previous in series || table of contents || next in series


Professor George H. Sorter
Professor-emeritus of Accounting
London School of Economics
March 27, 1979

[Introductory note: Professor George H. Sorter is Professor of Accounting and Chairman of the Department of Accounting, Taxation and Business Law at New York University. He is also Associate Director of the New York University Ross Institute of Accounting Research. Prior to joining New York University, he was Professor of Accounting at the University of Chicago Graduate School of Business.

Professor Sorter earned his bachelors, masters, and doctoral degrees at the University of Chicago. He is a CPA (State of Illinois).

He was Research Director of the AICPA Study Group on Objectives of Financial Statements. He continues to serve as Consultant to Government, Educational Institutions, Industry, and Public Accounting Firms. For many years, Professor Sorter was a member of the Board of Editors of the Journal of Accounting Research, and is currently a member of the Journal of Accounting, Auditing and Finance.

He has written a number of books and monographs and his articles have appeared in numerous professional journals.]

The title "Beyond Emptiness and Blindness" was taken from Kant. The actual quote consists of two parts, the first of which I will use to start the speech, and the second to end it, with a mercifully brief time in between. The first part of the actual quote is, "Thoughts without content are empty, intuitions without concepts are blind." It seemed to me as well as to others who have shared a dissatisfaction with accounting research, that both emptiness and blindness has characterized accounting research until the recent past. I want to examine two phases in accounting theory and research. This examination will be far from exhaustive, although it may be exhausting. The first phase is roughly up to 1960, and can be characterized as both blind and empty. The second phase of research, from 1960 to the present, is blind and bloated, rather than blind and empty. First, let's consider the blind and empty phase. The general impression is that the period up to 1960 was a golden age of theory, perhaps empty of data and experience, but rich in farsightedness. This was the golden age of the classics we studied in our youth: Hatfield, Sprague, Paton, Littleton, Canning and Vatter, all the figures that we revered and that we nostalgically remember as the giants of pure theory. What a disappointment it was, in preparation of this speech, to riffle through these works once more and find out that while these were indeed works empty of data, they were also, unfortunately, devoid of substance and theory. What issues did these works deal with? The major, burning questions seem to be, should we use cost or value, is goodwill an asset, and should the accounting unit be thought of as a proprietorship or as an entity? But how can such questions be answered absent the specification of what the purpose of accounting is and how it is to be used? And yet, we search in vain, throughout all these works, for any specification of what the purpose of accounting may be. Consider cost vs. value. Both cost and value are non-false descriptions of differing attributes of an object or a state. How should the choice between these be made, or does a choice indeed need to be made? Why not use both of them? Where is a theory developed in these books that suggests what choice we should make, apart from personal prejudice and subjective preference? I didn't find any.

The next momentous question that most of these works dealt with is, is goodwill an asset? Well, that obviously depends on the definition of an asset. But does not the definition of assets, in turn, depend on, or should it not depend on what the purpose of accounting is, and how this purpose can be furthered by alternative and substantive definitions? This would seem to be necessary unless we're satisfied with the splendid definition of an asset, advanced in Terminology Bulletin 1 of the AICPA, which states elegantly, that "an asset is something represented by a debit balance that is, or would be, properly carried forward."

Now, we come to a really important issue, entity versus proprietorship. Who are we accounting for, an entity or a proprietorship? Books were written about this, dissertations, articles, the world was flooded with competing arguments. This debate just raged on and on, but what difference does it make? What accounting qualification or classification would change in any way, if we thought of the accounting unit as an entity rather than as a proprietorship? Revisiting these works, I come to the regretable conclusion that these works are not theoretical, at least as theory is defined in the dictionary, as "a coherent group of general propositions used as principles of explanation of a class of phenomena," or the alternative definition, "a particular conception of something to be done, or the method of doing it."

These theories did not prescribe, and indeed, they did not even describe in conceptual terms. A theory does not need to be prescriptive, it can be descriptive in order to illuminate and explain the world as it is. How illuminating are these works when the following fundamental questions were not dealt with? What is more fundamental than double entry? None of these works defined double entry operationally, that is, what does double entry really mean? Operationally, double entry is defined by three propositions. One: that each accounting event produces two and only two effects. Two: that each of these two effects is quantified by the same absolute number. And three: that the world consists of two and only two classes of objects. But what are the reasons for these propositions? What are the costs, what are the benefits, why did they arise, why do they continue to be observed? You will look in vain in any of these classic, theoretical works, or in any discussion of the costs or benefits of these propositions. Must all three necessarily exist together?

The absence of theory during these times was surprising; now the absence of data was expected. After all, this was before the days of empiricism, before the days of the Journal of Accounting Research. None of these works generated or tested hypotheses. Only Canning, who was a non-accountant, can be said to be an empiricist in any sense of the word, because he surveyed actual accounting practice in order to identify implied rules which govern the accounting practice. This is not what we presently think of as empiricism but it is. We equate empiricism today as numbers, but empiricism means "based on experience", and our experience is not restricted to numbers.

Official accounting pronouncement during this time, of course, was devoid, also of both theory and data, or even common sense. Accounting Terminology Bulletin #2 defines income as follows. Income: "amounts resulting from the deduction from revenues of cost of goods sold, other expenses and losses, or some of them." No wonder this needs to be labeled a definition. This definition is used later on in ARB 43 as follows, "The main objective, in terms of inventory valuation, is the matching of appropriate costs against revenues, so that there may be a proper determination of income." Now, given that definition of income, I leave it to your imagination how and what a proper determination of income is. Costs should be such that a proper income results and income is what results after planning costs properly. The definition that was cited is too much for me. Perhaps this circular definition is caused by accountants running around in circles, trying to find the window side in modern, windowless buildings.

All things pass, and the golden age of empty blindness gave way in the sixties to bloated blindness calculated to cause indigestion. In the sixties, the wonders of methodology burst upon the minds of accounting researchers. We entered what Maslow described as a mean-oriented age. Accountants felt it was their absolute duty to regress, regress and regress, certainly to a preverbal phase which forbade even a verbal appendix to the methodology papers since, if such a verbal appendix was published it would be sufficient cause for denial of tenure. It was an age which is characterized by two anecdotes attributed to Freud. Freud describes the person who continuously polishes his glasses and is so busy polishing the glasses that he doesn't put them on to see with. The other, perhaps, more damaging indictment also attributed to Freud is the lovely story of the drunk looking for his glasses. There is a drunk who looked for his glasses on the street corner underneath a street lamp. He searched for his glasses nearly being blind and soon attracted a crowd of people who helped him. After twenty minutes of fruitless search, they questioned him, "Where did you lose your glasses?" "I lost them two blocks down that dark alley." "But why are you looking for them here?" "Because the light is better." The notion that we suit research to the available methodology rather than the other way around is apparent to anyone who looks at the research of that period. Let us look at some of this seminal research. One is the information contents studies started by Ball & Brown. The Ball & Brown study demonstrated that accounting numbers had information content. This was both good news and bad news for the accountant. The good news was that accounting income numbers had information content. "Hurray!" cried the accountants, but most of that information was already impounded and used. That wasn't such good news, but what does information content mean? Information content was defined as just common sense, to demonstrate an association between accounting income and stock prices. If there was an association between accounting income and stock prices, accounting income numbers had information content. Why don't stock price changes then have information contents for accounting reports? Isn't that an equally true statement to make? Of course, there is an association between accounting income and changes in stock prices. Suppose you have larger than expected sales during a period. Would you expect that share prices would respond favorably to such a development? The answer is yes. Would you also expect that accounting income would go up? Of course you would. Does that mean that the accounting income number is used in terms of setting these share prices, or that it had information content?

The other main type of research of the period demonstrated that the choice of accounting method "makes no difference" -- the market is not fooled by "bean counters." "Makes no difference" is essentially defined to mean that there is no discernable effect around the "announcement date" of annual reports. These two types of accounting research lead to two unanswered questions and one that answers itself.

1. You are the chairman of the FASB. A suggestion has been made that accounting income in annual reports be the change in the stock prices, plus dividends from 1/ 1 / X to 12/ 31 / X. This income figure would have maximum "Information content" as measured by the conventional definition of "information content," would reduce audit fees, and would be speedily available so that annual reports could be issued on a more timely basis. How do you respond to this suggestion?

2. You again are chairman of the FASB. You have just defended the FASB's action for successful efforts by proving that the allegation that stock prices and debt financing would be adversely affected is in error. The market is efficient; there would be no effect on stock prices, debt or anything else. The question is then directed to you -- if that is so, why do you care, why not let everybody do what they want?

3. Why did you have trouble answering the above questions? If there is no model about how accounting information is used, we do not know what this type of research means -- we are truly blind. I'm not the first to have made this criticism. Neil Haakinson and Hector Anton said very much the same thing, much better. We need both model and observation. Either is incomplete without the other.

We don't have a theory! We don't have objectives! You might quote the Objective Study but that study was very incomplete even though I had a major share in it. It wasn't such a bad study. It did identify the proper objectives of financial statements but it didn't suggest a model by which the financial statements meet those objectives. This is my complaint. Without such a model you really are blind, you continue to be blind. There are certain hopeful developments that people are beginning to think about of possible uses of financial statements that take into account that financial statements are issued with a time lag, and that you cannot expect that financial statements provided fresh news at point of impact. In a recent paper a colleague and I suggest that they provide a mechanism which allows users to interpret new information as it comes along. By identifying certain crucial relationships it allows users to interpret new sales data when it comes along in terms of what a change in sales means to Chrysler vs. General Motors. Statements allow users to determine the significance of a wage settlement. In addition, financial statements can be used to validate alternate information sources and users predictions.

Well, let me try to sum up. Despite all the high-powered technology, despite all our glorious history, we're still not thinking in accounting. The very people that argue that it doesn't make any difference in an efficient market, whether you reveal things in terms of footnotes or in the body of the financial statements, will state in their research that they limited their sample to Compustat firms. Now, think about that for a moment. Their research is constrained by things reported in Compustat; footnotes are not reported in Compustat. Therefore, format makes a difference in their research but not in what they are researching. We need to encourage risk taking so that we can attack these fundamental problems. Thinking is fun, it's exciting. Accounting is fun and exciting, and thinking about accounting is splendid. Let me end as I started with the second half of the quotation from Kant. It goes on, after the quote that I gave before, to say, "The understanding can intuite nothing, the senses can think nothing. Only through their union can knowledge arise." Let such a union finally arise in accounting.


George, you and I will come out in the same end because we really very, very much agree, so what it is that I'm going to be doing now is merely trying to explore or expand some of the thoughts that you raised during your very fine presentation. Since I do have tenure and I'm "risk-free" let me try to take those three questions that you've put to those poor, scared-cats who are taking the doctoral examinations. I'll try to answer the third question first as to why I would have difficulty answering the first two, by merely saying that foolish questions deserve foolish answers, but I'm too scared to label your questions as being so. With respect to the first question about giving the stock market data and the like, actually, you have the answer, or the answer in objectives, so stock market data are given ex-post, whereas the readers of financial statements are trying to discern data ex-ante, so it is that the ex-post data are interesting statistics, but not responsive to the objectives of financial statements which, you say you've deduced as a result of empirical research when you were pursuing that objective study. Now, the second question is a very vital one, as to why the FASB, excepting the fact that it doesn't make a tinker's damn whether they use efficient, whether they use successful efforts or full-costing and the like, but yet, we know that the FASB, for mistaken reasons, are obsessed with the idea that they want to develop unitary, single-minded rules, rules instead of standards. And it's because they want rules, even though it doesn't make any difference, they're going to spend all of their time spinning their wheels and finally driving themselves into the quagmire, by coming up with these very silly dicta like an 8 and 2 and 19 and others along the way, including with leases, where they then tried to button up everything along the way.

George, there is one very important area, where I can't help but feel the sense of disagreement, and you were probably doing this in order to state a position, and you did it very beautifully and with a great deal of humor and many illustrations. I can't help but feel though, George, that when I entered accounting in the 1930's, and as I pursued it in the 40's and 50's and 60's and 70's, that I was standing on the shoulder's of giants, who through their various pursuits, whether you labeled it as research or not, they were very, very important in doing what? Making us think about what our accounting objectives are, and what we are committed to doing. As they kept writing and describing, they, as a result, caused us to react to what they were saying. I can't help but feel a very important debt to Payton and Littleton, for the felicity of phrase with which they put what it is that accounting is committed towards. I can't help but feel that we owe a debt, collectively, to those who gave us cost accounting concepts in the 1930's, in the 40's, in the 50's, whether you call it research or not. Developing a concept to see how costs might appear, and how they might be made to flow. So it is, that I feel that the sweep was far too general, and as a consequence of the theme, so very, very general, you had to even label your best shot, namely your being an executive director, or the research, or whatever it is, of the Trueblood study, as being non-research, and giving rise to essentially, naught, but a lot of platitudes. But yet, platitudes or not, it's something to think about, and to move us forward. Let's not be that harsh with respect to the past. And possibly one other point. You labeled blindness throughout as a pejorative, but yet, if you go back to the story of Sophocles, Oedipus Rex, he demonstrates, ironically, that the blind see better than those who are seeing.

I'm tempted to say that if foolish questions deserve foolish answers, then wise statements don't need answers. As Prime Minister Begin said yesterday, "I agree with you but with a small amendment." In rereading the giants of the past, they don't seem so tall, somehow they don't seem so wise, and if they stimulated us to think about the objectives and the concepts, then those people that they stimulated sure didn't do a very good job. As far as the blindness, it's only blind theory to which I object.

I'd like to talk about the giants whose shoulder's we're standing on, but don't think that would help me any after what Mr. Briloff did. I think more of the robber barons, and all we've heard about them; you look back on them, now they don't really seem so crooked. But the government started then and has continued to try to protect us from these people. Isn't that really one of the things that we're making our biggest mistake on, thinking that what we do can really help protect people from their own foolishness. We should protect ourselves from our own foolishness but not try to protect them from theirs. And I think that that's one of the big mistakes of a lot of this research in accounting, trying to help investors stop being suckers. I'd appreciate if you'd let me end there, even though that wasn't a very well-formulated question.

Well, I agree with you in one sense, I abhor paternalism, I abhor paternalism in any sort, and people should be free to make mistakes, to dig their own graves; I just want them to have all the information available.

I've been reading some research, I guess, for want of anything better to do, and what you've said about their thinking capacity leads me on another point. I found some of their written English so very poor, at times I just couldn't understand them. I was wondering if there was some degree of corellation between their ability to think and their ability to write.

I would think so.

May I add to what has been said already about our enjoyment of your talk which has been extraordinarily interesting and stimulating. I think perhaps one might summarize it by saying that you're surprised and disappointed that the amount of research that's being done is so unproductive. Well now, wouldn't it, perhaps, be helpful to consider the history of accounting in this context, and I hope Professor Briloff won't contradict me when I say that an eminent person, much in your position a century ago, described accounting as the one and only perfect science. Now, what has gone wrong between then and now? I think the answer is that we're expecting too much of something which was designed, or which grew up with very moderate objectives, and after all, the reason why bookkeeping was invented, the reason that shaped it's evolution, was the desire to keep track of the debts that were due to one, and what were owed to other people. And added, there was added to that the very reasonable desire to keep track of cash, to see that there wasn't waste, there wasn't fraud, and to keep one's relations sweet with other people. Now that was what could be described as the perfect science, bookkeeping could do it splendidly. But on top of that we've tried to build something else. The income statement, which is originally a kind of bonfire into which we put the items that didn't seem worth keeping any longer. The balance sheet grew out of a trial balance which performed the admirable role of check on our accuracy, and that was that. And if we said that an annual report shows that the controller of the company had managed to balance his books, and it seemed that petty cash hadn't been balanced, then we would know that it had done something good and has achieved a worthy objective. Well, on to this splendid basis, we now try and grasp something ambitious. I don't quite know why, I suspect that accountants feel a little bit ashamed of their skill at bookkeeping, and I'm sure that the teachers of accounting feel bored and ashamed and the easiest way to get out of it is to learn some statistics. An accounting department really consists of a lot of first rate accountants trying to become third rate statisticians. Well, if then, you accept this line of reasoning, shouldn't you say that really, the remarkable thing is that the superimposed structure hasn't been worse than it is. And the final suggestion: if we really do want all this highbrow research, we do look for all the results that seem desirable, oughtn't we to scrap the basis that we have and invent something entirely new?

That was Professor Baxter, and as usual he says things more worthwhile and says them better than anybody else. I agree with you, with an amendment. I think, I believe that that's gone wrong, we knew what we were about but didn't write it down, or didn't communicate very well, what the purpose of accounting was. Then the world changes, the needs change, and we didn't consider what the legitimate and proper response was. This lack of thinking, at that stage, that caused us the grief, and this continued lack of thinking, continued to cause us grief. Our objectives need to evolve, need to change, and most of all, need to be made explicit.

Referring to the accounting research, what will go for the future systems? What accounting research can be expected or anticipated for the future? For instance, we have developed the accounting system from the various bookkeeping stages to the management accounting level; in costing we introduced budgeted control and standard costing and other developments. So what can we anticipate or foresee for the further development for administrative accounting, like Board Room or high-control accounting? In management accounting we have again the report to the Board of Directors to summarize the operating research. Question 2: We have so far, for many centuries, been using the balance sheet. In what way can we recast or define in the future, instead of a balance, some other financial statement, which can be so meaningful for the disclosures for the outsiders as well as stock holders? We have recently come up beyond the balance sheet, the cash flow or working funds statement, to disclose in the annual report. Is there any possibility in research to improve the balance sheet of the present system? There is one way, in which the balance sheet is broken down in the Double Entry system, in industries like electrical, railway, and utility services, and broken down into permanent and fixed assets, and fixed liabilities, are broken down in the statement of capital... So is there any possibility in the future, or do you forecast?

Yes, two things quickly, in your point of view. In terms of the internal or the managerial accounting, I think there's good news and bad news. I think there's probably going to be much more advances made in that, but it's also going to be much less interesting, as it seems to me because by and large, the decision models are more specified, the inputs that are necessary are more known, and therefore, it just says, how do you create these inputs that go into known decision models? As for the external accounting, of course there can be improvement, but before we even know what improvement is we have to answer, what data do we (a) want, and (b) how can we use it?

I was just going to say that I'm not tenured so I'm not going to ask a very risky question, but it seems I'm on the last flight so I don't have much choice. You criticized the Trueblood Report, that it didn't provide a model for any empirical research, but it really does set up a criteria for ranking accounting principles. In other words, if accounting method A allows the user to better predict future cash flows than accounting method B. Now, when I read the report I was very excited about it, and I thought a lot of people were just going to sit there and now start writing articles and saying, well look, lets analyze this method and see which one is better. Now I did some work on it in the Fifo-Lifo area, but there's been almost no other research in that, trying to use the criteria. Why is that?

First of all, I may have been unduly harsh to the Trueblood Report, and if so, I apologize to Professor Sorter. My criticism was that there's no specific, explicit model or statement as to how the objectives can be attained, how accounting information is used or usable in order to accomplish the objectives of the Trueblood Report. I stand by the objectives of the Trueblood Report but they need to be made operational.


previous in series || table of contents || next in series

Newman Library Digital Collections