Summarize this POSITIVE ACCOUNTING THEORY and write what you understand

According to Watts & Zimmerman (1986, p.2), positive accounting theory is concerned with the explanation and prediction of accounting practices. It provides reasons for observed practices and predicts the occurrence of unobserved accounting phenomena. It is "positive" because it deals with "facts" about how the world works as opposed to normative statements and theories which are concerned with prescriptions about how the world should work. Claiming the authority of Milton Friedman and other economists for this distinction, the authors appear oblivious to the considerable critical literature about "positive" economics (see, for example, Caldwell, 1982, Chapters 8,9; Coddington, 1972; Eichner, 1983; Musgrave, 1981) and the limitations of empiricist philosophies of science. They also appear ignorant of recent discussions about the appropriateness of natural scientific methodologies for the human sciences and of engineering models for social intervention technologies (Bhaskar,1979, Chapters 2,3; Fay, 1975, Chapters 2,3; Stockman, 1983, Chapters 6,7; Thomas, 1979, Chapters 3,4; Turner, 1980). The idea that accounting researchers can simply construct general theories of accounting practices which will enable practitioners to predict the outcomes of their decisions in an uncertain world by following methodological rules based on neo-classical equilibrium economics is, to say the least, tendentious, and open to considerable scepticism.

The ability of Watts & Zimmerman to ignore much of the literature in the philosophy of the natural and social sciences while making methodological assertions exemplifies the fragmented nature of accounting research as an intellectual field and the lack of strong theoretical norms integrating research results (Whitley, 1984). Watts & Zimmerman consider recent accounting research to be largely constituted by empirical "tests" of economics-based theories about the role of accounting information in market changes. Although they do refer to some other work, the bulk of their discussion of positive accounting theory focuses on the extension of equilibrium theories to capital markets and the disequilibrating effects of new accounting information. Thus, there are chapters summarising research on the Efficient Markets Hypothesis and the Capital Asset Pricing Model, the effects of earnings announcements, disclosure regulations and changes in accounting conventions on share price movements, on agency contracting and on the managerial selection of conventions. Practically all this research presumes the adequacy of neo-classical economics models of markets and "tests" hypotheses by statistical analyses of standard data bases derived from annual reports and similar public information. It thus purports to follow the model of scientific work popularised by logical positivist and logical empiricist philosophers of science in which scientists develop theories and general laws which are then tested against observation statements based on experience. According to Watts & Zimmerman, this sort of research has demonstrated progress because theories have been replaced by more refined ones and technical errors have been corrected. Thus they summarise the development of empirical accounting research as a series of theoretical improvements and technical refinements.

Leaving aside the question as to whether this selection of research is fully representative of recent accounting research, and whether "positive" research is the only type worth summarising, it seems doubtful whether these studies do indeed follow the methodological rules promulgated by philosophers such as Hempel and Popper. Christenson (1983) has already pointed to the failure of the "Rochester School" to test their theories systematically and the ad hoc adjustments made by Watts & Zimmerman when faced with apparent falsifications of their hypotheses. Similarly Lowe et al. (1983) produced empirical evidence which appeared to contradict their account of the development of accounting theories in their 1979 paper, yet the bulk of that paper is reproduced as an explanation of the growth of accounting research in this book. This cavalier approach to empirical tests is echoed by their rather simple account of intellectual change and development which assumes that scientific progress consists of methodological improvements to account for "anomalies". Whereas Popper ( 1968, pp. 81-111 ) suggested that scientists should reject theories which were found to be inconsistent with observation statements, and progress consisted of replacing falsifted theories by those with greater empirical content and verisimilitude, Watts & Zimmerman seem to applaud the continued acceptance of the EMH and the CAPM despite their immunity to empirical tests (Findlay & Williams, 1980) and are also unable to demonstrate that the empirical content of accounting theories has increased. Indeed it is not clear how any theory which claims to describe equilibrium states of "perfect" markets in which information is costless and true could be considered to have empirical content (cf. Whitley, 1986). Since, however, similar points have been made about economics, this sort of accounting research may be no worse than other pseudo sciences (Blaug, 1980, pp. 253--264; Caldwell, 1982; Deane, 1983; Hutchison, 1976, 1984).

If the research summarised by Watts & Zimmerman does not follow the methodological precepts which they claim characterise "science", then the epistemological status of its knowledge claims is clearly in doubt. A further possibility, of course, is that Popper's methodological rules are incoherent and/or in[1]appropriate to knowledge about these sorts of phenomena. In particular, they may not be applicable to research in the human sciences (Findlay & Williams, 1985). This latter possibility is not mentioned by Watts & Zimmerman because they consider the scientific method of hypothetico-deductivism to be applicable to all phenomena and to be the only valid way of generating and assessing truth claims in all the sciences. The overall adequacy of Popper's account of scientific progress is likewise not considered by them, they offer no reasons for preferring it to, say, realist accounts (Bhaskar, 1975; Harr6, 1970) or show any indication that they are aware of alternatives. In fact, the only justification for claiming the authority of Popperian epistemology seems to be its use by Blaug (1980) in his analysis of economics.

Generally, Watts & Zimmerman appear to rely a great deal on the dominant orthodoxy of neoclassical economics in justifying positive accounting theory and their book could be seen as presenting a methodological case for reducing accounting research to a branch of economics. Depending on one's view about the epistemological status of orthodox economics and its validity as an intellectual enterprise, this foundation for accounting research may seem less than secure, especially if researchers are concerned to explain why social events and phenomena occur and change and help to resolve practical problems. As Findlay & Williams (1980, 1985) point out, drawing policy recommendations for financial analysts and managers in an uncertain and internally related world from theories that reduce uncertainty to risk is unlikely to be very effective. Similarly, presumptions of "perfect" competition, homogenous expectations, costless information and negligible transaction costs render much financial economics and positive accounting theory remote from the world of practitioners and incommensurate with their concerns (Whitley, 1986). The dominance of such presumptions in economics does not, pace Jensen (1972) and Ross (1978) automatically justify their correctness or demonstrate their relevance for everyday problems in uncertain and interdependent social worlds.

The epistemological status of recent accounting research has been discussed by a number of writers in the past 5 or so years, (e.g. Ryan, 1982; Thomas, 1981; Tomkins & Groves, 1983). In many cases, a particular philosophical position 634 R.D. WHITLEY has been assumed and then used to evaluate a body of research without a great deal of attention being paid to the purpose and adequacy of that position. As we have seen, Watts & Zimmerman claim the authority of Popper for their view of science and its generalisability to accounting research, a commitment that is shared by some of their critics, such as Christenson (1983). In considering how appropriate such a commitment is for accounting researchers and those urging particular sets of methodological rules for them, it is clearly important to understand the purpose of such rules and the extent to which they fulfil it as well as its appropriateness for analysing the epistemological status of accounting research. Before, then, discussing the issue of whether theories of natural scientific knowledge are necessarily applicable to the human sciences, I shall briefly raise a number of issues about the nature of the Popperian research programme in the philosophy of science and its success in providing a theory of progress of scientific knowledge. Although I focus on Popper's work, many of the points raised are applicable to other forms of logical empiricism