analogous to that of advertising. Even though advertising tells the literal truth, certain ads can mislead through omissions, half-truths, out-of-context statements, and the like. When times are good for a company, they believe that the firm's annual report is highly credible. When the company has problems, they seek objective corroboration or the advice of competent analysts. This questionable credibility is one reason that individual investors (and professionals as well) use so many different sources of information. [SRI, p. 60]
Questionable credibility is not perceived as a problem for most investors, however; they expect it, and they find ways to overcome it. They recognize the natural biases of management and subjectively make appropriate adjustments. They rely heavily on other information sources for decision making. The professionals, especially, are confident of their ability to recognize exaggerations, favorable accounting treatments, and significant omissions. [SRI, p. 60]
Even though investors do not regard questionable credibility as a major problem, issuers of annual reports should take it seriously for three reasons. Firstly, management exacerbates rather than enhances its image by glossing over problems and poor performance. A noncredible annual report raises the question of management credibility in general, and can even cast a shadow on management competence. Despite some short-term discomfort, full disclosure will, in most cases, actually enhance management credibility. Secondly, the professionals are irritated that some executives seem to regard them as gullible enough to believe that all is well when in fact all is not well. Finally, those few companies whose annual reports are considered to be highly credible (Berkshire Hathaway and Quaker Oats are frequently cited examples) earn high marks and expressions of great respect from professional investors. [SRI, p. 60&62]
The Credibility of Annual Reports
Individual Investors Professional Investors Statement Agree Neutral Disagree Agree Neutral Disagree "Most annual 15.4% 41.3% 36.3% 13.8% 41.0% 45.2% reports are candid in their discussion of company performance." "Annual reports 86.4 11.1 2.2 85.6 11.2 3.2 are writtento project the most favorable impression of company management." "Annual reports 84.9 10.9 2.5 88.1 8.0 3.8 would be more useful to me if the management discussion was frank about reporting poor company performance." "Annual reports 44.8 32.9 21.7 - not asked - would be more useful to me if they were less promotional."
Source: SRI International survey, 1986. [SRI, p. 61]
__________
[Sometimes,] earnings quality . . . seem[s] to be related to "representational faithfulness," and management's forthrightness in disclosure. For example, one analyst reported that an extreme drop in the reported tax rate of a company "caused some to doubt the quality of (its) earnings". Another expressed concern about earnings quality on the basis of the amount of costs included by a company in the determination of cost of goods sold. [Also included in 1(b), 1(c), and 5(a)] [PREVITS, p. 16]
___________
[Context] The AIMR position paper provides a two-paragraph summary of the section (pages 20-23) entitled "Qualitative Characteristics of Financial Statements." The second paragraph pertains to timeliness of information:
Some attention is paid in this section to the need for timely reporting. It introduces the view of AIMR that mandated quarterly reporting not only is essential, but that moves to abolish it appear to be based on incorrect premises, blaming quarterly reporting requirements for "short-termism" when the blame can better be placed elsewhere. [Also included in 2(a)] [AIMR/FAPC92, p. vii]
[Context] Because they have that focus, excerpts on timeliness, which normally would be included in 2(a)-Relevance (of which timeliness is a subset in FASB's concepts statements), are included in 11(a)-Frequency of interim reporting.
The first paragraph is as follows:
The qualitative characteristics of accounting that we find most important to the needs of financial analysts are relevance, reliability, both verifiability and representational faithfulness, timeliness and neutrality. First, analysts need to know economic reality -- what is really going on -- to the greatest extent it can be depicted by accounting numbers. The information must be relevant to the process of analysis, one reason why much space in the early part of the report is devoted to describing the analyst's work. [Also included in 2(a)] [AIMR/FAPC92, p. vii]
[Context] It introduces the following excerpts on reliability and neutrality:
There is general agreement that accounting and other financial data should have certain characteristics. The FASB's Statement of Financial Accounting Concepts No. 2, "Qualitative Characteristics of Accounting Information," creates two groups of these characteristics under the headings "relevance" and "reliability." That grouping is appropriate because in many cases the format and content of accounting data requires a trade-off between the two. Certainly, financial analysts desire information that is both relevant and reliable, but their bias is towards relevance. In a phrase, analysts prefer information that is equivocally right rather than precisely wrong. Inexact measures of contemporaneous economic values generally are more useful than fastidious historic records of past exchanges. A short discussion of several characteristics of accounting quality and our views of them follows. . . . [Also included in 2(a)] [AIMR/FAPC92, p. 20]
Reliability: General
The two primary components of reliability are verifiability and representational faithfulness. The former refers to the likelihood that different accountants, availing themselves of the same evidence, will draw similar conclusions. The latter refers to the likelihood that the accounting measure depicts accurately the nature of the object being measured. [AIMR/FAPC92, p. 21]
Reliability: Verifiability
This characteristic is intimately related to the attest function. For financial reports to be useful, they must be trustworthy. The report of the independent auditor is essential. The auditor however can verify only that which can be documented or confirmed. Perhaps that is one reason for the extensive amount of detailed guidance provided with current accounting standards. As the standards-setting process has infiltrated areas in which the measurements are less than precise (pensions and other postemployment benefits, financial instruments, recognition of fee revenues, etc.) the rules have become more detailed. Detailed rules may also be perceived as necessary to serve the needs of both financial statement preparers and their independent auditors. Verifiability implies that two unrelated parties considering the same facts independently will draw similar conclusions. It is possible that detailed rules are now the only way to inculcate verifiability into measurements that otherwise are subject to honest differences of opinion. Can better ways be found? We hope so and are heartened by the issuance of FASB Statement 109, "Accounting for Income Taxes," which we regard as a step in the right direction. [AIMR/FAPC92, p. 21]
Another aspect of verifiability is knowledge of its absence. Most accounting numbers have an appearance of precision. But, other than contemporaneous exchanges involving cash, accounting numbers are determined by estimates of various degrees of inexactitude. Analysts need to know how indefinite those numbers are and they need to know the degree to which the same economic event or condition could have been reported differently using alternative measurement methods. More information of that sort incorporated in financial reports would be exceedingly welcome. [Also included in 9] [AIMR/FAPC92, p. 21]
Reliability: Representational Faithfulness
Assets and liabilities are probable future economic benefits and claims against those benefits, and users of financial statements expect to see them depicted accurately. There are two aspects to representing them faithfully. One is to select the appropriate attribute to measure; the other is to measure it accurately. There are too many examples to cite them all, but one may be instructive. [AIMR/FAPC92, p. 21]
Under current accounting practice, intangible assets are recorded at cost only when they are purchased from another entity, either separately or as part of a business combination. The effect is that self-developed intangibles are not recorded at all or at the nominal amounts spent to assure monopoly rights. Furthermore, the costs of both purchased and self-developed intangibles are amortized over arbitrary future time spans, even though their value may decrease in some other pattern or, in many cases, increase as the enterprise makes additional expenditures to maintain and/or enhance their value. Those accounting practices cause severe noncomparability between and among companies. [AIMR/FAPC92, p. 21-22]
Also, regardless of whether intangibles are recorded at the cost of purchasing them or at the nominal amounts to develop them internally, many of the future benefits to be obtained from them are more speculative and conjectural than those to be received from tangible assets which at least may have some value in alternative use. So, only at the date on which purchased intangibles are acquired do the financial statements assuredly reflect amounts that can purport to be representationally faithful of economic reality. Moreover, there may well be no accounting measure that is capable of expressing well over time that the sole economic benefit of intangible assets is their potential contribution to the future cash flows of the enterprise. Our specific recommendations for accounting for intangible assets are discussed later in this report. [AIMR/FAPC92, p. 22]
Neutrality
In addition to timely dissemination, fairness also requires neutrality, presentation of data that are without bias. Investors both buy and sell securities. Financial reports should inform both sides of a transaction in such a way that neither is favored. Much of what applies here was discussed above under the heading of relevance. Historic costs, even more so lower of cost or market procedures, tend to introduce bias in favor of buyers of securities by suppressing good news while revealing the bad straightaway. The absence of adjustments to reflect price changes, even as supplementary information only, in North American accounting standards institutes a bias that varies in proportion to: (a) the rate of price change; (b) the dispersion of those changes among the various goods and services traded; and (c) the holding period for assets whose prices change. [AIMR/FAPC92, p. 23]
__________
[Context] Meeting of the Investor Discussion Group on January 13, 1993. Part of the meeting was devoted to the topic of value information. During the discussion, comments were made on the reliability of value information.
Committee/Staff/Observer
[Participant I-12] said that part of determining fair value depends on what your expectations are as to how you are ultimately going to realize the asset or settle a liability. Even if you want to adjust the information you receive, why wouldn't you want to know what management's expectations are with respect to fair value? [Also included in 4] [TI 1/13, p. 16-17]
Participant I-12
Management lies all the time. [Also included in 4] [TI 1/13, p. 17]
Committee/Staff/Observer
So it's a reliability issue? [Also included in 4] [TI 1/13, p. 17]
Participant I-12
Yes. [Also included in 4] [TI 1/13, p. 17]
Participant I-10
Have you ever seen a management who thought that their stock was overvalued? With fair value, you're giving them a platform to induce people to believe that there is an enormous gap between what the market price is and what they believe the business is worth. I think that plenty of room would be given for deception. Not everyone is a professional investor; there are millions of people who believe what they read! It's somewhat dangerous. [Also included in 4] [TI 1/13, p. 17]
Participant I-15
Going back to management's perceptions. When you look at some of the opportunities that companies take, when they realize a large gain, to do these restructuring charges and write-off assets, it shows you that management is often shortsighted and unreliable. You can't believe what management tells you many times. [Also included in 4] [TI 1/13, p. 18]
Participant I-12
Going back to the question of why don't we just use market value? The market is nothing more than a value at a point in time. All the market is is collective judgments as to the fair market value at a point in time. I'm saying that those judgments have a lot of unreliability in them. Is [name deleted] really worth $8.50, the price it closed at at the end of 1991, or is it worth $21.50? Those differing prices simply reflect the collective decision-making of analysts. I'm not convinced that that's an appropriate benchmark that gives us the kind of reliability and stability that we need to make our own judgments about what the market might do. We're getting into some circular logic here that bothers me a lot. [Also included in 4] [TI 1/13, p. 18]
Committee/Staff/Observer
Last question on fair value, which deals with the volatility or "noise" issue in reported earnings. First, do you agree that fair values in financial statements would introduce unhelpful noise in reported earnings? Anybody think it will not introduce unhelpful noise? The answer seems to be no. Next question: would your answer to this question differ depending on how the impact of fair value measurements is reported in the income statement? [Also included in 4] [TI 1/13, p. 22-23]
Participant I-7
And reliable? [Also included in 4] [TI 1/13, p. 23]
Committee/Staff/Observer
Take the reliability issue off the table. [Also included in 4] [TI 1/13, p. 23]
Participant I-11
I don't think you can separate those issues. My complaint is this notional fair value is fundamentally an unreliable number. What is the fair value of the [large commercial office building]? [Also included in 4] [TI 1/13, p. 23]
Participant I-8
If I am not mistaken, there is already a distinction made in the income statement in the way in which accounting is done for earnings of foreign companies depending on whether the currency is considered to be stable or the "banana republic" type thing. The accounting profession is saying that in the latter case, the accounting will be done a specific way because the currency fluctuations should stay within a specific band. [Also included in 4] [TI 1/13, p. 23]
Committee/Staff/Observer
There is no interest for running value changes through the income statement. I'm wondering what your reaction is to the accounting in the pension arena when value changes are in effect spread to eliminate volatility. That's kind of a compromise in the market value arena; is that good or bad? [Also included in 4 and 9] [TI 1/13, p. 23]
Participant I-7
You're not in the business of putting businesses out of business. In some instances, if you didn't spread, you would really create a problem. [Also included in 4] [TI 1/13, p. 23]
Participant I-8
I think it's good; it reflects the realities of the world and to that extent it's good. The real question is whether the actuarial assumptions are valid or not, not the interim fluctuations in the assets that happen to be held at that moment. [Also included in 4 and 9] [TI 1/13, p. 24]
Participant I-12
I have a problem with the actuarial assumptions. We all know of companies that are still using 7-10% accumulation rates. This comes back to the notion of reliability. I don't have a problem in trying to reflect in some manner the cost of employee health care benefits; on the other hand, what happens if we socialize medicine and get deflation? [Also included in 4 and 9] [TI 1/13, p. 24]
Participant I-5
The question of reliability is fine and good, but the fact is the present historical book value that is recorded is significantly less reliable than someone's best guess of fair value today in 95% of the cases. For example, [name deleted] gets an appraisal every year; $1.8 billion 4 years ago, then $1.7 billion, then $1.6 billion. Meanwhile, the bond is trading as if it's worth maybe $700 million. If the company shows it in the balance sheet at historical cost of $600 million and didn't tell you about the $1.8, $1.7, and $1.6 billion, what is the best measure? I think the best guess of what someone says it's worth today is valuable to have relative to what the cost was in 1936. [Also included in 4] [TI 1/13, p. 24]
[Context] Meeting of the Investor Discussion Group on January 13, 1993. Part of the meeting was devoted to the topic of disclosure about operating opportunities and risks. During the discussion, an investor answered a question on reliability.
Committee/Staff/Observer
Before you answer that, the next question we have is right on that point. The second framework that we talk about in the materials is the SEC's MD&A requirements. In MD&A, the discussion and analysis of results of operations is to focus on events and uncertainties known to management that would cause the reported information to not be a good indicator of future operating results. It is also to describe known trends or uncertainties that have had or that management expects will have a material impact on net sales or revenues or income from continuing operations. Our question is: do those MD&A requirements provide a workable framework for categorizing and disclosing what you need to know about operating opportunities and risks? Is it a promising starting point or basis from which to develop an accounting standard requiring disclosure of information about operating opportunities and risks? And there is also [committee/staff/observer]'s question; do you believe what you get? [Also included in 10(b) and 13] [TI 1/13, p. 48]
Participant I-7
Reliability is in the mind of the issuer. I think it goes beyond reliability. There are certain managements that are "ept" and others that are inept. So when you read the MD&A or have a discussion with management, for the most part, they're trying to give you as reliable information as they possibly can. But within the context of a competitive environment, some are being inept. [Also included in 10(b) and 13] [TI 1/13, p. 48]
For example, management thinks the company is going to have a 10% sales increase this year in the motor industry; 6% increase in units and 4% increase in sales price. The statement is absolutely true until you go out into the marketplace and find that 25% of the business is distributor-related; so the distributor will also increase its price by 4%, but salesmen of the 75% segment of the business will be under pressure to get the price increase down to as close as 1% as possible. So 10% is going to be wrong if you're setting up your cost structure on that basis. [Also included in 10(b) and 13] [TI 1/13, p. 49]
[Context] Meeting of the Investor Discussion Group on March 17, 1993. Part of the meeting was devoted to the topic of auditor involvement. During the discussion on the scope of auditing, an investor made a comment about reliability of information.
Participant I-12
The MD&A may also be an area that would be more appropriately handled by the SEC, in terms of what kinds of things belong there. I agree with the notion of having an independent opinion of internal control and information systems because that leads back to the reliability of the information that we're getting. Also, the auditor could play a role in the MD&A in terms of bringing out environmental changes and the company's exposure to those changes. And perhaps greater discussion of assumptions that underlie all the numbers. [Also included in 9 and 17(b)] [TI 3/17, p. 10]
[Context] Meeting of the Investor Discussion Group on March 17, 1993. Part of the meeting was devoted to the topic of conservatism, volatility, reliability, and neutrality.
Committee/Staff/Observer
Question 8 wraps up our discussion of auditor involvement and relates to the concept of conservatism. Let me address the question in parts. Parts (a) and (b) talk about attributing a higher quality of earnings and a better multiple if a company uses conservative accounting practices. First question: is accounting conservatism widely used in financial analysis as a key factor in assessing the quality of earnings of a company? [TI 3/17, p. 29]
Participant I-12
I think there's a lot of lip service applied to quality of earnings. I heard analysts tell me that the [name deleted] had very conservative accounting principles 6 months before they blew up. As analysts, we have a spiritual search for high quality earnings and, in many respects, we use that to justify a position to buy or sell a stock when that may or may not be the reality. I have a problem with this issue of quality of earnings. [TI 3/17, p. 29]
Committee/Staff/Observer
On page 9 of the meeting materials, we give 5 items that describe conservatism.
[1. Conservatism means to anticipate no profits but anticipate all losses
2. Conservatism means providing for losses when profits are high so that a company's resistance to recording losses in low-profit periods will not result in overstated assets or understated liabilities
3. Convservatism offsets the natural over-optimism of managers who tend to view the expected operations of their company through rose-colored glasses
4. Conservatism makes it likely that possible errors in measurement will be in the direction of understatement rather than overstatement of net income and net assets and thus future surprises are likely to be pleasant
5. Conservatism means that the uncertainties that inevitably surround many transactions should be recognized by exercising prudence in preparing financial statements but does not justify creation of secret or hidden reserves.]
Which one(s) fits your definition? [TI 3/17, p. 29]
Participant I-16
The one I prefer is 5. However, what financial analysts really like is a company where you can take a straight edge and describe the trend in earnings. That is one problem I have with conservatism; it sounds like a nice thing but the more conservative you are, the more leeway you have to manipulate the trend. I'm always wary when accountants are putting in a change in accounting principles that appears to be more conservative because my suspicion is that there might be more room to manipulate the trend. [TI 3/17, p. 29]
Participant I-11
I agree. [TI 3/17, p. 29]
Committee/Staff/Observer
Still on conservatism, question (d) talks about chosing between two concepts: conservatism and neutrality. We note that conservatism conflicts with reliability and neutrality which are two key qualities that make financial information useful. Financial reporting cannot be reliable and neutral and at the same time conservative. Our question is: do you accept that reliability and neutrality are more important qualities than conservatism? Or do you think that accounting should emphasize conservatism to avoid reporting values that may not be realized even if their realization is likely? [TI 3/17, p. 30]
Participant I-11
As long as you're dealing with estimates, I'm not sure that there is such a thing as neutrality. It's a principle that I think is good but, in fact, you're making judgements in making those estimates and they are subjective and you can't be neutral. The auditor's responsibility lies in making his professional judgement about those estimates; probability and magnitude are things that enter into that judgement and each one has to be dealt with in light of the particular circumstances. My bias is that if you have to err, err on the side of conservatism because it does less harm. [TI 3/17, p. 30]
Committee/Staff/Observer
[Participant I-11], if everybody agrees that a certain outcome is most likely, it's the best guess, but there is a more conservative outcome that is at least reasonably possible but not the best guess, you're not advocating going to the most conservative guess? [TI 3/17, p. 30]
Participant I-11
No, the best guess and I want the auditor to tell me what the best guess is without bias whether it's conservative or liberal. But if it's 50/50, then let's go with the conservative. [Also included in 17(c)] [TI 3/17, p. 30]
Participant I-7
Reliability and neutrality are the most important qualities for me, much more important than conservatism. [TI 3/17, p. 30]
Participant I-16
Our job is to put a value on an enterprise, not to make sure that a company won't go out of business; that's not what investors are trying to do. Conservatism is putting a floor estimate on the company's earnings, cash flows, and value. It's highly unlikely that the true value is below that number. For most purposes, that would not be a useful number because most people using financial statements are not trying to come up with a floor number. Most of us are trying to get to a realistic estimate of value and efficient capital markets require that. [Also included in 1(a)] [TI 3/17, p. 31]
Participant I-12
There has to be a balance between the two. I don't want a company ignoring reality in the interest of conservatism. We have all seen companies that are having terrific results through reasonably generated numbers, not flaky accounting, and the CFO decides that rather than growing 20% this year, they'll grow 16%, and sock earnings away in reserves for future years. As analysts, we will go back and make adjustments if we know that this is happening. Similarly, if things are horrible, I want to know they're horrible. There has to be a balance between the two principles and I can't say that I prefer one over the other. [Also included in 1(c)] [TI 3/17, p. 31]
Participant I-16
One example of the use of conservatism is the extent to which a company writes down physical or intangible assets or recognizes expenses that will really benefit future periods. Companies get away quite often because it's viewed as conservative but I would like to see when that happens the auditors give some recognition of the fact that the company is boosting its future reported earnings. I can recall a company that acquired a business from another company and did not acquire the brand name; as a result, as part of the cost of the acquisition, they wrote off the next two years of advertising. They said that they were being conservative in writing down the value of what they had acquired, but what happens after two years when you have to start expensing for advertising? Is that really conservative? [Also included in 17(c)] [TI 3/17, p. 31]
Committee/Staff/Observer
Page 10(e) of the meeting materials presents two statements that summarize what we said about conservatism.
[ Conservatism should not connote deliberate understatement of assets or overstatement of liabilities. Nor should financial reporting attempt consistent understatement of income, which in any event is impossible to achieve because decreasing income of one period inevitably increases income of a later period or periods.
Conservatism is a prudent reaction to uncertainty to try to insure that uncertainties and risks inherent in business situations are adequately considered. For example, if two estimates of amounts to be received or paid in the future are about equally likely, conservatism suggests using the less optimistic estimate. However, if two amounts are not equally likely, conservatism does not dictate using the more pessimistic amount rather than the more likely one. Neither does it require deferring recognition of income beyond the time that adequate evidence of its existence becomes available nor justify recognizing losses before there is adequate evidence that they have been incurred.]
Do you agree with the two paragraphs? If not, why? [TI 3/17, p. 31]
Participant I-11
I agree; those two bullet points put it very well. [TI 3/17, p. 32]
Participant I-12
Yes. [TI 3/17, p. 32]
Participants I-7 and I-16
Yes. [TI 3/17, p. 32]
Committee/Staff/Observer
Question 9 deals with volatility of earnings and the question is in parts. Parts (a) and (b) say that it is our understanding that investors attribute a higher quality of earnings and a better multiple if a company's earnings shows a stable trend and a lower quality of earnings and multiple if it shows variability and volatility. Our first question related to volatility is: do we have a reasonable understanding of the way stability, variability, or volatility affect the quality of earnings in financial analysis? Further, is stability, variability, or volatility of earnings widely used in financial analysis as a key factor in assessing the quality of earnings of companies? [TI 3/17, p. 32]
Participant I-12
I'd like to start with the second part first. Yes, it's widely used; we all look at the variability of earnings. There are cyclical and secular factors; I'm not convinced that higher and lower multiples necessarily are attributed to companies with more stable earnings. If you analyze multiples across the S&P 500, the highest multiples are on the companies with the worst earnings, and it's purely a mathematical exercise. At a cyclical bottom, the multiple on [name deleted] earnings is going to be much higher than it is at a cyclical peak, and that's just because there is no earnings. The market is anticipating future earnings and is assigning a normalized multiple to those earnings. As analysts, we would tend to attribute a more stable multiple. The range of multiples assigned to a [one company] will be vastly different from a range of multiples assigned to [other companies]. So variability of earnings is something we look at a lot and it is important to us. [TI 3/17, p. 32-33]
Participant I-16
Clearly, one pays a premium for stability because it is presumed to be an indicator of lower future risk and uncertainty, thus, it should get a higher valuation. I think the market has gotten a little more sophisticated in viewing stability and the evidence for that would be the low multiples for more diversified businesses versus less diversified businesses; in the latter, you can understand whether reported volatility is reality much more easily in a one-product business or one-industry business, like [name deleted], than would be the case with a highly diversified company. A lot of large diversified companies have broken up, cognizant of the fact that the market penalizes companies if you can't understand how that trend comes about. Just showing a nice trend that investors don't believe represents reality will not provide a value as high as about 10 years ago. [TI 3/17, p. 33]
Committee/Staff/Observer
The last question about volatility is: what problems related to volatility do you have with the information you receive from external financial reporting? The meeting materials on page 11 identified 4 possibilities. [Also included in 4] [TI 3/17, p. 33]
Participant I-12
Fair value accounting. Running changes in the value of a bond portfolio through the income statement is going to make that statement incredibly volatile, and it may be a faked volatility because those quarterly gains or losses may or may not be realized. For example, anybody who sold their stocks November 1, 1987 probably realized substantial losses; anybody who waited 6 months probably made out just fine. So the realization of gains and losses is vastly different from the paper effect. Fair value accounting would just make the volatility of earnings that much worse. [Also included in 4] [TI 3/17, p. 33]
Participant I-16
Does that make it worse or does that just recognize the reality? [Also included in 4] [TI 3/17, p. 33]
Participant I-12
What's reality? [Also included in 4] [TI 3/17, p. 34]
Committee/Staff/Observer
I don't want to get into an argument but I could argue that the November 1 person holding the securities as opposed to the person who sold those securities is not presenting a very correct balance sheet. I'm not talking about how to handle that in the income statement, that's a different issue. [Also included in 4] [TI 3/17, p. 34]
Participant I-12
But in terms of volatility, that would introduce volatility in the income statement that wasn't there before. [Also included in 4] [TI 3/17, p. 34]
Committee/Staff/Observer
Assuming you put the unrealized November 1 loss in the income statement. [Also included in 4] [TI 3/17, p. 34]
Participant I-7
For me, earnings volatility has nothing to do with earnings quality for the companies I follow. In terms of measuring quality, if I know that the company is following good accounting procedures, earnings volatility has nothing to do with earnings quality. [TI 3/17, p. 34]
Committee/Staff/Observer
We've talked about the quality of earnings; how would you define quality earnings? We've talked about the concept of core earnings, which I would define as the earnings to which a multiple greater than one is applied. Can any analogy be drawn between core earnings and quality earnings? Could I think about quality earnings in terms of a multiple? Are good quality earnings earnings that you apply a multiple greater than one to, lower quality earnings something you apply a multiple of one to? [Also included in 5(a)] [TI 3/17, p. 34]
Participant I-16
There are two ways of looking at it. One is the conservatism aspect; for example, companies using accelerated depreciation using the same useful lives as another company using straight-line, are clearly more conservative and are perceived as having better quality of earnings. The second aspect is predictability and stability. If you believe a company can report earnings of at least that much in the next year, it's worth more than if you have no idea. For example, if I had the earnings of [name deleted] for one year and ask how much it's worh, I wouldn't have any idea because I don't know whether they made any money in the prior year and whether they would make any money in subsequent years. [Also included in 5(a)] [TI 3/17, p. 34-35]
Participant I-7
One of the problems I have in answering that question is that earnings quality is only one aspect entering into the valuation of a company. For me, earnings quality is only one measurement of valuation. [Also included in 5(a)] [TI 3/17, p. 35]
Participant I-12
I would focus on the concept of earnings quality equals predictability. For example, [name deleted] is considered among the highest quality in the brokerage business, a highly volatile business. The company typically gets a substantial discount to the S&P multiple because they have a merchant banking operation where they periodically take gains. You take those numbers out and then look at the P/E and it gives you an entirely different perspective, because the market is looking at the predictable elements. I think of core earnings as operating earnings; the merchant banking part is not an operating business. What I assign a multiple to is the portion of the earnings where I have some ability to predict them. [Also included in 5(a)] [TI 3/17, p. 35]
[Context] Meeting of the Investor Discussion Group on March 17, 1993. Part of the meeting was devoted to the topic of databases. During the discussion, comments were made on the reliability of information provided by databases.
Participant I-16
I have spent most of my career as a generalist, which means that I have been working on companies that I haven't worked on for very long, so I don't have a database or history. So I use a database as a way to allocate my time. You can screen, using 10 years of financial statement numbers, an enormous number of companies in a couple of minutes, perhaps pop out some anomalies that are worth investigating. I wouldn't buy a stock based upon a screen because I don't know if there is enough reliability. But if it can limit the universe and enrich the likelihood of finding something good in that universe, then it's a very useful screening device. I know there are some firms that use databases to make investments, but I think most of us use them for screening purposes. It helps narrowing down my search. [Also included in 1(c) and 16(a)] [TI 3/17, p. 48-49]
Committee/Staff/Observer
How do you know what's in databases is accurate or complete? [Also included in 16(a)] [TI 3/17, p. 49]
Participant I-7
You don't. [Also included in 16(a)] [TI 3/17, p. 49]
Participant I-12
Until you check it against the financial statements. If a database shows the assets of a company at $10 billion and the financial statements show total assets at $15 billion, it makes you wonder about the database. [Also included in 16(a)] [TI 3/17, p. 49]
Participant I-5
Unfortunately, a lot of them [database vendors] adjust [information] in different ways. You're not certain exactly how a database treats different items. As an analyst, it takes long enough to look at a company's financial statements for the year as they're presented; to take them as they are somehow massaged in a database, it does not help you much to get an opinion on a company. Where I found databases useful is in getting aggregate numbers; for example, the median cash flow coverage ratio for all single B credits. You can get that from a database without being that far off. [Also included in 1(c) and 16(a)] [TI 3/17, p. 50]
[Context] Meeting of the Creditor Discussion Group on February 2, 1993. Part of the meeting was devoted to the topic of value information. During the discussion, comments were made on qualitative aspects of external reporting.
Participant C-17
One of the difficulty with fair market values is they're so volatile. I that as additional information, it's helpful, because it gives you a reference point. Knowing what the spread is and getting some sense of fair market versus historical is important. I don't base my decisions solely on it. The thought that comes to me is sometimes, if I'm trying to choose between a secured and unsecured debt, for example, I may want to factor in how much capital support is really there. And I may be swayed to some extent by the reliability of the values that I see. You use fair values with a certain amount of prudence. [Also included in 4] [TC 2/2, p. 3]
__________
Participant C-13
The historical cost model seems to suit our requirements the best. But fair value information, providing you can satisfy yourself as to reliability, is important, particularly if there's a large disparity between fair and book value. The [name deleted] example is a good one on one side. The life insurance industry in the 1980's is a good example on the other side, where the much lower market values book indicated some pressures and strains on the industry. [Also included in 4] [TC 2/2, p. 5]
__________
Participant C-7
I guess we rely on historical cost because of consistency and the relative objectivity. I think fair value becomes an issue whenever you think that there's a variance; the degree of variance somehow correlates to your interest in fair values. It's when you see a major variance that you become interested. You want to abandon the historical cost concept, and then get into current value. [Also included in 4] [TC 2/2, p. 7]
Participant C-11
I would agree with all of these comments. I think that if we're talking about going concerns, the need for fair value information and its reliability and usefulness, in terms of knowing how well the business is doing, is lot less and definitely that puts it into supplemental status. I think we have a great problem in general as to knowing when a company is in distress, and when we have to take a different accounting approach. So far, all the comments have been focused on revaluing at market values specific types of assets. Nobody's mentioned liabilities. But I think we can't forget that. I want to make a comment that in an increasingly distressed situation, a company doesn't have to, necessarily, sell one particular type of loan or securities or whatever. There is often an option of selling part of its business. And so when you're talking about what is the fair or market value of an entity, it isn't necessarily just individual assets. It can be a business component. And the way you value the component of the company's business is going to be a lot different then. And it may be even more successful a way to take care of a distress situation than just selling its individual loans. I think if you're thinking about market value, you have to think in a more complex way and not just value the specific individual assets and liabilities and think you've done the job. I feel strongly about that. I'd also make just a general comment about supplemental information. I don't ascribe more importance to something because it's in a footnote, as opposed to being in a supplemental schedule of some sort. We have all kinds of supplemental schedules that are required and that's where you can get some of your best data. As a user, I don't have a phobia about needing to have it on the balance sheet or a footnote, per se. [Also included in 2(a), 4, and 5(d)] [TC 2/2, p. 7-8]
[Context] Meeting of the Creditor Discussion Group on March 11, 1993. Part of the meeting was devoted to the topic of conservatism, volatility, reliability, and neutrality.
Committee/Staff/Observer
Question 9 deals with conservatism, neutrality, reliability and volatility. The fundamental question is whether or not conservatism is something you seek or not seek in financial statements and how do you define conservatism for purposes of the analysis work you do? [TC 3/11, p. 39]
Participant C-5
I am probably even a minority in my own institution in saying that conservatism is a bane for me in terms of that whole approach, whether it's financial statements or just an approach to doing business. It is frustrating because I don't know the degree of conservatism or the degree of bias that's inherent in the financial statements. Things like inventory accounting methods where conservative inventory treatment is aggressive income statement treatment and vice versa. As far as neutrality, that's assumed and I haven't seen accounts that have been other than neutral. Volatility, there is absolutely nothing in the financial statements that lets me understand the volatility of the reported numbers. For example, I can have a company holding a rental property where loss of a key tenant representing 78% of the rental could change the whole value assumption whereas one that's a tenant at will that looks like it has no long term leases but they've been in there for 30 years, the building has got a real continuity to its value. It is important to understand what is the value but also to put the context around it to allow me to understand how volatile those assumptions are. [TC 3/11, p. 39]
Participant C-13
I think it's important to recognize conceptually that the investor in securities is confronted with an array of securities which he may buy, hold or sell or decline to buy. And if through some misperception of conservatism, either in the presentation and valuation or in some kind of smoothing of earning sense, the investor doesn't get a true picture of what the company operations are, then the financial statements haven't fulfilled their obligation. My definition of conservatism would be much closer to number five here in your list (in the premeeting materials). [TC 3/11, p. 39]
Committee/Staff/Observer
With respect to number five would you seek that level of conservatism or does that frustrate you? [TC 3/11, p. 40]
Participant C-13
No, that level of conservatism is not necessarily frustrating. I think that's clearly the closest definition that you've got here for conservatism and I would very vigorously reject number two, for instance. [TC 3/11, p. 40]
Participant C-17
The operative words for me are really reliability and volatility. I want to know that the information is reliable. And the other part of it is when you see statements that show a great deal of volatility, I tend not to trust them. [TC 3/11, p. 40]
Participant C-13
Well then the reality is volatile. [TC 3/11, p. 40]
Participant C-17
Well, if the reality is volatile, then I tend not to trust management. [TC 3/11, p. 40]
Participant C-13
You trust the business, though? [TC 3/11, p. 40]
Participant C-17
I'm talking about sudden and dramatic changes in asset accounts or extraordinary write ups that occur from out of the blue or extremely complex footnotes that would take Solomon and twelve others to try to understand what it is they're really talking about because probably nobody really knows. [TC 3/11, p. 40]
Participant C-12
I like the C-4 definition which isn't real conservative. But it's been my experience that very strong companies tend to be this sort of conservative in reporting results and that less strong companies tend to be a little more aggressive, to push a little harder to keep up with the very strong companies. [TC 3/11, p. 40-41]
Committee/Staff/Observer
But of the choices, you'd prefer four? [TC 3/11, p. 41]
Participant C-12
Yes, I like four. [TC 3/11, p. 41]
Participant C-1
The way you phrased the question, it's an equity question. Because for me, a lot of the write-offs and reserves that are done are more to justify a bad year; let's dump everything we possibly can into it so that our earnings on a net income basis will improve next year whereas from my standpoint cash flow is more important. So I've got to go back and adjust for these reserves that were set up in 1992 for plant closings that are not going to happen till 1993. The other part is that conservatism relates to this concept of hidden assets on the balance sheet where you've got inventories and receivables that are based on cost and maybe should be also based a little bit more on market. Market value might be lower but, at the same time, the market value of fixed assets also based on cost could be dramatically higher. There's no way to really tell whether assets and/or liabilities have any type of reality to the true market value of the company. [Also included in 4] [TC 3/11, p. 41]
Committee/Staff/Observer
Let me ask you a follow up question with regard to inventory and receivables. If cash flow is the target, isn't there still a conservatism question, though, in terms of the valuation of inventory and receivables vis-a-vis the potential cash flow? And if so, how conservative an estimate of future cash flows do you want? [TC 3/11, p. 41]
Participant C-1
I'd rather have assets that you anticipate in some way liquidated over a short period of time, one year, to be extremely conservative. And I don't mean being hidden but I think that there are assets which are not truly short term assets that are put in that section. [Also included in 5(b)] [TC 3/11, p. 42]
Participant C-4
From a creditor's standpoint, conservatism is something that we would always like to see. Quite often, management will come to us and say that their numbers are conservatively stated. We will be much more liberal in applying our underwriting standards to a company that did consistently conservative reporting. Probably leaning towards number four here as far as our overall feelings about conservatism. [TC 3/11, p. 42]
Participant C-11
I basically don't like the word conservative at all because it could be misinterpreted in the way that others have just spoken of in terms of "if there's any question make it a very low number and end up having undervalued assets and ineffective reserves". The word realistic or relevant to the situation is more appropriate. And consistency also is important. Consistency of application so that you don't get what I would call artificial, arbitrary swings in numbers because, in this case, they've been valued too low and then something transpired and they had to be written back up or whatever. [Also included in 2(c)] [TC 3/11, p. 42]
Committee/Staff/Observer
Let me give you an example that was given to me so I can test your feel for conservatism. Real estate market in Europe is apparently weakening. Realistically, perhaps the collateral is salable now at the value of the loans on the books. However, conservatism might tell you that you believe that the trend will be down. So a very conservative approach would be to project what the real estate market will be three years hence and write the loans that are collateralized down to that level now. Is that appropriate use of conservatism? [Also included in 4] [TC 3/11, p. 42]
Participant C-17
No, the move to write the banks' assets down on their performing assets to collateral value would be pretty strenuously objected to. Conservatism may create some real hardships in terms of their ability to lend or their willingness to lend. To just simply do it because it appears to be the most conservative approach is not a good idea. [Also included in 4] [TC 3/11, p. 42-43]
Participant C-5
Reasonableness or neutrality would suggest that the auditor has to review the likely ability to hold the asset through this cycle. U.S. financial institutions just showed record profits in 1992 of $32 billion. They also showed losses in 1991 and maybe, in fact, that they didn't have either one of the two but their concept of conservatism has both made banks look tremendous in 1992 and totally abominable in 1991. So conservatism is really showing its true colors in that kind of scenario. The banks are not that much better than they were and they weren't as bad. [TC 3/11, p. 43]
Participant C-1
Just in terms of the inventory, I would just take out of current assets anything that the banks wouldn't lend against. Because then it's not current. [Also included in 5(b)] [TC 3/11, p. 43]
Participant C-17
But that's not always the case. If I'm sitting as a secured lender to the inventory, I'm looking at it in terms of what's it going to bring to me if it liquidates. So a lot of times, a lender's going to advance against his perception of liquidation value, not the normal operating cycle. [Also included in 5(b)] [TC 3/11, p. 43]
Participant C-12
I think one of the problems with bank lending on real estate generally is for the first year, maybe even two years into the process, everybody's reasonable estimate of the value of the collateral, even most people's worst case, isn't bad enough. But a number of U.S. banks today have levels of reserves which are at or above the level of problem loans. [TC 3/11, p. 43]
Committee/Staff/Observer
The next question is: when companies are doing well, should accounting encourage them to put away larger reserves so that when things aren't going so well, they have those to fall back on or should the lean years and the fat years essentially stand on their own? [TC 3/11, p. 43-44]
Participant C-14
I think most of us are going to say no to accounting smoothing earnings and I would say that net worth is your reserves for cyclical or seasonal peaks. [TC 3/11, p. 44]
Participant C-11
If a company's business has swings in it due to sales or unexpected cost changes, those kinds of things are very important to show in the financial statements. On questions of reserving, if you're a financial institution and you're making loans, by definition they're risky. There is a loss content in the aggregate in all lending. So rather than just letting it all hang out there should be a recognition of loss content in loans at the time that the loans are made and you're getting income relating to that. [TC 3/11, p. 44]
Participant C-5
I would not agree that accounting should smooth the volatility in earnings but I also would suggest it should not exaggerate the volatility and it has to understand the company's movements and activities throughout these cycles that exist. I repeat the bank example; they exaggerated the volatility that has occurred over the last two years in the reporting of the financial information. The statement that higher quality earnings were attributable to consistency, that's true. An investor, a creditor is very comfortable with a stable source of period earnings or a steady trend line of earnings. If a company is in a less cyclical business, it is able to operate with lower leverage. It is able to operate with lower multiples of coverage and so forth and that's just the advantage of that business. But cyclical companies have to make a lot of money in the good times because of that and investors accept that. So I'm not troubled by the fact of volatility or lack of volatility as a reflection of quality of earnings. That's understood and the risk factors that go into it are there and the capital bases required to deal with that are appropriate. [TC 3/11, p. 44]
Participant C-13
I agree with that. But [participant C-5], with all due respect to your former life, it was the regulators not the accountants who caused the volatility. [TC 3/11, p. 45]
Committee/Staff/Observer
[Participant C-13], let me ask a question. [Participant C-5] says that there's more comfort, and I presume that means that there's better credit given, to stable earnings. Doesn't that tend to push companies then to do what they can to smooth earnings, to see if they can achieve a lower cost of capital? And if so, is that a bad thing? [TC 3/11, p. 45]
Participant C-13
I agree that stable results would tend to lower cost of capital. And that therefore there is an incentive to try and report stable results in order to lower your cost of capital. So the tendency is always in that direction. Investors have to be appraised of the true volatility in order to make correct credit judgments in allocating capital. So it's important that the financial statements reflect the underlying reality, getting back to the word that [participant C-11] used earlier. [TC 3/11, p. 45]
Participant C-17
I'd rather see real reserves for real expected expenses. I think it's something of a conflict because the other part is that the market does reward stable earnings. What you see happen a lot is the attempt to keep stable earnings. But when there's nothing more to lose because you're going to have "a bad year anyway", that's when an asset suddenly gets written off. That's when those issues are faced. [TC 3/11, p. 45]
Participant C-12
In general I don't want the accounting to prove the quality of earnings. On the other hand, looking at something like bank lending on real estate, maybe we should recognize that that is a cyclical business, that the value of the collateral the banks are lending on is cyclical and that the higher the cycle and the longer the up part goes, the deeper the trough is going to be and maybe that should be factored into earnings each year. Because it would be hard to argue that the banks didn't, in effect, open the port to earnings in the late eighties. [TC 3/11, p. 45-46]
Participant C-5
I don't disagree. It exaggerated both ways. It exaggerated the boom and it exaggerated the bust. So the accounting which thought it was being conservative actually didn't help either way. [TC 3/11, p. 46]
[Context] Meeting of the Creditor Discussion Group on March 11, 1993. Part of the meeting was devoted to the topic of the impact of litigation on auditing and external reporting. During the discussion, a comment was made on qualitative aspects of external reporting.
Participant C-5
I realize the dilemma in what we get in that it [litigation] does constrain. I know from the other side of the fence that it constrains us in the way of disclosures for both management and then the accountants. And it gets back to conservatism; you're unlikely to be sued for being conservative. I think in any attempt to remove this, what you need to do is increase the disclosure, not of the actual financials but the disclosure of the work that's been conducted. There's still a lot of flexibility on the part of individual firms to make their own determinations. In our environment, the potential for us to suggest a CPA change is almost nonexistent. [Also included in 18(b)] [TC 3/11, p. 62]
[Context] Responses to the postmeeting questionnaire to the March 17, 1993 Investor Discussion Group meeting. QUESTION 9 a. Indicate your degree of agreement or disagreement with the following descriptions of conservatism: SA - Strongly Agree A - Agree N - Neutral D - Disagree SD - Strongly Disagree Strongly Agree Neutral Disagree Strongly Agree Disagree 1. Conservatism means to anticipate no profits but anticipate all losses. 1 1 2 1 2. Conservatism means providing for losses when profits are high so that a company's resistance to recording losses in low- profit periods will not result in overstated assets or under stated liabilities. 3 2 3 Conservatism offsets the natural over-optimism of manager s who tend to view the expected operations of their company through rose-colored glasses. 2 2 1 4. Conservatism makes it likely that possible errors in measurement will be in the direction of understa tement rather than overstatement of net income and net assets, and thus future surprises are likely to be pleasant.1 2 1 1 5. Conservatism means that the uncertainties that inevitably surround many transactions should be recognized by exercising prudence in preparing financial statements but does not justify creation of secret or hidden reserves. 4 1 6. Conservatism should result in using the less optimistic estimate if two estimates of amounts to be received or paid in the future are about equally likely. 2 1 1 1 7. Conservatism should not dictate using the more pessimistic amount rather than the more likely one if two amounts to be received or paid in the future are not eually likely. 2 3 8. Conservatism should not require deferring recognition of income beyond the time that adequate evidence of its existence becomes available nor justify recognizing losses before there is adequate evidence that they have been incurred and can be reasonably estimated. 2 2 1 9. Something else. Please describe: Participant I-16: Awareness of uncertainty which is more likely to be unfavorabe that favorable, management optimism which may bias estimates and the fact that the impact of unfavorable developments may be greater than that of favorable developments. Be aware of tendency to assume favorable resolution of unknowns. b. Please indicate your degree of agreement or disagreement with each of the following statements about the proper role of conservatism in financial accounting and reporting by inserting the appropriate letter(s) in the blank. SA - Strongly Agree A - Agree N - Neutral D - Disagree SD - Strongly Disagree SA A N D SD 1. A company's use of conservative accounting is so important to the quality of its earnings that conservatism should take precedence over relevance, reliability , and neutrality in financial reporting. 3 2 2. Consistent use of conservative accounting procedures is desirable because while it generally does not decrease total reported income -- accounting procedures that decrease income of one period almost inevitably income of a later period or periods -- it shifts income to later periods and shifts losses to earlier periods. 1 2 2 3. Accounting should avoid reporting values that may not be realized, even if their realization is likely, because a cushion of unrecognized gains provides a needed margin of safety. 1 1 3 4. Credibility of accounting information rests on its reliability and neutrality--users can depend on it to represent faithfully the economic things or events that it purports to represent without bias intended to attain a predetermined result or to induce a particular mode of behavior--with which conservatism tends to conflict. 4 1 5. Conservative accounting favors buyers of securities over sellers by suppressing good news while revealing or anticipating bad news. 1 4 6. Evolution of conservatism from deliberate understatement of assets or overstatement of liabilities to a prudent reaction to uncertainty to try to insure that uncertainties and risks inherent in business situations are adequately considered should continue. 3 2 Participant I-11: I think the issue here is one of degree. For instance, in #3, I would feel different about a large value with a 51% probability and a small (but still "material") value with a 99% probability. That, I suppose, is where appropriate reserve levels come into play. [PMQI 3/17, p. 14-17] QUESTION 10 Please indicate your degree of agreement or disagreement with the following statements about volatility of reported income by writing the appropriate letter(s) in each of the spaces below: SA - Strongly Agree A - Agree N - Neutral D - Disagree SD - Strongly Disagree a. Investors usually attribute a higher quality of earnings if a company's earnings or net income show a stable trend and attribute a lower quality of earnings if a company's earnings show variability or volatility. 4 1 b. Stable earnings reduce the cost of capital.4 1 c. Companies that report significant swings in earnings are more difficult to analyze. 3 1 1 d. Companies should be encouraged to use accounting methods that spread the effects of unusual transactions rather than create volatility in reported year-to-year earnings. Participant I-9: If they can properly do so. Otherwise, my answer is inconsistent with "f" 2 2 1 e. Companies should be encouraged to dampen reported volatility by increasing reserves in periods of high earnings and reducing those reserves in periods of low earnings. 1 1 3 f. Companies whose businesses are volatile should faithfully report that volatility. That is , they should not smooth earnings to appear less volatile than the underlying business.. 3 2 [PMQI 3/17, p. 17-18] QUESTION 11 Please answer the following regarding volatility. For each item, please indicate whether the described accounting 1) is a significant problem to users of financial information, and 2) whether the problem, if any, can be solved through a) expanded disclosure alerting users to the degree to which the chosen accounting fluctuates in response to economic events or b) standards changes that eliminate the problem, if any, imposed by the accounting. Problem Solution YES NO DISCLOSURE STANDARDS CHANGE a. Accounting procedures - such as for changes in foreign exchange rates, oil & gas activities, pensions benefits, and futures contracts - make earnings appear more volatile and companies appear more risky than they probably are. 4 3 b. "Historical cost" accounting procedures - such as those that ignore prices of assets unless they are bought or sold or those charging equal amounts of expenses to each period by systematic allocations - make earnings more stable and companies less risky than they probably are. 3 2 3 c. Accounting procedures that recognize transa ctions and other events as they occur - such as accounting for changes in foreign exchange rates, pensions and other postretirement benefits, and investments in marketable equity securities - often are corrupted by provisions that smooth reported earnings, such as those that either spread the effects of price changes over several periods or allow the resulting gains and losses to bypass reported earnings or net income. 3 2 2 1 d. Accounting procedures - such as creating reserves that do not represent liabilities and capita lizing costs that do not represent assets or represent questionable or doubtful assets - permit companies to manipulate reported earnings by shifting costs and revenues between periods to create an illusion of consistency, leading many investors and creditors to underestimate the riskiness of the company.5 5 1
COMMENTS
Participant I-16: Companies take charges as "one time" events while taking favorable adjustments as part of "on-going operations." The problem is both with standards and disclosure. Investors reward companies for stable earnings trends, creating substantial incentives for distortion.
Participant I-11: I don't think most of these are major issues, and I think that often the standards that have been enacted to deal with perceived abuses have created larger problems than they solved.
[PMQI 3/17, p. 18-19]
[Context] Responses to the postmeeting questionnaire of the March 11, 1993 Creditor Discussion Group meeting.
QUESTION 10
a. Indicate your agreement or disagreement with the following descriptions of conservatism:
SA - Strongly Agree
A - Agree
N - Neutral
D - Disagree
SD - Strongly Disagree
Participant C-5: I don't like conservatism.
SA-1,D-11,SD-1
1. Conservatism means to anticipate no profits but anticipate all losses.
Participant C-11: (D) But some FASB opinions and auditors use this definition.
A-1,D-8,SD-4
2. Conservatism means providing for losses when profits are high so that a company's resistance to recording losses in lowprofit periods will not result in overstated assets or understated liabilities.
A-3,N-5,D-5
3. Conservatism offsets the natural overoptimism of managers who tend to view the expected operations of their company through rosecolored glasses.
SA-1,A-10,N-2
4. Conservatism makes it likely that possible errors in measurement will be in the direction of understatement rather than overstatement of net income and net assets, and, thus, future surprises are likely to be pleasant.
SA-4,A-8,N-1
5. Conservatism means that the uncertainties that inevitably surround many transactions should be recognized by exercising prudence in preparing financial statements but does not justify creation of secret or hidden reserves.
A-1
6. Something Else.
Please Describe:
Participant C-4: Always error to the conservative side! Particularly when doing/using estimates.
Participant C-17: Conservatism means the firm carefully analyzes future trends and events and reserves at a level consistent with the probable future operating environment one to two years out.
b. Please indicate your agreement or disagreement with each of the following statements about the proper role of conservatism in financial accounting and reporting by inserting the appropriate letter(s) in the blank.
SA - Strongly Agree
A - Agree
N - Neutral
D - Disagree
SD - Strongly Disagree
A-1,N-1,D-6,SD-5
1. A company's use of conservative accounting is so important to the quality of its earnings that conservatism should take precedence over relevance, reliability, and neutrality in financial reporting.
A-6,N-2,D-5
2. Consistent use of conservative accounting procedures is desirable because while it generally does not decrease total reported income -- accounting procedures that decrease income of one period almost inevitably increase income of a later period or periods -- it shifts income to later periods and shifts losses to earlier periods.
A-5,N-2,D-4,SD-2
3. Accounting should avoid reporting values that may not be realized, even if their realization is likely, because a cushion of unrecognized gains provides a needed margin of safety.
Participant C-12: assuming "may not" means a 5-10% or more likelihood of the value not being realized, not a 0.5% or less probability.
Participant C-17: Rather because uncertainty always exist. No profit should be reported before its time. The probability and amount, however, should be revealed in discussion (MD&A) or footnotes.
[PMQC 3/11, p. 15-17]
QUESTION 11
Please indicate your agreement or disagreement with the following statements about volatility of reported income by writing the appropriate letter in each of the spaces below:
SA - Strongly Agree
A - Agree
N - Neutral
D - Disagree
SD - Strongly Disagree
SA-2,A-7,N-1,D-3
a. Creditors usually attribute a higher quality of earnings if a company's earnings or net income show a stable trend and attribute a lower quality of earnings if a company's earnings show variability or volatility.
Participant C-11: Quality is not good if the tend is stable, but the "real" underlying trend is variable.
Participant C-4: Could be cushioning earnings, hiding volatility, I want the facts.
SA-1,A-7,N-3,D-2
b. Stable earnings reduce the cost of capital.
Participant C-11: if the stability is for real.
Participant C-21: Stable earnings because industry and company are stable.
SA-1,A-9,N-2,D-1
c. Companies that report significant swings in earnings are more difficult to analyze.
Participant C-21: But if that is the nature of their business or industry and thus a risk that needs to be understood, not buried in an accounting treatment.
A-1,N-2,D-8,SD-2
d. Companies should be encouraged to use accounting methods that spread the effects of unusual transactions rather than create volatility in reported year-to-year earnings.
Participant C-9: Depends on transaction.
A-2,N-1,D-8,SD-2
e. Companies should be encouraged to dampen reported volatility by increasing reserves in periods of high earnings and reducing those reserves in periods of low earnings.
Participant C-11: If the reserves reflect underlying risk present - such as loan loss reserves.
Participant C-5: Unless high earnings create high risk, staff and cost buildup, lower quality loans, etc. I would argue that (high earnings) result from underreporting of risks, opportunity costs, etc.
SA-6,A-7
f. Companies whose businesses are volatile should faithfully report that volatility. That is, they should not smooth earnings to appear less volatile than the underlying business.
Participant C-14: Quality of earnings reflects degree of predictability. Even companies with volatile/variable earnings can have predictable earnings to the extent that the volatility is caused by cyclical/seasonal factors. Predictability is a reflection of management quality and results in lower cost of capital.
[PMQC 3/11, p. 17-18]
QUESTION 12
Please answer the following regarding volatility. For each item, please indicate whether the described accounting 1) is a significant problem to users of financial information, and 2) whether the problem, if any, can be solved through a) expanded disclosure alerting users to the degree to which the chosen accounting fluctuates in response to economic events or b) standards changes that eliminate the problem, if any, imposed by the accounting.
Problem? Solution?
Y - YES D - Disclosure
N - NO S - Standards Changes
Y-4,N-9 D-3,D&S-1
a. Accounting procedures - such as for changes in foreign exchange rates, oil & gas activities, pensions benefits, and futures contracts - make earnings appear more volatile and companies appear more risky than they probably are.
Y-6,N-7 D-5,S-1,D&S-1
b. "Historical cost" accounting procedures - such as those that ignore prices of assets unless they are bought or sold or those charging equal amounts of expenses to each period by systematic allocations - make earnings appear more stable and companies appear less risky than they probably are.
Y-7,N-5 D-5,S-3
c. Accounting procedures that recognize transactions and other events as they occur - such as accounting for changes in foreign exchange rates, pensions and other postretirement benefits, and investments in marketable equity securities - often are corrupted by provisions that smooth reported earnings, such as those that either spread the effects of price changes over several periods or allow the resulting gains and losses to bypass reported earnings or net income.
Y-9,N-2 D-6,S-1,D&S-1
d. Accounting procedures - such as creating reserves that do not represent liabilities and capitalizing costs that do not represent assets or represent questionable or doubtful assets - permit companies to manipulate reported earnings by shifting costs and revenues between periods to create an illusion of consistency, leading many investors and creditors to underestimate the riskiness of the company
COMMENTS:
Participant C-20: Under a. Sufficient disclosure already. Under c. These are good reasons for smoothing (e.g., actuarial gains and losses under pensions). Under d. Present GAAP does not permit questionable assets. This is an issue regarding the ethics of management.
Participant C-14: I am not sure that the way I am reading the question is as intended.
Participant C-11: Regarding c: Bad question. Too many different issues present. Regarding d: This doesn't sound like GAAP - I don't know what issues are here.
Participant C-9: D - This area can be a great unknown. Users may not have sufficient information to make an educated assessment. Standards may provide potential to window dress - couldn't that also lead to creditors overestimating the riskiness due to lack of supporting information?
[PMQC 3/11, p. 19-20]
__________
[The CIC] oil industry subcommittees [complimented oil companies] regarding the quality and timeliness of information made available to investors and the awareness of most managements of their obligation to those who own the company. [However,] the Insurance Subcommittee, [commented]: "In general, comments from subcommittee members showed a growing frustration with the lack of candor and insight into the numerous problems of both the life and property-casualty industries provided by many insurance management teams. There is a sense that too many companies are not being managed in an effective manner. . . .It is hard to believe, but the quality of the industry's reporting to shareholders continues to deteriorate." [Also included in 2(a) and 16(b)] [AIMR/CIC92, p. 1]
Most [CIC] subcommittees agree . . . [that] the following suggestion seems appropriate: [Also included in 1(b), 2(c), 3(b), 3(d), 5(a), 5(d), 11(a), 13, and 16(b)] [AIMR/CIC92, p. 3
Quarterly reports with timely data presented in a format comparable to that of the annual report. [Also included in 1(b), 2(c), 11(a), and 16(b)]
[AIMR/CIC92, p. 3]
__________
The questions increasingly raised about the validity of Precept 1, below, give rise to our conviction that more has to be done by the FASB, the SEC, and accounting teachers and researchers to reassert the objectives established for accounting standards by the Trueblood Study Group and the FASB. [AIMR/FAF91, p. 18]
Precepts by Which the FASB Operates
Precept 1 [To be objective in its decision making and to ensure, insofar as possible, the neutrality of information resulting from its standards]. Users wholeheartedly support the precept that the FASB be objective in its decision making and ensure neutrality of information. Any other approach would render financial statements useless to investors and, we firmly believe, to the economy and society at large. We believe that the FASB has been diligent about trying to achieve neutrality, despite the various pressures brought to bear by its various "constituencies." [AIMR/FAF91, p. 18]
No preparer of financial statements, however diligent, can know in advance the goals, objectives, and needs of any given investor, i.e., whether an investment is made in the hope of income over a definite or indefinite period, or in the hope of capital appreciation, again over a predetermined or indefinite period of time, or a combination of both, or for some other reason. The interplay of the various interests, goals, and objectives of investors is what creates the efficiency of the markets and, not at all secondarily, enhances opportunities for capital formation. [AIMR/FAF91, p. 18-19]
The FASB's Statement of Accounting Concepts No. 1 (Objectives of Financial Reporting by Business Enterprises) summarizes the objectives and role of financial reporting as investors understand them this way:
Financial reporting should provide information that is useful to present and potential investors and creditors and other users in making rational investment, credit, and similar decisions. The information should be comprehensible to those who have a reasonable understanding of business and economic activities and are willing to study the information with reasonable diligence. (Paragraph 34)
. . . The role of financial reporting in the economy is to provide information that is useful in making business and economic decisions, not to determine what those decisions should be. . . . To the extent that financial reporting provides information that helps identify relatively efficient and inefficient users of resources, aids in assessing relative returns and risks of investment opportunities, or otherwise assists in promoting efficient functioning of capital and other markets, it helps to create a favorable environment for capital formation decisions. However, investors, creditors, and others make those decisions, and it is not a function of financial reporting to try to determine or influence the outcomes of those decisions. The role of financial reporting requires it to provide evenhanded, neutral, or unbiased information. (Paragraph 33) [Emphasis added.]
[AIMR/FAF91, p. 19]
The conclusions stated in the Concepts Statement did not, like Venus, spring from the ocean depths fullborn. The Trueblood Study Group discussed a list of the qualitative characteristics of reporting in its Objectives report. Some paragraphs on "Freedom from Bias" are worth repeating. [AIMR/FAF91, p. 20]
Preparers and users, borrowers and lenders, buyers and sellers, special interest groups, and others have primary interests in financial statements. While any information affected by judgments necessarily has some bias, there should be no purposeful bias favoring any group. Absence of bias, which may be characterized as neutrality and fairness, has long been recognized in accounting, although the perception of what is neutral and fair has changed with the times and needs.
In financial statements, avoiding bias which possibly benefits the interests of one group at the expense of another requires that the application of conservatism be carefully considered. Conservatism for its own sake may actually introduce bias. Confining financial statement representations to the results of transactions and other events for which substantial evidence exists and recognizing the varying degree of uncertainty would seem to aid in avoiding bias.
Degrees of uncertainty in accounting information are not disclosed at present. Instead, to avoid possible adverse consequences and to counter possible management bias, the most conservative among alternative accounting treatments has generally been favored. Thus losses, but not gains, tend to be anticipated.
If financial statements do communicate information about varying degrees of uncertainty, about the judgments made and the interpretations applied, and about the underlying factual information, then the impact of surprises--pleasant or unpleasant--will diminish greatly. This should result in a substantial lessening in the belief that conservatism is essential.7
[AIMR/FAF91, p. 20]
__________
Focus Group Comments, Analysts: The comments made by analysts in the focus group meetings were generally consistent with and supportive of the survey results. Although direct comparisons are not possible, inferences were drawn. The table below presents the main conclusions from the survey with responses from the focus groups: [Also included in 1(b), 1(c), 2(a), and 4] [KPMG BANK STUDY, p. 39]
Indicated concern over the amount of subjectivity involved in making fair value estimates and questioned the ultimate usefulness of the results [Also included in 2(a) and 4] [KPMG BANK STUDY, p. 39]
User Survey Results, Users: The comments made by analysts in the focus group meetings were generally consistent with and supportive of the survey results. Although direct comparisons are not possible, inferences were drawn. The table below presents the main conclusions from the survey with responses from the focus groups: [Also included in 2(c), 4, 5(a), 5(b), 5(d), and 13] [KPMG BANK STUDY, p. 39]
Generally believed fair value disclosures of financial instruments would be useful provided they were reliable and comparable [Also included in 2(c) and 4] [KPMG BANK STUDY, p. 39]
The FASB, the Securities and Exchange Commission (SEC) and other regulatory bodies are currently considering a requirement to prepare financial statements based on market values in place of financial statements prepared on a historical cost accounting basis. The questions in this section relate to this issue: [Also included in 1(b), 2(a), 2(c), 4, 10(b), 11(a), and 15] [KPMG BANK STUDY,
p. A-9]
Indicate whether you believe fair value accounting should be the primary accounting basis for the preparation of an institution's financial statements.
10% Yes
90 No
0 No opinion
[Also included in 2(a), 2(c), 4, and 15] [KPMG BANK STUDY, p. A-9]
[One user commented] no. Grave doubts exist as to the usefulness and accuracy of 'fair value'. [Also included in 2(a), 4, and 15] [KPMG BANK STUDY, p. A-9]
[One user commented] no. Fair market value can be easily manipulated for many financial instruments. In fact, many had commercial real estate loans made based upon 'estimates of market value'. [Also included in 4 and 15] [KPMG BANK STUDY, p. A-9]
[One user commented] no. It is difficult to determine the fair value of many assets and liabilities. This could distort financial statements and hinder comparability. [Also included in 2(c), 4, and 15] [KPMG BANK STUDY,
p. A-9]
[One user commented] no. Fair value accounting is too judgmental and too susceptible to external factors (i.e., market fluctuations) to serve as the primary accounting basis. [Also included in 4 and 15] [KPMG BANK STUDY, p. A-10]
[One user commented] no. [There is an] inability to accurately adjust values of all asset and liability categories. [Also included in 4 and 15] [KPMG BANK STUDY, p. A-10]
[One user commented] no.
--Misrepresents "lending to maturity" aspect of bank loans.
--Concern about behavioral impact on bankers.
--Costs more to gather information than benefits users.
--Too much estimation required; comparability and integrity [are] questionable.
--Misuse of information by less-sophisticated users.
[Also included in 2(c), 4, and 15] [KPMG BANK STUDY, p. A-10]
High degree of subjectivity, and the inherent uncertainty of forecasts on which valuations are based, will diminish both the consistency and comparability of financial institutions' reports. [Also included in 2(c), 4 and 15] [KPMG BANK STUDY, p. A-10]
Increase potential for accounting abuses.
[Also included in 4 and 15] [KPMG BANK STUDY, p. A-10]
[One user commented] no. Too judgmental once one gets away from liquid and marketable instruments. [Also included in 4 and 15] [KPMG BANK STUDY, p. A-10]
[One user commented] yes, however, I would qualify my answer by acknowledging that valuation of many loans is subjective. Thus I have serious concerns regarding:
1. Comparability among more conservative and less conservative banks
2. Restrictions in lending/credit crunch involving borrowers, which are very difficult to value. [Also included in 4 and 15] [KPMG BANK
STUDY, p. A-11]
There are current accounting rules that require the disclosure of fair values, realized and unrealized gains and losses, cash flow information and maturities and yields of investment securities. Considering that this information is already available, indicate whether you believe the historical cost based accounting should be replaced with fair value based accounting.
8% Yes
88 No
2 No opinion
2 No response
[Also included in 2(a), 2(c), 4, and 15] [KPMG BANK STUDY, p. A-11]
[One user commented] no. Much of the additional information that would be available with fair value accounting must be based on estimates which are likely to incorporate varying assumptions and therefore, is unlikely to be reliable or consistent. Further, much of what is proposed is irrelevant for valuing a banking company. [Also included in 2(a), 2(c), 4 and 15] [KPMG BANK STUDY, p. A-11]
[One user commented] no. Most adjustments would be unreliable approximations of improbable or impossible transactions; they would create destabilizing volatility of earnings that would not fairly reflect realities of going concerns. Lack of timeliness of problem also. [Also included in 4 and 15] [KPMG BANK STUDY, p. A-12]
For an institution that has the intent and ability to hold assets for the foreseeable future (defined as 12 to 18 months), indicate whether you believe fair value accounting is appropriate.
30% Yes
60 No
8 No opinion
2 No response
[Also included in 2(c), 4, 10(b), and 15] [KPMG BANK STUDY,
p. A-13]
[One user commented] yes. Some effort at fair value accounting is still useful for these organizations. However, if the subjectivity involved is too great for certain loans so that comparability is destroyed, I would favor historical cost with an explanatory footnote. [Also included in 2(c), 4, and 15] [KPMG BANK STUDY, p. A-14]
Estimates of fair value may vary by institution because of different assumptions, methodologies and the practicability of such disclosure. The following questions relate to the reliability and comparability of fair value estimates: [Also included in 1(b), 1(c), 1(d), and 4)] [KPMG BANK STUDY, p. A-20]
If fair value estimates for financial instruments were only included in the footnotes to the financial statements, select the letter which best describes the amount of measurement error in fair value estimates you would tolerate before you consider these estimates to be misleading. Measurement error is defined as the variance between the precise fair value and the disclosed estimated fair value:
a. Within plus or minus 1%
b. Plus or minus 1 to 5%
c. Plus or minus 6 to 10%
d. Plus or minus 11 to 20%
e. More than 20%
f. Other
g. No response
_______________________________________________________________ A B C D E F G _______________________________________________________________ 33% 45% 13% 5% 4% Equity investment securities 35 53 5 5 2 Debt investment securities 55 30 5 3 5 2 Purchased mortgage servicing rights 50 35 5 3 5 2 Excess mortgage servicing rights 10 50 23 3 8 5 1 Loans 18 48 15 3 5 5 6 Demand deposits 20 45 15 3 5 5 7 Time deposits 23 50 10 5 5 7 Long term debt 53 30 5 3 5 4 Financial guarantees 5 45 25 10 5 5 5 Commitments to extend credit 3 53 28 5 5 5 1 Letters of credit 10 55 18 5 3 5 4 Swaps, options, futures, etc. 3 Other ____________________________ A B C D F _______________________________________________________________ 58% 25% 5% 3% 9% Equity investment securities 63 25 3 9 Debt investment securities 40 33 15 3 9 Purchased mortgage servicing rights 40 35 13 3 9 Excess mortgage servicing rights 50 23 15 3 9 Loans 50 30 8 3 9 Demand deposits 55 25 8 3 9 Time deposits 53 30 5 3 9 Long term debt 35 43 8 3 11 Financial guarantees 38 35 13 3 11 Commitments to extend credit 35 40 10 3 12 Letters of credit 48 33 3 3 13 Swaps, options, futures, etc. 5 Other
The papers are a summary of a committee and staff members' discussions with selected sell side analysts at GoldMan Sachs.