FINANCIAL ACCOUNTING: AN EPISTEMOLOGICAL RESEARCH NOTE

This research note is the result of the authors’ refl ections on epistemological issues in respect to the fi nancial accounting fi eld. From an epistemological perspective, this document attempts to trace the philosophical, historical, sociological, and discursive research perspectives that have guided academic research in the fi eld of fi nancial accounting. In order to do so, this document explores the distinctions and connections between accounting theory and accounting practice, which we believe is the fi rst step towards understanding accounting as a scientifi c discipline. We analyze the theories underpinning fi nancial accounting research, discussing its purposes, historic evolution, and scientifi c methods used. This document also discuss the sociological and discursive contexts of fi nancial accounting in order to demonstrate that, like every other social science, accounting research is based upon assumptions about the nature of it players, or social networks. This document does not have the pretension to cover or close the discussion about all the pitfalls of this complex topic. In this sense, we try to document our analysis and draw some arguments in order to offer evidence for further discussion.


INTRODUCTION
One of the great attractions of conducting applied research in accounting or any other business fi eld is that the research topics extend beyond the normal boundaries.It brings worldviews into confl ict, improves accounting practice, and raises doubts about professional social and values.
This research note is the result of the authors' refl ections on epistemological issues in respect to the fi nancial accounting fi eld.The objective is to discuss some of the principles that guide current applied research in fi nancial accounting, without presuming to cover or propose solutions to all the pitfalls of this complex topic.More specifically, from an epistemological perspective, this document attempts to trace the philosophical, historical, sociologi-cal, and discursive research perspectives founded in the fi nancial accounting literature.
The remainder of the paper is organized as follows.In the fi rst section, we explore the distinctions and connections between accounting theory and accounting practice, which we believe is the fi rst step towards understanding accounting as a scientifi c discipline.The second section outlines the purposes, historic evolution, and scientifi c methods of fi nancial accounting research.In the third section, we attempt to clarify the sociological and discursive contexts of fi nancial accounting.Finally, we draw some conclusions.

ACCOUNTING THEORY AND ACCOUNTING PRACTICE
What is accounting?It is amazing how this simple, basic question has never been answered precisely (KAM, 1986).A simple and widely-held concept of accounting is the process of identifying, measuring, recording, and communicating economic information about an organization so that it may be used for sound decision-making.This concept was derived from WELLS (1976), and like other concepts of accounting, it emphasizes the application aspect of accounting knowledge.Viewing this defi nition from an epistemological perspective, one might argue that the object of study is not well defi ned, the methodology (truth criteria) is not identifi ed, and the purpose of accounting research is poorly delimited.The aim of this document is not to criticize this specifi c definition, but to argue that one of the diffi culties in understanding accounting as a scientifi c discipline resides in its defi nition as stated in the literature.Among others, the importance of viewing accounting as a scientifi c fi eld is that fundamental or applied research is the only way to generate and improve knowledge in a scientifi c fi eld.In other words, the relevance of, and incentives for, conducting research in a specifi c discipline like accounting depend on the extent to which specifi c methods may be applied to improve the discipline's body of knowledge.
Following this line of reasoning, we believe that in order to perceive and appreciate accounting as a scientifi c fi eld, a fi rst, essential step would be to understand the distinctions and associations between accounting theory and accounting practice.According to the framework proposed by KUHN (1972), for example, we may conjecture that accounting theory is a body of statements or propositions connected by rules of inferential reasoning (i.e.testable hypotheses or premises and conclusions) that form the general frame of reference for the development or explanation of accounting practices.The study by HENDRIKSEN (1982) corroborates this argument, adding that accounting theory may be defi ned as logical reasoning in the form of a set of broad principles that: (1) provide a general frame of reference by which accounting practice can be evaluated, and (2) guide the development of new practices and procedures.
According to these principles, we argue that the next step in perceiving accounting as a scientifi c fi eld would be to identify the accounting theories that are being developed and how they are verifi ed.In this respect, POPPER (1982) suggests that accounting knowledge is a body of normative and positive empirical theories built around inductive inferences.
"Normative" means that accounting theories contain imperative value judgments stemming from factual statements about the object of study, e.g., the market value of fi rm equity.Another justifi cation is that normative conclusions are very often the origin of policy recommendations, which may or may not be adopted by practitioners in the fi eld.According to WATTS & ZIMMERMANN (1986), normative theories are almost entirely devoted to the examination of questions of "what ought to be done."Thus, this theory attempts to prescribe what information ought to be communicated and how it ought to be presented.In other words, the normative theories attempt to explain what accounting "should be" rather than what accounting "is." On the other hand, positive theories attempt to explain why accounting is what it is.They describe not only what accounting information should and how it should be communicated to its users, but also why accountants do what they do and the effects of all this on people and resource utilization (CHRISTENSON, 1983).However, as suggested by SCHROEDER and CLARK (1995), ideally there should be no such distinction (normative versus positive) because a well-developed and complete theory encompasses both what should be and what it is.
The empirical and inductive attributes of accounting theory are easier to justify.In fact, according to STERLING (1970), only mathematics and logic can be classifi ed as non-empiri-cal sciences.Accounting theories in particular are fundamentally based on experience and observation.For example, the qualitative and quantitative variations of fi rm equity studied in Financial Accounting, or the dysfunctional behaviors of budgetary control investigated in Management Accounting.However, accounting premises and conclusions are connected by inductive inference.Double-entry bookkeeping system can serve to illustrate this point.The double-entry system is based on noting changes in the wealth of a fi rm and an attempt to translate the qualitative and quantitative variations in the fi rm's equity.The double-entry system, perhaps the fi rst and most important paradigm 1 of accounting science, was invented in the commercial city-states of medieval Italy in response to the emergence of trade and commerce.According to ROVER (1938) the double-entry was born when people came to see that you could not take something out of one pigeonhole without putting it into another.It has emerged as a natural outcome of the evolutionary process to the need of times (KAM, 1986).The fi rst published accounting work was written in 1494 by the Venetian monk Luca Pacioli (1450-1520).It summarizes principles that have remained essentially unchanged to this day.
Subsequent works written in the 16th century introduced the fi rst formulations of the concepts of assets, liabilities, and income.In keeping with this theme, LAKATOS (1978) suggests that a theory is constructed by a body of concepts.From this perspective, assets, liabilities, income, and other notions derived from these such as long and short term, revenue, costs, expenses, operational, no operational, etc., have a specifi c (or rather particular) meaning in the accounting fi eld, and are fundamental elements for the building and understanding of accounting knowledge.In the same line of thinking, the study by GLAUTIER AND UNDERDOW (1976) suggests that the concepts of fi nancial accounting are particularly signifi cant to the development of accounting theory in two ways: (1) they are themselves part of an empirical process for developing rules of fi nancial accounting, and (2) they refl ect the infl uence of institutional forces which shape the philosophy of accounting in a given and social environment.
Much later, the Industrial Revolution drove the need for accounting practices that could handle mechanization, factory-manufacturing operations, and the mass production of goods and services.With the rise of large, publicly held business corporations owned by absentee stockholders and administrated by professional managers, the accounting role was further redefi ned.According to SCHOROE-DER, CLARK and CATHEY (2005) the Industrial Revolution brought the need for more formal accounting procedures and standards.In terms of epistemology, these two events may be interpreted as a crisis (KUNH, 1972) in accounting science.Organizations were immersed in a new social and economic reality.New paradigms were imposed onto management activities, calling for new accounting theories to support the new accounting practices.From that point on, research in the accounting fi eld split off into two directions: fi nancial accounting and management accounting.While the fi rst focuses on the outside user of accounting information, the second focuses on the internal user and the decision making process.However, independently of this split, theorists continue their quest: explain accounting practice.In the next section we discuss the purposes, evolution and methods of fi nancial accounting research.
1 Both inductive and deductive inferences may generate positive or normative theories.Positive theories attempt to explain what and how accounting information is presented and how it should be communicated to users.Normative theories attempt to prescribe what data ought to be communicated and how they ought to be presented, that is, they attempt to explain what should be rather than what is.WATTS & ZIMMERMANN (1986).

FINANCIAL ACCOUNTING RESEARCH: PURPOSE, EVOLUTION, AND METHODOLOGY
Financial accounting is commonly considered as the process by which a company discloses information about its fi nancial and economic activities to different users outside the organization (CHAMBERS, 1966).Following this line of reasoning, we may argue that fi nancial accounting theories are responsible for setting the rules and principles that guide current fi nancial accounting practice.More specifi cally, economic theories such as agency and stewardship theories are the underpinnings of accounting standards that require the disclosure of fi nancial information.Also, they are the foundations of the frameworks used by external users to interpret this information.For example, the United States' FAS -Financial Accounting Standards' body of rules, among the most well-known standards in fi nancial accounting, is used as a reference worldwide for preparing, disclosing, or interpreting fi nancial statements.
In the United States for example, fi nancial accounting studies proliferated after the passing of the Securities Acts of 1933 and 1934, which created the Securities Exchange Commission (SEC), a government regulatory agency.Regarding SEC's role in the development of accounting principles, Section 13 (b) of Securities Acts of 1934 states: "The commission may prescribe, in regard to pursuant to this title, the forms in which required information shall be set forth, the items or details to be shown in the balance sheet and the earnings statement, and the methods to be followed in the preparation of report,……" At the same time, SEC was also granted the authority to determine external auditing practices.SEC's laws insti-tutionalized the corporate audit and regulated the fi nancial disclosure of corporations listed on the stock exchange.In this respect, (1982) suggests that the creation of the Security Exchange Commission represents a sea change in basic accounting thought.The SEC's regulatory role changed the goal from presenting information to management and creditors to providing relevant fi nancial information to investors and stockholders, a completely new paradigm for fi nancial accounting.
Although SEC had the power to regulate accounting practices and disclosure, in general it has heavily relied on accounting professional and used its power to set constraints and exert veto power (BELKAOUI, 1993).In this same line of reasoning, KAM (1986) states that the philosophy of the SEC has been to permit the private sector to take the leading in formulating accounting principles.In this sense, accounting theorists have became increasingly concerned with prescribing what information should be reported and how.As a result, the accounting fi eld has seen the creation of a plethora of committees to discuss, investigate, and propose new accounting principles and procedures.According to WATTZ and ZIMMERMAN (1986) "accounting theorists became much more concerned with prescribing how fi rms should report.Very little concern was exhibited for the empirical validity of the hypotheses on which the normal prescriptions rested."Under this normative approach, accounting properties are evaluated in terms of their perceived proximity to the ideal criteria.
That being said, we must also emphasize that the normative approach adopted by fi nancial accounting research in its early stages still predominates today.Two factors appear to have infl uenced this thinking.First, the main guidelines given to accounting researchers was, and still is, that specifi c accounting theories should conform to the general principles, which were believed to be grounded in current fi nancial accounting practice.As HENDRIKSEN (1982) points out, if we wish to evaluate current accounting practices, it may be necessary to start identifying the procedures and rules applied in the accounting practice, and from that attempt to establish general presuppositions and theories.
Second, from a sociological perspective, the evolution of fi nancial accounting research was hampered by the involvement and regulatory oversight of governments and professional accounting associations.For instance, the American Accounting Association (AAA), which committees' comprised mainly of accounting educators, and has as a mission "to foster worldwide excellence in the creation, dissemination and application of accounting knowledge and skills", has infl uenced the development of accounting thought by its research and publication (KAM, 1986).
According to RYAN (1992), simultaneous signifi cant research efforts were being carried out by economists and economics-oriented accountants in the United Kingdom, mainly at the London School of Economics (LSE).Accountants interested in distilling principles from existing practices (the normative approach) were producing inductive theories based on rationalizations of prevailing practices, whereas economists were using economic analysis to critique basic accounting methods.WHITTINGTON (1987) defi ned the fi rst approach as empirical inductive and the second as deductive.These different approaches began to converge after the Second World War, leading to the research of the 1960s, known as the "golden age" of a priori research in fi nancial accounting, apparently under the strong infl uence of the critical ethos of POPPER (1982).According to our analysis, we argue that fi nancial accounting researchers embraced a hypothetical deductive methodology, and attempts were made to derive measures of income that conformed to economic theory and at the same time satisfi ed the requirements of existing accounting practices.The works of CHAMBER (1966), STERLING (1970), andMATTESSICH (1970) date from that era.It was the fi rst major turning point in accounting methodology.
At that time, one of the primary assumptions in fi nancial accounting was that income measures were needed to assist shareholder and investor decision-making in their ownership interests.This supposition led to the development of the predictive ability criterion in fi nancial accounting research.The predictive approach steered a new set of empirical studies towards the effects of accounting information on decision-making.These studies examined, for example, the behavior of decision-makers faced with different types and reporting intervals of accounting information.Owing to the diffi culty of determining future cash fl ows, these studies generally contented themselves with examining the relationship between accounting information and current share prices.The seminal works of BEAVER (1968) and BALL and BROWN (1968) represent the shifted in accounting research, as the normative approach began to be replaced by the informational approach.This is refl ected in fi nancial accounting research in information economics, securities prices, and behavioral sciences (BEAVER, 1998).
This kind of research enjoyed great popularity during the 1970s, and was mostly grounded on fi nance theories such as the Capital Asset Pricing Model (CAPM) and the Effi cient Market Hypothesis (EMH).Accordingly, accounting studies from that period are characterized by the progressive introduction of fi nance research methods, accompanied by the growing presence of positive theory and methodology.As a result, empirical works were marked by contradictions between the positive concept and hypotheses that underlay the normative prescriptions of early accounting.
This shift to an informational approach in fi nancial reporting is also emphasized in the Financial Accounting Standards Board's Statement N o 1 (1978), which states that: "Financial reporting should provide information that is useful to present and potential investors and creditors and other users in assessing the amounts, timing, and uncertainty of prospective cash receipts" According to Beaver (1998) one of the reasons for this shift was that the concept of economic income had not been well-defi ned when there were imperfect or incomplete markets.
In 1986, a new methodology was formally introduced into fi nancial accounting research, with a signifi cant impact on research directions.It was called Positive Accounting Theory,2 and was invented in North America at the Universities of Chicago and Rochester by WATTS & ZIMMERMAN (1986).Positive Accounting Theory represented an extreme form of empiricism, and a reaction to the normative methodology of the a priori theorists that had reigned for previous decades.
The main argument of WATTS & ZIMMERMAN ( 1986) is that scientifi c research may concern itself only with "what is" questions, that is, it must follow a predictive approach.In other words, accounting research cannot be used to answer "ought to be" questions (the prescriptive or normative approach discussed previously).Thus, positive accounting postulates that accounting research can be used to predict stock market reactions to the disclosure of accounting information, but cannot prescribe how income ought to be measured in fi nancial statements or how stock prices ought to be evaluated based on accounting information.The authors of the positive accounting theory also advised careful discrimination between positive and normative propositions.For example, the positive perspective is concerned with how the world works and how accounting information is interpreted.Let's say a fi rm switches its inventory evaluation method from FIFO to LIFO3 .Because the stock market has not anticipated the change, the stock price rises.This result can be interpreted under the positive perspective ("what happens").A normative proposition however, being prescriptive, would suggest that, because prices are rising, the LIFO system should be adopted.This epistemological change in fi nancial accounting theory can be thought of as a revolution (KUHN, 1972) in accounting science, in which the consensus associated with the old paradigm is replaced by consensus on the new paradigm, in this case, a predictive approach.
The Positive Theory was generally welcomed into financial accounting research, inspiring Effi cient Market Hypothesis testing and research based on the assumptions of neoclassical economics, portfolio, and agency theories.Since then, these topics have been the predominant focal areas in fi nancial accounting research.Notwithstanding, in recent years the Positive Accounting Theory has been subjected to harsh critiques from positive theorists like CHRISTENSON (1983) and WHITTINGTON (1987).Its detractors have pointed out the impossibility of divorcing empirical testing from theoretical assumptions.The most telling argument is that the design of any empirical test requires theoretical constructs, and a positive theory inevitably contains theoretical assumptions.
In our opinion, here again the evolution of fi nancial accounting research, theory, and practice can be viewed through KUHN's (1972) concept of science.According to this author, criticism should be interpreted as a pre-paradigm stage, during which a body of phenomena is examined by scientists espousing competing schools of thought with no common body of belief.As FEYERABEND (1979) puts it, "La science est anachiste" (science is anarchist).If this turmoil is inherent in the very scientifi c process itself, then accounting and fi nancial accounting theory must necessarily be a contentious process.
As the new millennium approached, fi nancial accounting theory and practice seemed to be doing quite well, as fi nancial markets were booming across capitalism countries.However, in the beginning of this decade, accounting has suffered a huge loss, as public confi dence in fi nancial reporting started to erode.Fraud scandals involving corporations like Enron, WorldCom, Adelphia, Global Crossing, Parmalat, Lucent, Tyco and Xerox have resulted in fi nancial statement's credibility loss, raising questions about the integrity of the accounting and the auditing professions.
As former SEC's president states LYNN TURNER (1999) states "Given the fact that accounting is done by people, I do not expect us to be able to totally eliminate fraud.However, we shouldn't ignore it but rather proceed with timely and appropriate enforcement actions." Following this line of reasoning, we seem to be entering into a new paradigm, where the main concern is to recovery public trust in fi nancial reporting.In the United States, for example, the Sarbanes-Oxley Act, enacted in 2002, wishes to attribute higher accountability to executives and companies involving in fi nancial statement fraud.Accounting research has also been infl uenced by these events as topics like corporate governance, business ethics, earnings management, report restatement, manager's compensation and incentives, fraud red fl ags, forensic accounting, continuous auditing and whistle-blowing have been lately addressed by the scientifi c community.
In the next section we follow this epistemological analysis of the fi nancial accounting discipline exploring its sociological and discursive roots.

THE SOCIOLOGICAL AND DISCURSIVE PERSPECTIVES OF ACCOUNTING
The studies by LATOUR (1989) and WHITLEY (1984) sug gest that the sociological and discursive perspectives of a science are basically determined by the extent and intensity of its interaction with society.Like every other social science, accounting conducts its research based upon assumptions about the nature of social science and the nature of society (BELKAOUI, 1997).As it happens, fi nancial accounting may be analyzed from both the sociological and discursive perspectives.Thus, accounting may be viewed as a "socio-systemic" structure, with input, process, and output.The idea is that fi nancial accounting knowledge does not affect only the accountants and accounting practices, but also (directly or indirectly) impacts the management context in all its ramifi cations.As BEAVER (1998) suggests, the current fi nancial report environment consists of various groups (investors, information intermediaries, regulators, managers, auditors, etc.) who are affected by and have a stake in fi nancial reporting requirements.Hereafter, the sociological and discursive perspectives of accounting will be analyzed assisted by the strong interdependence between science and society (the science "players").Our argument is that, as an applied science, the accounting discipline is no exception to the rule.
In this sense, the process of constructing accounting theories has been analyzed, culminating in the conclusion that market pressure, tax laws (institutional infl uence), management decision needs, and macroeconomic factors such as infl ation are the main inputs to a sociological accounting system.These inputs are the starting points for an accounting translation process.Thus, the discursive result is the creation or improvement of accounting practices, while the sociological contribution is the correct incorporation of these aspects into the accounting framework to address user needs and serve as interpretation models.Therefore, the accounting "socio-output" is represented by better assessments of a fi rm's fi nancial health by investors and stakeholders and improved decision-making by managers.Thus, selection of a fi nancial reporting system might be viewed as a social choice, where bargain power will determine whoever gets their desires fulfi lled.
In a number of countries, such as the United States, where fi nancial reporting information is directed primarily toward the needs of investors and creditors, decision usefulness is the overriding criterion for judging its quality (MUELLER, GERNON and MEEK, 1994).However, in some other countries, such some Latin American countries, fi nancial accounting is designed primarily to ensure that the right amount of tax is collected.In this sense, accounting is shaped by the environmental forces in which it operates.
At the same time, scientifi c research in accounting has also been infl uenced by social and environmental forces, which resulted in two different streams of research: the North-American and the European.According to LOPES and MARTINS (2005) research in accounting cannot be considered independently of the social environment in which it is inserted.The research itself is a product of the social environment.
The North American stream of accounting research, which is known as the mainstream, has been based on the economic concepts and in a framework based on the positive method, which basically relies on: (i) hypotheses development (ii) economic theories to support the hypotheses (iii) empirical tests using econometrics techniques (iv) conclusions that wish to construct a theory in order to explain and predict particle This line of research has been largely disseminated by the Elite Schools (Chicago, MIT, Rochester, Stanford, etc.) and their PhDs programs.This research has also been stimulated by premier scientifi c journals like The Accounting Review (TAR), Journal of Accounting Research (JAR), Journal of Accounting and Economics (JAE), Contemporary Research (CAR) and Review of Accounting Studies (RAS).
However, an alternative stream of research has emerged with the foundation of the journal Accounting, Organization and Society, in England.Here, we call it the European or British stream, as most of the researches were originally British like Antony Hopwood, Michael Power and Peter Miler.However, we might note the existence of British authors that are adopters of the "North-American approach" and vice-versa.
The theoretical approach used by the British stream of research has been based on disciplines like sociology, psychology, history and political economy.In this line of research, the accounting phenomenon cannot be viewed within the best possible option (normative) or a set of hypotheses to be tested (positive); instead the proposition is that forces that shape accounting should be elaborated within a set of social interactions that act in a debate arena (LOPES and MARTINS, 2005)

FINAL REMARKS
This brief epistemological overview of the history of fi nancial accounting research demonstrates how it gained importance as a hands-on activity before the accounting theorists arrived on the scene.Consequently, accounting practices were shaped by accounting practitioners and the government authorities, which took a keen interest in the protection of capital markets and creditors.The capital market still wields a strong infl uence over the sociological and discursive branches of fi nancial accounting science.Research programs have been supported by regulatory bodies such as AICPA (USA) and CICA (Canada) and professional accounting associations such as AAA (USA) and CAAA (Canada)4 .Financial accounting research has also been impacted by the corporative infl uence.This infl uence has taken the form of standards designed to control fi nancial accounting practices instead of fostering discussion on the anomalies between the reality and evaluation of fi rm equity.As a consequence, few paradigms or accounting theories have been put forward to guide research avenues in fi nancial accounting.
When Positive Accounting Theory brought to accounting a theory-testing approach, researchers embarked on an efficient capital market approach, which led to improved utilization of rigorous research methods and statistical analysis.These factors may have shielded fi nancial accounting from criticism, and therefore creativity, compared to other management disciplines, where there was more incentive for qualitative and interpretative investigation.
At the same time recent accounting scandals involving highly known corporations have raised questions about fi nancial report's reliability, which seem to somehow shifted the focus back to regulation that could result in less information usefulness, in order to recover the integrity of accounting information.As such, researchers in fi nancial accounting need to be aware of the many dimensions and realities that they are attempting to "account for" and represent.Numerical Accounting highlights aspects of organizational reality that are quantifi able and built into the accounting framework, but oftentimes ignore aspects of organizational reality that are not quantifi able in this way.That said such challenges are part and parcel of all scientifi c fi elds.