Normative Accounting Theories
Md. Humayun Kabir*
Abstract This paper reviews five important works on normative accounting theory – MacNeal (1939), Paton and Littleton (1940), Littleton (1953), Chambers (1966), and Ijiri (1975) – with emphasis on recognition and measurement issues in accounting. It shows that there is a lack of agreement among these theorists on basic assumptions and hypothesized information needs of the users. Even where there is agreement on an assumption, different implications have been drawn therefrom by the concerned theorists. These differences lead to different recognition and measurement proposals. This paper also shows that many of the ideas in accounting that seemed to be novel were anticipated by these early theorists. Finally, it assesses the present accounting practice in the light of these theories.
Keywords: Normative accounting theory, accounting recognition and measurement.
Introduction
This paper examines select normative accounting theories. Starting with the twentieth century,1 normative accounting theorists had been preoccupied with developing/constructing accounting principles. The primary concern had been recognition and measurement issues. Hence the focus of this paper shall be on these theorists’ proposals on accounting recognition and measurement and the arguments/theoretical structures behind these proposals. It also compares the works reviewed.
There are controversies among accounting academics regarding what an accounting theory is. Watts and Zimmerman (1986: 2) posit that accounting theory seeks to explain and predict accounting practice.2 Positivists like Watts and Zimmerman (hereinafter W & Z only) cite
* Md. Humayun Kabir, D.B.A., is an Assistant Professor in the Department of Accounting & Information Systems, University of Dhaka. The author is grateful to Professor Santi Narayan Ghosh and Professor Saroj Kumar Saha of the University of Dhaka and Professor Hidetoshi Yamaji of Kobe University of Japan for their helpful suggestions on an earlier draft of this paper. Comments of an anonymous reviewer are appreciated.
1 The debate about accounting principles did not start seriously before the twentieth century (MacNeal 1939: 69). Accounting practice was primarily concerned with the development of bookkeeping techniques up to the beginning of the twentieth century. Early bookkeeping books seem to have been concerned with the methods of making bookkeeping entries. In fact, the early books did not mention even such basic issues as the use and calculation of depreciation.
2 They label their theory as ‘Positive Accounting Theory.’ They borrow the phrase ‘positive theory’ from economics. Friedman (1953) popularized this phrase. What Watts and Zimmerman do actually is propagate a specific version of accounting theory. Methodologically they subscribe to the Kuhnian (1970) view of science. It is, however, to be noted that most of the research in economics, from which Watts and Zimmerman (1986) borrow their term positive, has normative consequences. When economists analyze a social phenomenon, their analysis contain in it normative implications. For example, information economics deals with the implications of information asymmetry and contains the suggestion that mechanisms should be devised so that the problem of information asymmetry can be mitigated. Agency theory highlights the agency problem arising from the separation of ownership from control in large corporations and suggests that corporate governance mechanisms should be designed and implemented to mitigate the agency problem. The same can be said of Adam Smith’s the Wealth of Nations. The Wealth of Nations did not merely explain a capitalistic economy. In fact the English economy was hardly capitalistic in today’s terms when Smith wrote the book. The Nation propagated the notion and benefits of a capitalistic economy. The book was written and used as an intellectual weapon by the bourgeois against the monarchy (Ashton 1968). Positive accounting theorists have studied the incentives of earnings management and have identified situations when earnings management may be present. But they hardly suggest a remedy to this problem. Sterling (1990) has rightly attacked this position of the positivists. This note, however, should not be taken as an attempt to belittle the achievements of positive accounting research. It has yielded many interesting and useful insights into financial
economics and natural science disciplines such as physics, chemistry, etc. in the defense of their method and call their method ‘the scientific method’ (W & Z 1986: 2), thus probably implying that there is only one method in science. This is highly disputed by accounting academics pursuing other strands of research (Christensen 1983; Chua 1986). Science is not a unified structure. There is no a unique scientific method. Science knows many methods (Feyerabend 1993). Hence, even if one wishes to study accounting as a scientific discipline, there is more than the method advocated by W & Z (1986).3 One major criticism of W & Z’s view of accounting theory is that it unnecessarily narrows the area of accounting research (Chua 1986; Whittington 1987). For our purpose, we adopt the following definition of accounting theory: “--- the business of accounting theory is to examine beliefs and customs critically, to clarify and extend the best from experience, and to direct attention to the genesis and outcome of accounting work” (Littleton 1953: 132). This definition accommodates different strands of accounting research such as research in normative accounting and empirical accounting as well as research in interpretative accounting.
A normative accounting theory seeks to prescribe some basis of accounting measurement, particular accounting procedures, and the contents of financial reports (Ijiri 1975; W & Z 1986).4 Ijiri views normative theories as a special case of deductive theories. Deductive theories that start with some goal assumptions and deduce accounting procedures therefrom are labeled normative theories.5 Thus, there are two important elements of a normative theory: (a) goal assumption, and (b) deduction. A theorist may set his own goals that are not inherent to current accounting practice. Chambers (1966) falls in this group. Again, a theorist may inductively derive goals from accounting practice and use those goals to suggest improvements in current practice. Ijiri falls into this group. Such theories are also categorized as normative in this paper. It is to be noted that not every theorist is explicit on goal statement. Some state the basic assumptions and deduce accounting measurement from these. Paton and Littleton (1940) fall in this group.
So far three approaches have been employed in normative accounting research. These are (a) inductive model, (b) deductive model, and (c) the decision usefulness approach. In induction, a general statement (X) is induced from some empirical observations, hypothetical phenomena, or non-empirical concepts (O). The implications of X include and go beyond O. It may be noted that many Xs may be induced from O. The contribution of an inductive model is in coming up with an X as an explanation of O. On the other hand, the opposite process is followed in deductive models. Here O is deduced from X. X is a set of theories, or assumptions that have already been accepted. In a deductive model, O is a special case of X. In the decision-usefulness, decision model approach, ‘information relevant to a decision model or criterion is
accounting and reporting issues. The point this note intends to make is that normative accounting research has, and should have, a rightful place in accounting theory and research.
3 Accounting also knows many methods. Positive accounting theory uses one of these methods. See Hopwood and Miller (1994) for a sample of studies using methodological assumptions/world views different from those used in positive accounting research. See Sawabe and Yamaji (1999) for a useful discussion of various approaches employed in the research of accounting. Specifically, they discuss various research approaches falling under the broader umbrella of institutional accounting research.
4 Watts and Zimmerman do not use the phrase ‘normative accounting theory’. Rather they (p.7) use the phrase ‘normative propositions.’
5 Ijiri differentiates between normative theories and policies. In normative theory, the researcher does not commit himself/herself to the goal assumed. In accounting policy, however, the researcher is committed to the goal. Normative theories can thus be scientifically verified, while policies are based on value judgement of the researcher. As Ijiri recognizes, this distinction is blurred in accounting. Accounting theories and policies are often intermingled.
2
isolated and various accounting alternatives are compared to the data presumably necessary for implementing these decision models’ (AAA 1977: 10).
Inductivists such as Hatfield (1927), Littleton (1953), Ijiri (1975), etc. examine extant accounting practice and have tried to rationalize and, sometimes, justify major elements of extant accounting practice. Among the inductivists, Ijiri (1975) is very explicit in his adoption of the inductive approach to accounting theory. He expresses his preference for inductive models over deductive models in the following words:
This type of inductive reasoning to derive goals implicit in the behavior of an existing system is not intended to be pro-establishment or to promote the maintenance of the status quo. The purpose of such an exercise is to highlight where changes are most needed and where they are feasible. Changes suggested as a result of such as a study have a much better chance of being actually implemented. Goal assumptions in normative models or goals advocated in policy discussions are often stated purely on the basis of one’s conviction