Accounting is an important aspect of management control. Budgets are unarguably the most obvious form of utilising accounting data to monitor and punish or reward strategic business units and consequently employees, regardless of whether they are managers or workers on the shop floor, according to their performance in relation to budgeted targets. “The budget is a financial plan for implementing the various decisions that management has made” (Drury 1997, p. 8). Participation in the formation process of budgets by those ultimately affected is practised in companies with the aim to generate a better-performing workforce. Empirical evidence on the effects of participative budgeting is ambiguous and the literature is fragmented. In this paper, I shall mainly review research on participative budgeting as well as other issues in budgeting and some critical perspectives on budgeting as a means of management control.
Historical evidence of accounting and management control
Chan, Lew and Tong (2001) analysed the accounting and management control practices described in the popular Chinese novel “A Dream of the Red Mansion”. They found that big family households of the early Qing Dynasty (18th century) recognised the importance of, and made distinct achievements in, accounting and management controls. “They mastered the segregation of duties, the control of cash, the use of budgets for planning, the containment of costs and the efficiency of operations” (Chan et al 2001). Besides providing evidence of accounting practices and management control systems, their work offers insights into the conditions and requirements for control systems in a specific and social environment. The latter could prove useful for Western corporations acquiring state-owned enterprises in China.
Ezzamel (1997) found evidence of preliminary forms of management accounting, control and accountability in ancient Egypt. He ascertained that accountability was used to legitimise the authority or power of state officials and that it was of ceremonial value in justifying social status.
Behavioural aspects of budgeting
Long before the papers on participative budgeting to be discussed later were written researchers started concentrating on the influence budgetary systems have on behaviour and action. In his book, Hopwood (1974) emphasises the need to see the process of standard setting and budgeting in its entirety and respond to it as a complex human and technical problem rather than one standing in technical isolation. ”Budgets are used to motivate members of the organisation by serving as targets and mechanisms for gaining involvement and commitment” (Hopwood 1974, p. 44). Hopwood (1974) realises that budgeting is concerned with human action and the significance of its behavioural, political and social dimensions. He went on to express his amazement at the dominance of a purely technical approach and called for a merger of both technical and behavioural aspects.
In 1955, Argyris looked at the problems organisations faced when introducing participative management by drawing on prior research for a concept of organisation and human personality. In Argyris’ (1955) invented scenario, employees were dependent on a leader. This lead him to question how much participation and democracy this allowed criticising that previous studies had failed to address that question. He concluded that participative management implied a different set of organisational principles and that executive training should thus focus on understanding the effects of basic dependency relationship on employees and how it conflicts with their normal personality needs. This dependency could be reduced by the introduction of participative management.
Participative budgeting and management accounting control systems
Brownell (1982a, p. 124) defines budgetary participation as “a process in which individuals, whose performance will be evaluated and possibly rewarded on the basis of their achievement of budgeted targets, are involved in and have influence on, the setting of these targets”. Brownell is a key researcher in the field of participative budgeting and has published an impressive number of studies throughout the 1980s and 1990s.
Before Brownell’s research, Argyris (1952) stated that the most significant contribution from budgeting activities would occur in the case of subordinates’ being allowed to participate in budget creation. Milani (1975), on the one hand, was unable to find evidence for a significant positive association between performance and budgetary participation but, on the other hand, found evidence that budgetary participation improves both attitude toward the job and the organisation. For this study, a continuum for the degree of participation was developed.
Up to then, research had simply been concerned with the question what outcomes participative budgeting had on the people affected and the organisation as a whole. Brownell (1981), though, was interested in its antecedents, i.e. under what conditions participative budget is effective. He investigated the role of the personality variable, locus of control, as a moderator of the relationship between budgetary participation and performance. The result was that budgetary participation had a positive effect on individuals who feel they have a large degree of control over their destiny while it had a negative effect on individuals who feel their destinies were controlled by luck, chance or fate. A year later, Brownell (1982b) carried out further research in order to validate the results of the prior laboratory experiment using a field-study approach. The previous study was extended by considering the effects of participation and locus of control on job satisfaction as well as on performance. Results of the survey phase were consistent with those of the preceding experimental phase. No supporting evidence for effects on job satisfaction was found. The strong positive association observed by Brownell between performance and participation was in stark contrast to Milani’s (1975) findings discussed earlier. It is safe to draw such a strong direct comparison since Brownell used Milani’s (1975) six-item scale for measuring participation. By combing the two phases of his study, Brownell (1981, 1982b) created both causality and generalisability of the study’s findings.
“Brownell (1982a) developed a framework for organising research on participative budgeting that divided the literature into two parts” (Shields & Young 1993, p. 265). He considered the literature to be “fraught with contradiction, overlap and a general lack of conclusiveness on the question whether participation works or not” (Brownell 1982a, p. 124). He divided the literature according to which variables they dealt with into:
- Cultural variables
- Organisational variables
- Interpersonal variables
- Individual variables
The conclusion was that most literature had overlooked the issue of participation as a response to cultural and organisational circumstances, which lead Brownell (1982a) to develop a unifying framework, in which he distinguishes between antecedent moderators (preconditions dictating the need for participation: cultural and organisational variables) and consequence moderators (influences on the effects of participation: interpersonal and individual variables). He advised that rather than researching whether participation is effective or not in general, the conditions under which it is effective should be specified.
Consequently Brownell (1983, 1985, & Abernethy 1997, & Abernethy 1999) focused on researching participative budgeting in different contexts taking into account a number of different variables.
Brownell (1983) undertook an empirical study with the aim to explore the effects of management-by-exception in a budgetary context on individual motivation. Budgets enable managers to operate a system of management-by-exception, “which means that a manager’s attention and effort can be concentrated on significant deviations from the expected results” (Drury 1997, p. 194). Hypothesising that there would be a positive relationship between management-by-exception and motivation as well as interaction between management-by-exception and budgetary participation affecting motivation, the results showed little statistical significance and the study’s explanatory power was marginal at best.
In another empirical study, Brownell (1985) investigated how different functional activities within an organisation [marketing and research and development (R&D)] influence the effectiveness of participation and the reliance on accounting information and control. Having found support for the hypothesis that there are significant differences between the R&D and marketing units he hypothesised that there would be a significant interaction between participation and functional area affecting managerial performance. The only conclusion from testing this hypothesis was that participation does not show simple and direct relationships with performance and interacts with other variables. He had also hypothesised a significant interaction between the reliance on accounting information and functional area affecting managerial performance. Reactions captured in interviews with the vice-presidents of the R&D and marketing units were consistent with Brownell’s last expectation, but empirical evidence for this could not be obtained. The main contribution of this study was the realization of the diversity of effects participation can have.
Continuing his research into antecedent moderators, Brownell & Abernethy (1997) criticised the dominance of manufacturing in management accounting research and hence looked at the nature of control system design in R&D divisions of large enterprises. Brownell et al (1997) found empirical evidence for the expectation that the reliance on accounting controls is positively related to R&D management performance where tasks are highly analysable and exceptions are few. The hypothesis that where task are low in analysability but few exceptions occur, the reliance on accounting controls is positively related to R&D management performance, was not supported by the study’s results.
Brownell & Abernethy (1999) paired up again to undertake research, this time relating more specifically to participative budgeting. They identified strong theoretical support for management accounting control systems’ being able to serve an active role in shaping organisational change. “There is, however, little broad-based empirical research examining how these systems are used in organisations facing change and with what consequence” (Brownell et al 1999, p. 189). The hypothesis of the study was that the relationship between strategic change and performance would be moderated by the extent to which budgets are used interactively. Interactive refers to the style of budget use with diagnostic located at the other end of the scale. Budgets that are used diagnostically serve the traditional purpose of evaluating performance, comparing this to budgeted targets and attributing responsibility for outcomes to corresponding organisational functions or members. When budgets are used interactively, in contrast, a continual exchange between top management and lower levels of management, as well as interactions within various levels of management but across functions takes place. “This interaction involves not only participation between subordinates and superiors in the budget setting process, but also an ongoing dialogue between organisational members as to why budget variances occur, how the system or behaviours can be adapted and even whether any action should be taken in response to these variances” (Brownell et al 1999, p. 191). Hence, the concept of interactive budgeting encompasses participative budgeting involving a more sophisticated and interactive evaluation process, thus facilitating organisational learning. Their hypothesis was confirmed. The contemporaneous relationship between strategic change and performance was found to be more positive when the style of budget use was interactive compared to when it was diagnostic. Additional analyses were performed in order to test several of the novel model’s assumptions. It was confirmed that the extent of strategic change was important in determining the degree to which top management used budgets interactively, rather than the direction of the change process.