Uncertainty as a determinant of performance measurement and compensation systems

Performance measurement and compensation systems (PMCS) are a critical means to align individual and organizational goals. Accounting researchers have devoted considerable effort to identifying the determinants of PMCS as well as the conditions that influence its effectiveness. The problem becomes challenging when the decision context is uncertain as it is difficult to identify measures that adequately capture the performance of individuals. The ‘wrong’ measures can lead to misdirected effort and worse fraudulent behaviour which is not only costly for the firm and the individuals within it but also influences the efficient function of the market and hence has wider implications for society’s well-being. We only need to look at the consequences of Enron, Worldcom, and, more recently, the global banking crisis to appreciate the economic and social consequences of PMCS. One of the challenges in designing PMCS in uncertain decision contexts is that it is difficult to implement measures that adequate capture the efforts of managers. Environmental uncertainty, for example, can lead to outcomes that are independent of the actions taken by employees. It is thus not surprising that much of the empirical research explicitly or implicitly is concerned with the impact of uncertainty on the design of these systems.

We adopt a problem-focus in our research in our approach by examining empirical research relating to one of the major challenges facing organizations, namely how to ensure that individuals behave in ways that achieve organization goals and objectives efficiently and effectively. This is a complex problem, particularly in organizations faced with uncertainty in the decision making process. We draw on empirical papers published between 2003-2013 in seven international accounting journals in our attempt to synthesize the literature. We focus on the design and use of performance measurement and compensation systems (PMCS) and synthesize the literature around an organizing theme, namely uncertainty. Our aim is to develop a framework for future research. Our motivation is simple. The literature is not integrated and recent experiences in teaching doctoral students highlight the difficulties. Students cannot easily identify what we have learned and what there is left to learn; nor can they understand why researchers adopt different paradigms that typically don’t talk to each when addressing this critical management problem. We attempt to develop a framework that will enable us to provide direction to those wishing to contribute to this challenging problem!

We use the term uncertainty to capture a variety of contextual (exogenous) factors. Thompson (1967, p. 159), one of the very early researchers of organization behaviour, identified uncertainty as the fundamental problem for complex organizations and coping with uncertainty is the essence of the administrative process. Researchers have included different variables in their empirical models; this often reflects the different paradigmatic approaches to the research problem. For example, those researchers drawing on principal/agent models typically conceptualize uncertainty as information asymmetry or as a function of task interdependencies. These researchers typically control for factors such as growth, competition and volatility given the effect that these factors might have on the design of PMCS. Those drawing on organizational theory typically focus at the firm or business unit level and describe the decision context in terms of task uncertainty or environmental uncertainty. The psychology and social psychology literature focuses on the uncertainty at the individual level, whether it relates to the uncertainty an individual faces in performing their task or the inability of a superior to evaluate performance of her subordinates due to their cognitive limitations in complex decision settings.

Our overview attempts to integrate prior research in order to better understand what we have learned from the studies undertaken. Merchant et al (2003) provide evidence of the absence of cross referencing across the alternative paradigms. For example, economics and psychology frameworks both assume that human beings exhibit bounded rationality. However, their frameworks are typically based on different assumptions. The traditional principal/agent model assume that agents are opportunistic, risk-averse, prone to moral hazard and slacking, and have conflicting interests to those of the principal. In contrast, those adopting a psychological perspective, assume agents have a co-operative or collaborative nature, inherently seek fairness and reciprocity, and prefer a working environment based on trust. Common to both approaches, however, is the importance of uncertainty, in numerous representations, as a key determinant of PMCS design.

In summary our objectives are as follows:
1. to establish the state of knowledge in the accounting literature of the determinants, the use and the effects of PMCS;
2. to explore and analyse the role of uncertainty as a critical determinant in the design of PMCS in complex settings;
3. to identify common themes, highlight issues, and indicate potential topics for future research.
4. to develop a framework which integrates the alternative approaches to PMCS with a view to providing directions for future researchers interested in addressing this complex problem.