Ross L. Watts and Jerold L. Zimmerman have written a remarkable paper in accounting research which attempt to provide positive theory about factors influencing companies in lobbying on accounting standards. The paper also discuss the question whether the size of the companies affect their opinion on certain standards. This paper is important because it makes contribution to the positive accounting theory development. It might be the first paper which attempt to provide empirical evidence on positive accounting theory about factors influencing management attitude on certain accounting standards. The data and the issues discussed were new in that period. It contributes to “a literature that has uncovered empirical regularities in accounting practice” (Holthausen and Leftwich, 1983 and Watts and Zimmerman, 1986 as discussed in Watts and Zimmerman, 1990 p 131). The literature has been developing since the publication of this paper in 1978 (Watts and Zimmerman, 1990 p 132). Though, the paper also brought considerable debate among researchers.
This essay intends provide critical review and highlights some important points from the paper by reviewing in four main areas: (i) research paradigm, (ii) theoretical framework, (iii) methodology, and (iv) research findings.
The following section discusses about research paradigm used in the paper. The next section explores criticisms on the theoretical framework of the research. Section four reviews the methodology used in the research. Section five examines important finding from this paper. Finally, the last section presents some conclusion about the paper.
The researchers discussed an interesting topic in accounting theory. They use positivism paradigm in the research. This paradigm gets many criticisms when it is used in the social science research. According to Collis and Hussey (2009), positivism is criticized based on five main arguments. First, “it is impossible to separate people from the social context in which they exist”. Second, “people cannot be understood without examining the perceptions they have of their own activities”. Third,” a highly structured research design imposes constrains on the results and may ignore other relevant findings”. Fourth, “researchers are not subjective, but part of what they observe. They bring their own interests and values to the research”. Fifth, “capturing complex phenomena in a single measure is misleading”(Collis and Hussey, 2009, p. 56). Furthermore, Sterling (1990, p. 98) argued that “Instead of trying to determine how to improve the present state of accounting we are told to try to determine the present state of accounting”.
However, Watts and Zimmerman argued that they “adopted the label ‘positive’ from economics where it was used to distinguish research aimed at explanation and prediction from research whose objective was prescription” (Watts and Zimmerman, 1990, p.148). Therefore, the evidence generated from researches using this paradigm will improve the development of accounting theory by providing explanations and predictions for normative research. Moreover, positivism as one of paradigm in research is used for many researches in social science today (Collis and Hussy, 2009). This paradigm is “underpinned by the belief that reality is independent of us and the goal is the discovery of theories, based on empirical research” (Collis and Hussy, 2009).
The paper starts with the discussion about factors determining management attitudes toward financial accounting standards. The discussion based on two assumptions. First, “individuals act to maximize their own utility”(Watts and Zimmerman, 1978, p.113). This assumption means that management attitudes on certain accounting standards are based on its interest to maximize its own utility. Second, “management’s utility is a positive function of the expected compensation in the future periods (or wealth) and a negative function of the dispersion of future compensation (or wealth)” (Watts and Zimmerman, 1978, p.114). The expected forms of compensation for management that are used in this research are incentive compensation and share price increase.
The discussion then continues with the description of factors affecting management wealth. Those factors are taxes, regulation, political costs, information production costs, and management compensation plans. Each factor gives different impact on the way management chooses accounting standards. The researchers argue that management tend to “choose accounting standards which report lower earnings (thereby increasing cashflows, firm value and their welfare) due to tax, political, and regulatory considerations than to choose accounting standard which report higher earnings and, thereby, increase their incentive compensation”(Watts and Zimmerman, 1978, p.118). They also argue that this prediction depend on the political pressure on the firms and whether they are regulated or not. Another consideration that management should consider is the cost of information production as a consequence of the implementation of new standards.
Based on the combination of factors affecting management wealth, they created two models that describe two possible conditions. The first model depicts the condition when the change in accounting standards will decrease the earnings of the firms and the second model depicts the condition when the earnings are increase. In short, the theory pronounces that there are three possible attitudes of the companies when the change in accounting standards decreases their earnings: submit favorable opinion, submit unfavorable opinion, or do no submit opinion. On the other hand, only two possible attitudes of the companies if the change in accounting standards increases their earnings: submit unfavorable opinion or do no submit opinion. The choice taken by the companies depends on the size of the companies.
One of the important points explained in the paper is that the researchers omit some variables in developing the theory. Two of them are described in their paper. First, they realized that, generally, the factors influencing the regulation cost of utility companies is not only net income but also operating earnings. Second, they omit political sensitivity of the firms’ industry because there is no underlying theory to discuss this factor at that time.
The theories used on this research got many criticisms. It needs “verification and replication” (Holthausen and Leftwitch, 1983 p. 100). The researchers should be able to understand and explained it well before conducting their research (Christenson, 1983 p. 20). Hines (1988 p. 661) suggested clarifying “whether their theories were universal or probabilistic, and if the latter, explained on what conceptual grounds they should be held to be so”. Therefore, it is better to test the theories first using another hypothesis before analyzing data using the theories (Hines. 1988 p. 661).
However, Watts and Zimmerman argue that although the criticisms are relevant, those criticisms placed “unreasonable demand on study” (Watts and Zimmerman, 1990, p.149). In addition, their theories provide theoretical basis for other researches in accounting. For instance, Milne (2002, p.371) argued the theories “provide the stated theoretical basis for a number of social disclosure studies”. Moreover, this paper was probably “the only paper which predicts corporate lobbying behavior as a function of firm-specific factors'” (Holthausen and Leftwitch, 1983 p. 95).
Another important point in the paper is that the researchers only focus on shareholders, creditors, and government in explaining some factors influencing management accounting choice. In fact, there are many users of financial statements have not discussed thoroughly in the paper. Those are employee, analyst-adviser, supplier and trade creditor, customer, competitors, and the public. The public includes ‘taxpayers, consumers and other community and special interest groups, such as political parties, consumer and environmental protection societies and regional pressure groups’. Each user has different interest on the firms. Therefore, they might influence the accounting choice. For instance, the accounting choice taken by the firm which has a significant impact on increasing its earnings will stimulate employees to ask for bonuses or even increase in their salaries. Conversely, if the accounting choice taken by firm has a significant impact on decreasing its earnings, it will make employees worry about their job security. Then, it might increase employee the turnover in the firm.
Finally, it would be interesting if the paper also discuss cultural differences between companies. It could be one important factor influencing management accounting choice. Each firm consists of groups of people who share their culture based on their top management’s policies. This culture will also determine the firm accounting choice. The link between accounting value and cultural value (Alexander et all. (2009) discussed research by Gray (1988) based on Hofstede (1984) classification scheme) could be used to analyze the accounting choice of the firm. For instance, some firms might choose a more conservative accounting choice because their cultural characteristic is uncertainty avoidance. It might also explain why some big size companies did not submit their opinion on the discussion memorandum.
The researchers used voluntary disclosure regarding GPLA Statements for some companies that published such disclosure to predict the position of the companies on the new standard. For the others companies, they used the studies from Davidson and Weil (1975) and Davidson, Stickney, and Weil (1976) who develop an adjusting procedure which relies solely on published financial statements and GNP deflators. Furthermore, they also constructed proxy variables based on unadjusted depreciation and net monetary assets. They argue that these proxy variables can “serve as a surrogate for the effect of GPLA earnings”.
One of main the problems is the dependent variable used by the researchers. They divided the responses into two categories, the companies which agree and disagree with the memorandum. This dichotomy does not represent the actual condition since some responses might consist of agreement on certain points in the memorandum but disagreement on other points (Holthausen and Leftwitch, 1983 p. 104). Another problem is the assumption that the companies which did not mention the compensation or reply the questioner assumed no compensation plans. This assumption may lead to the conclusion bias.
The next important point is the use of firm’s size to proxy for political cost. The decision to use this method might be affected by condition at the time the paper was written. During the 1970s, “oil companies dominated the largest of the US companies, and they were also subject to much public outrage and political scrutiny during and immediately following the ‘oil crisis’ of 1973” (Milne, 2002, p.374). However, Hilnes suggested to avoid “the use of crude proxies such as the use of firm size to operationalize the concept of ‘political cost'” (Hines. 1988 p. 661). Watts and Zimmerman (1990) as discussed by Milne (2002, p.377) finally realized that the size is too noisy as a proxy. Subsequent to their research, “empirical studies have tended to use or suggest a wider range of measures to proxy for political costs “(Milne, 2002, p.377).
They may needs to elaborate the effect of firm’s stage of growth in accounting choice. ‘The nature of management’s solution’ (including the accounting choice) ‘to each revolutionary period determines whether a company will move forward into its next stage of evolutionary growth’ (Greiner, 1972). The firm should choose the right solution (including accounting choice) in order to survive. Companies which are in the earlier stage of development will tend to maximize their profit while more mature companies will focus on sustainability issues. The focus on sustainability issues might stimulate companies to respond differently in order to survive in their environment. This argument could be an explanation why companies with different size have different respond to certain standards (Watts and Zimmerman, 1978).
In terms of analysis, the focuses of the research on political cost especially on the effect of government intervention might reduce its external validity. The analyses conducted in isolation of other factors are “inappropriate” because “accounting is used for many purposes” (Fields et all. 2001, p. 300). The proposition that corporations will do anything to support the decision of accounting procedures that minimize reported earnings might not suitable for the companies with accounting-based debt covenants. The empirical evidence shows that those companies are “more likely to choose income-increasing accounting methods” (Begley, 1990 p. 138). Another research by Sweeney (1994) on the companies which violated debt covenants found that managers tend to do accounting change to increase the income figure in the years surrounding technical defaults.
In summary, the researchers should be careful in using assumptions. They have to realize that their focus on political cost especially on the effect of government intervention might reduce external validity of the research.
Finally, the research provides important findings generated from this research. First, the empirical evidence supports the argument that the firms tend to “choose accounting standards which report lower earnings due to political and regulatory considerations”. Those firms are the firms that have contact with the government directly through regulation or indirectly through possible government intervention. Their findings also confirm that big firms are more likely to be subjected to governmental interference. The existence of costs generated by government intervention may have more fundamental and important effects on its investment-production decision if the potential costs of government interference become large.
Based on this research, McKee et all. (1984. P. 658) then replicate the research using a larger sample. Unfortunately, the results from this research are not satisfactory. The evidences do not support the theories provided by this research. McKee et all. (1984. P. 647) argued that there are some problems regarding the test procedures and its underlying theories.
However, the findings generated from this research are now become important consideration in conducting researches in accounting especially regarding earnings management. Research in this topic usually aims to find companies that tend to choose accounting standards which will affect their earnings. Therefore, most of researchers in this topic eliminate companies in regulated industries (utilities and financial companies) from their sample to minimize the effects of political and regulatory considerations (For example: Madhogarhia et all., 2009; Roychowdhury, 2006).
In summary, this paper is a remarkable paper in accounting research. It gives starting point in the development of positive theory in accounting. It might be the first research that investigates factors influencing management accounting choice. Though, there are some important points need to be improved in this research. The most important point is the improvement in theoretical framework.
Based on the discussion on the paper, it could be concluded that accounting choice is a function of internal and external influences. Internal influences comprise some factors such as culture and stages of growth of the companies while external influences comprise changes in tax, regulation, information production cost, and management compensation costs. Combination of those influences affects the attitude of the company to certain changes, including the attitude toward the change of certain accounting standards.
Iin order to improve the development of the theory in this topic, Fields et all. (2001, p. 299-300) made three recommendations. First, researchers should give more efforts to investigate the nature of the implications of alternative accounting methods rather than just testing the implications. Then, the research should not focus only to one accounting issue or one goal. Even though the tasks become more complex, they will give a better contribution to the development of theoretical framework in accounting. Finally, the researchers should “develop more powerful statistical techniques and improve research design” (Fields et all. 2001, p. 300).