1
FINANCIAL REPORTING
Inductive and deductive approaches to accounting theory
Inductive and deductive approaches to accounting theory Theories can be categorised as being discovered through inductive reasoning or derived from deductive reasoning. These differences in approaches to the development of accounting theory are relevant to an understanding of the role that a framework might perform. Both the inductive and deductive approaches to the development of theory share common basic elements. However, as illustrated in Figure 1, the logic linking the elements flows in opposite directions.
Figure 1 Comparison of approaches to theory development INDUCTIVE (specific ➝ general)
DEDUCTIVE (general ➝ specific)
Objectives
Objectives
Assumptions
Assumptions
Principles
Principles
Definitions/Actions
Definitions/Actions
2
FINANCIAL REPORTING
Discovering a theory (inductive logic) A person discovering a theory would begin the task by starting from the bottom of the left column in Figure 1—by observing definable activities and actions. Having recorded the definable activities and actions, the observer would then infer (or arrive at) the principles to which his/her observations conform. Having inferred certain principles on the basis of detailed observation, the researcher might then attempt to infer higher-level assumptions and objectives. In summary, a person discovering a theory always moves from the lower levels of the diagram to the upper levels, employing what philosophers refer to as ‘inductive logic’. Inductive logic is a process of reasoning whereby lower-level outcomes are used to identify or specify higherlevel ones. In the scientific literature, this is referred to as moving from the specific to the general (i.e. using individual observations of the phenomena to make an inference about the general population). Inductive logic can be used to develop a descriptive (or positive) theory. A descriptive theory sets out the way things are—that is, descriptive theory ‘describes’. Early attempts by members of the accounting profession to codify accounting activity on the basis of observation can be traced back to the 1920s. One of the most famous codification exercises was that carried out in the US by Paul Grady, an accounting researcher. After a long period of detailed observation, Grady (1965) produced an ‘inventory of generally accepted accounting principles’, more conveniently referred to as GAAP. In view of the inductive research methodology used, the expression ‘generally accepted’ is fitting.
Example To illustrate an inductive approach, let us consider how we might develop a theory to explain why some gains and losses are recognised in profit or loss, while others are recognised as part of other comprehensive income, that is, recognised directly in equity. The theory will be going beyond a simple explanation of compliance with accounting standards, to derive underlying principles and objectives reflected in the accounting treatment prescribed or permitted by accounting standards. Actions Under the deductive approach, we would commence with observations of gains/ losses that are recognised in profit and gains/losses recognised directly in equity. For simplicity, we will just consider a few observations of accounting treatment (actions). Gains/losses recognised in profit
Gains/losses recognised directly in equity
Downward revaluation of property, plant and equipment
Upward revaluation of property, plant and equipment
Change in fair value of held-for-trading financial instrument
Change in fair value of financial asset classified as available-for-sale
Change in fair value of a biological asset Principles The next step would be to identify principles that are consistent with the observed actions. For example, a principle that gains should be recognised in equity while losses should be recognised in profit or loss is not consistent with all of the observations. It holds for property, plant and equipment, but is inconsistent for biological assets where all changes in fair value are recognised in profit or loss. Your turn. Think of a principle that might explain one of the above observations and then test whether it is robust to the other observations.
3
FINANCIAL REPORTING
For the sake of the illustration, we will now assume that we have derived the following principles from our observations: 1 All fair value adjustments should be recognised in profit if they: relate to assets that are held-for-trading and which can be traded in a highly liquid market; or capture biological transformation that reflects the performance of the entity during the period. 2 Other fair value adjustments should be recognised in equity if the fair value is easily determined. 3 Where the fair value is not easily determined, the use of fair value should be discouraged by requiring downward revaluations to be charged against profit while not allowing upward revaluations to be included in profit. (Please note, these principles are merely made up for the purpose of illustrating inductive reasoning, and are not intended to suggest that these were the basis for the conclusions of the standard setters.) n
n
Assumptions Identify assumptions that have been made in identifying the principles, such as assumptions about performance measurement, what is meant by profit and the rationale for the observed actions. For example, the first principle assumes that profit is more relevant to measuring performance than is comprehensive income (in fact, the alternative proposition is made in accounting standards). The second and third principles assume that the measurement of the fair value of financial instruments is easily determined, while the fair value of property, plant and equipment is not easily determined. Objectives Lastly, broader generalisations are made from the principles and assumptions reached by the researcher or theorist. For example, a broad generalisation may be made about the objectives for the reporting of profit and comprehensive income, and about what they should comprise. This would provide general rules that could then be applied to new observations or emerging issues, such as how to account for changes in the fair value of assets arising from emissions trading schemes. Limitations of an inductive approach Inductive logic is useful for developing theories to describe and explain accounting practice. It is not well suited to the development of a set of conceptual and pragmatic principles that provide a general framework for accounting. The limitations of descriptive accounting theory in developing a conceptual framework include: ■ A tendency to maintain the status quo. The inductive approach does not question whether there might be better ways of doing things. Accounting is defined as ‘what accountants do’. ■ Lack of guidance on how to handle new or emerging issues and situations. As a consequence, inductively derived descriptive accounting theories encourage an ad hoc approach to problem-solving. It is clear from this that inductive accounting theories do not cope well with new ways of doing business. ■ There is no guarantee that descriptive accounting theories will produce internal consistency from a logical perspective. Contradictions can, and do, arise in that similar events could be treated differently in different circumstances. ■ Practice leads theoretical development, so that undesirable practices emerge in advance of the development of a principle, or principles, that may have prevented the emergence of the practices in the first place.
4
FINANCIAL REPORTING
Building a theory (deductive logic) A person building a theory through deductive reasoning would begin the task by starting from the top of the right column in Figure 1—by setting objectives. The developer of the theory sets whatever objectives he/she desires. Having set the objectives, the developer would then deduce (or derive) the assumptions (if any) that underlie those objectives. Once the objectives are specified, and any underlying assumptions deduced, the developer of the theory would then deduce the principles that flowed logically from both the objectives and the assumptions. In turn, the principles would then enable the inventor to deduce the definitions, activities and observable actions that should result. In summary, the inventor of a theory always moves from the upper levels of the diagram to the lower levels, employing what philosophers refer to as ‘deductive logic’. Deductive logic is a process of reasoning whereby higher-level outcomes are used to identify or specify lower-level ones. Deductive logic has also been described as the process of going from general principles to specific actions. Deductive logic can be used to develop a prescriptive (or normative) theory. Prescriptive theory sets out the way things should be done—that is, prescriptive theory ‘prescribes’. For example, Chambers used deductive reasoning to develop a prescriptive theory referred to as continuously contemporary accounting (CoCoA). An underlying premise of CoCoA is that users of financial statements need information about the financial capacity of an entity. Under CoCoA, assets are represented by current cash equivalents, measured at market exit price, if available. Profit is measured as the change in wealth after allowing for capital maintenance adjustments. (For further information about CoCoA, refer to Chambers 1955a, 1995b, 1966, 1980). Like many other conceptual frameworks of accounting, the FASB’s Conceptual Framework is derived using deductive logic. The FASB’s Statement of Financial Accounting Concept No. 1: Objectives of Financial Reporting by Business Enterprises states (1978, para. 9): Financial reporting is not an end in itself but is intended to provide information that is useful in making business and economic decisions—for making reasoned choices among alternative uses of scarce resources in the conduct of business and economic activities. Thus, the objectives set forth stem largely from the needs of those for whom the information is intended, which in turn depend significantly on the nature of the economic activities and decisions with which the users are involved. Accordingly, the objectives in this Statement are affected by the economic, legal, political, and social environment in the United States. The objectives are also affected by characteristics and limitations of the information that financial reporting can provide (paragraphs 17-23). The FASB’s Framework commences with the objective that financial reporting should provide information that is useful in making business and economic decisions about the allocation of scarce resources. This objective is underpinned by assumptions about the type of information needed by those making business and economic decisions. For example, it is assumed that the effectiveness of decision makers is enhanced by information that reflects the relative standing and performance of business enterprises (FASB SFAC No. 1, para. 16). The major problem often associated with deductively derived, prescriptive accounting theories is that they are not based on observation. For example, prescriptive accounting theories are not driven by the observation of existing accounting practices. Rather, they are based on the opinion of the developer of the particular prescriptive accounting theory about what ‘should’ happen. In the context of a conceptual framework, there may be disagreement about the desirability of an objective. For example, the objective of financial reporting proposed by the FASB and IASB, to provide financial information about the reporting entity that is useful to present and potential equity investors, lenders, and other creditors in making decisions, has been criticised for ignoring the stewardship function of accounting (for discussion, refer to O’Connell 2007). A related limitation of normative theory is the potential invalidity of assumptions underpinning the objectives and qualities of information that would meet users’ information needs. (For further discussion, refer to Walker 2003.) Although prescriptive theories are not driven by observations, there is potential for empirical accounting research to inform the choice of objectives and the assessment of the validity of assumptions. For example, considerable accounting research has been undertaken to assess the decision relevance of various financial and non-financial disclosures.
