Kaplan, R. S. 1993. Research opportunities in management accounting. Journal of Management Accounting Research (5): 1-14.
Summary by Mohamed Gomaa
Ph.D. Program in Accounting
University of South Florida, Spring
Field Studies Main Page | Research Methods Main Page
The main purpose of this article, as stated by Kaplan, is to offer some personal views on what he considers to be the more and less promising opportunities for research in management accounting. To illustrate his views, he emphasizes the areas in which he has had direct involvement, such as activity-based costing and performance measurement. He briefly examines areas in which he has had less direct experience, such as research in management control systems or incentive compatibility. The paper contains Kaplanís personal observations on the research that he considers most promising for management accounting.
Limitations of Statistical Analysis to Test Emerging Theories
Kaplan believes that applying traditional research methods, analytic modeling and statistical analysis is not the most promising for exploring contemporary management accounting issues. He illustrates his views with research that has been done on activity-based cost (ABC) systems. Using an activity-based cost system, the costs of resources used for organizational activities are measured, while the resources supplied to perform activities are recorded as expenses either in budgets or in the actual costs. Therefore, standard correlation or regression analysis of overhead costs against variety and complexity factors provide mis-specified tests of ABC theory. Kaplan believes that by attempting to correlate period expenses with measures of the variety and complexity of operations within a period, the alternative to the null hypothesis in this research would be:
(a) "the costs of resources used to handle variety and complexity, after controlling for volume effects, is greater than zero; and,
(b) managers adjust, each period, the supply of resources to the variety and complexity demands actually experienced that period."
If the ABC assumption stated in (a) is untrue or if assumption (b) is violated because managers donít adjust the supply of resources each period to actual demands may result in a failure to reject the null hypothesis. It is difficult to test the theory by examining existing practice if the implications from ABC theory are not already imbedded in the practice.
Kaplan believes that empirical analysis is not useless; it is just difficult to do with any credibility. He believes that researchers should try to guide their empirical investigations by the theory being tested. This will help them develop more powerful testing procedures. He believes researchers should not use research methods just because the methods have become "accepted" for testing other theories or because they have been trained to use those methods.
The Role for Analytic Research
Another research approach to studying ABC systems is to apply analytic methods. Using this approach, researchers construct scenarios in which the information produced by approximate ABC systems that they have designed leads to product costs that are further from the "true costs" than using a simple traditional system. Kaplan believes that this research approach has been beneficial in cautioning practitioners that not every ABC system design will benefit them.
Role for Design Research versus Analysis Research
When initial attempts to implement a new technical theory are unsuccessful, we should not conclude that the innovation is flawed or that further research to improve it is unwarranted. Research should be done to answer a more interesting question, which is how to design and implement the new technical theory without failures. Kaplan believes that the new research agenda for management accountants should encompass more design and less analysis. He believes we should take basic principles and apply them to the new environment in which management accounting is being practiced.
Role for Field Research
Several researchers have suggested that management accounting research make a more extensive use of field research to study management accounting phenomenon. Kaplan believes that standard field research can be used to study the match between the demand for new management accounting procedures and the supply of innovative practices. Traditional research methods such as empirical analysis of large data sets or analytic models are less helpful for management accounting research when major structural changes are occurring in organizations. Kaplan believes this leaves management accounting researchers with three choices:
1. find areas of study where widespread adoption has already occurred so that normal science methods can be productively employed to study "what-is";
2. engage in case study and longitudinal research methods to study "whatís-new," areas where adoption is underway; and
3. study areas where no adoption exists.
Kaplan describes two dissertations at the Harvard Business School that tested theories that were articulated twenty or more years ago in operations management. Both researchers could collect internal data from organizations with enough variation among the independent variables to analyze and test their theories about the dependent variables. Both studies could also be executed and evaluated against accepted standards developed for field study research. Kaplan uses these success stories to show that opportunities exist for management accounting researchers to identify new theories that have been adopted in practice in different management disciplines that have yet to be subjected to careful empirical testing.
"Whatís New" Research
Another research approach would have researchers observing and documenting the changes and innovations that are underway in organizations. Using this approach, researchers can associate themselves with these organizations to become more familiar with the circumstances of these experiments and the process of implementation and change. Kaplan believes that opportunities exist to study the extension of newly developed measurement procedures outside the manufacturing setting in which these procedures were initially developed. This type of research could also explore innovative management accounting systems in different countries. In order to perform this type of research, the researcher must identify opportunistically innovative companies and conduct in-depth observation and documentation to describe the management accounting innovation.
Several researchers have suggested that academic researchers in management accounting can be active participants in the change process. This action-oriented approach can take several forms:
1. New settings;
2. Implementation; and
Another stream of "to-be" research is to explore the wide set of issues that arise when attempting to implement new management accounting concepts. This research would allow behavioral and organizational accounting researchers to study actual phenomenon. A third research opportunity described by Kaplan would be to explore the integration of new cost and performance measurement systems with many other initiatives occurring today in organizations.
Who Does What?
Kaplan believes that doctoral students and new assistant professors should gain expertise in the use of normal science methods. He believes that if they donít gain these skills early in their careers, they are not likely to learn them later. Action research requires a greater maturity and knowledge of individual behavior. This knowledge needs to be accumulated over time and is usually not present during doctoral studies or non-tenured faculty years.