Absence and Variant Modes of Presence of Management Accounting in New Product Development – Theoretical Refinement and Some Empirical Evidence

Abstract

The purpose of this explorative multi-organisation study is to describe and explain absence and variant modes of presence of management accounting (MA) in new product development (NPD), and to problematise further the equivocal results and paradigm shift regarding the role and relevance of management control system (MCS) packages in innovative and uncertain environments, in particular, why there are environments without formal management controls such as MA. The study refines, and in certain respects develops the theory of MA absence by combining it with the theory base of MA change and stability. The theoretical analysis suggests that a wide variety of factors may explain the absence or the mode of presence of MA in NPD. These identified explanatory factors are analysed further within the developed theoretical framework. The strongest empirical evidence for MA absence arises from technical, economic and functional factors. Furthermore, lack of reasons-for-adoption of MA systems and other forms of control associated with engineering-oriented culture lead to MA absence in NPD. The empirical findings also reveal a number of variant modes of MA presence with only the ‘accounting thinking’ and concepts adopted in NPD. The contribution of the study extends from the refinement of the theory of MA absence towards increasing the understanding of the evolution, change, stability and relationship of various elements of MA systems, and the dynamics of MCS in general.

View correction statement:

Erratum

Acknowledgements

I am grateful for the comments and suggestions offered by the two anonymous reviewers, Tony Dávila, Markus Granlund, Seppo Ikäheimo, Frank Selto and Michael Shields at various stages during the time span of the study. In addition, I am indebted to the representatives of the case firms for their collaboration during the research project.

Notes

¹This is especially the case with knowledge-intensive companies specialising in high-technology business, which are constantly facing a challenging situation in NPD, where two apparently contradictory aims – encouraging a climate of innovation and simultaneously exercising enough financial control in order to meet stakeholder objectives – should be combined (Gleadle, 1999; Granlund and Taipaleenmäki, 2005; Lukka and Granlund, 2003; see also Ditillo, 2004). Cooper et al. (1981, pp. 175, 188) claimed that accounting systems ‘might be designed to facilitate creativity and innovation in the organizations’ and that they must include playful aspect, i.e. ‘provide sufficient creativity and range of interpretation in order to allow ambiguity of goals and diversity of meaning in explaining past events and predicting future events’. It is of paramount importance to note that also traditional industries are increasingly transforming towards knowledge-intensity, dependent on innovation, as they are adapting to the developments of the information society and adopting modern information technology (IT) in their business processes.

²According to some authors, contemporary business trends, such as process-orientation, have affected the role and the organisation of MA, transforming it into a business-oriented function working closely in a cross-functional interface (Granlund and Lukka, 1998b). It would be relatively easy to argue on the basis of these trends that MA should extend to the key cross-functional processes such as NPD. As regards R&D investments, it has been noted that external interest groups, particularly institutional investors also increasingly require information on the effectiveness of R&D for the valuation of R&D expenditure (Nixon, 1998).

³This uncertainty in NPD research may concern markets, technology and project scope (Dávila, 2000) and be a consequence of the limited controllability of the outcome-input relationship (Jørgensen and Messner, 2010). Dávila (2000) also concluded that, for example, many target costing studies (Kato, 1993; Tani, 1995; Hansen and Jönsson, 2005) concur on the role of the target costing method as a communication, problem solving and learning device.

⁴The studies focusing on MA in NPD have contributed mostly in analysing the presence and the roles of entire MCS or accounting systems, or even more typically of a single accounting technique in its textbook mode of presence in R&D or NPD (see, e.g. the literature reviews by Bisbe and Otley, 2004; Dávila, 2000; Dávila et al., 2009b; Ditillo, 2004; Shields and Young, 1994). Recently, Dávila et al. (2009a) studied also the reasons why these systems are adopted in the first place within the context of NPD. On the basis of empirical evidence, they also identified the certain reasons for maintaining an informal control approach, instead of a formal MCS in NPD. According to Dávila et al. (2009a), these reasons were (1) A team has worked together for a long time and their informal interactions are well-understood but not coded; (2) the management team believed that formal systems would kill creativity; (3) the organisation was not considered to be large enough to grant MCS; and (4) the management team did not have the knowledge to implement these systems.

⁵In Choudhury's (1988, p. 549) words:

‘The first level of analysis should seek to establish the social, psychological and technical conditions that prelude or do not encourage the presence of accounting. The second level of analysis should concern itself with the researcher's ideological framework that may interfere with the recognition of accounting voids’.

⁶For example, accounting routines may exist, while accounting rules are absent (Quinn, 2011; see also Feldman and Pentland, 2003).

⁷Libby and Waterhouse (1996) found that the change in MA systems was proportionate to the number of existing formal MA systems, consistent with the view that learning and change result from experience.

⁸One stream of literature suggests that research on MA change should be based on evolutionary theory (e.g. Burns and Scapens, 2000; Scapens, 1994). Johansson and Siverbo (2009), building on the framework by Scapens (1994), develop the evolutionary approach beyond the general belief that it describes only small and gradual, slow changes, and claim that both continuity and change are seen as evolutionary outcomes. They argue further that the selection and replication of existing routines lead to continuity or stability, whereas the successful selection and replication of variation lead to what is commonly referred to as change, i.e. increased awareness of variation is an important prerequisite for change. They also point out the criticism against studies guided by institutional theory, which may have focused almost exclusively on inertia and continuity, or in micro and single entity (Johansson and Siverbo, 2009, p. 159). For this study, the evolutionary perspective of accounting change is relevant due to the fact that, whereas revolutionary change involves a fundamental disruption to existing routines and institutions, evolutionary change is incremental with only minor disruption (see Burns and Scapens, 2000). Minor steps towards accounting presence can be considered to indicate evolutionary change, and presumably introducing totally new accounting controls would be regarded as revolutionary change, within the context of NPD with absence of MA, and hence such an environment would be more apt to resistance. MA change literature includes multiple dichotomies of the change: formal and informal, revolutionary and evolutionary, and regressive and progressive. These dichotomies all constitute extremes along a continuum and may mutually co-exist (Burns and Scapens, 2000; Granlund and Modell, 2005).

⁹They (2004) reported that they found negative support for the proposition that interactive use of MCS favors product innovation particularly in high-innovating firms, although the favorable effect may exist in low-innovating firms.

¹⁰Cobb et al. (1995) developed the accounting change model of Innes and Mitchell (1990) and claimed that the prior model ignored barriers, i.e. factors which hinder, delay and even prevent change. Hence, these generic barriers may also explain the absence of management accounting.

¹¹For example, the lack of resources or organisational culture can partly explain other control mechanisms, which override a management accounting system. Especially in product development, an engineering-oriented culture may lead to the use of other engineering-oriented forms of control. Similarly, Granlund (2001) stated that, in practice, it may sometimes be very difficult to separate the economic, functional, institutional and individual counterforces from each other when attempting to explain the stability of management accounting systems. Dávila et al. (2009a) also pointed out that the reasons-for-adoption are not exclusive of each other, and more than one of these reasons may be present at different stages for different systems within a company. Undoubtedly, this also applies to reasons-for-non-adoption. In addition, Innes and Mitchell (1990, p. 4) applied contingency theory to analyse management accounting change, and stated that, for example, it is not clear whether the environment and technology affect management accounting directly or through their impact on the organisational structure, or indeed whether there is reciprocity in the links between management accounting and internal contingent variables.

¹²Even if the management and control processes would require MA support, undoubtedly there are also cases where these needs have not been identified (Choudhury, 1988).

¹³Dávila et al. (2009a) identified that the reasons-for-adoption of MCS in NPD can be based on the role of MCS, or based on certain external and internal events, such as an organisational failure. Building on the evidence from the multi-method, multi-case field study of 69 technology-based companies by Dávila et al. (2009a), it can be expected that the lack of these most important reasons-for-adoption of MCS in NPD, and the like, would be associated with the absence of MA. Furthermore, the less information tools to deal with market, technology and NPD project scope-related uncertainty (Dávila, 2000), the less reasons there are for adopting MA in NPD. Similarly, focusing on the non-adoption of target costing in NPD, Dekker and Smidt (2003) found that by far the most common reason for non-adoption was that the method was not well applicable due to the nature of the company.

¹⁴Dávila et al. (2009a) conceptualised MCS in their field research as formal systems particular to NPD including project milestones, budget for development projects, reports comparing actual progress to plan, project selection process, product portfolio roadmap, product concept testing process and project team composition guidelines. Product innovation processes, particularly in mature medium-sised and large firms, are purposeful and structured (Dougherty and Hardy, 1996), a fact which suggests that other forms of control are typically present in NPD.

¹⁵Munro (1995) suggested that when management accounting controls are absent, the situation is particularly suitable for researching other types of control.

¹⁶On the other hand, many of the explanations discussed in this paper may explain the absence of generic, corporate-wide MA systems as well.

¹⁷Granlund (2001, p. 152) wrote that

‘the inclusion of the human factor category extends the analysis beyond the economic and institutional macro-level, and the macro-level is extended to cover organizational and political cultures (micro-level institutional analysis, see Burns and Scapens, 2000). The human factors do not fit the institutional framework as such, but are assumed here to underlie all organizational arrangements’. For accounting change within an organisational change context, see Broadbent and Lauglin (2005).

¹⁸For one of the adoption examples, see Mouritsen et al. (2001), who described a company, which adopted a target costing system for control purposes in an environment where the NPD function was outsourced.

¹⁹In terms of organisational culture, MA literature in the 1970s and 1980s recognised the cultural and knowledge gap between accounting and marketing professions (Bridges, 1971; Ratnatunga et al., 1989). It is a striking fact that writing dealing with the cultural differences at the accounting and engineering interface has not emerged extensively.

²⁰It has to be noted, however, that on a continuum of certainty, basic research and applied research can be claimed to be more uncertain than new product development, which is the focus in this study (Abernethy and Brownell, 1997; Rockness and Shields, 1984; see also Brownell and Dunk, 1991; Brownell and Hirst, 1986; Hirst, 1983).

²¹Either, the investment in MA software cannot be afforded, or the cost–benefit relation of MA information is not satisfactory, or there is no suitable technology to enable MA or its change (Ezzamel et al., 1997; Innes and Mitchell, 1990; Innes et al., 2000; Siti-Nabiha and Scapens, 2005). Even if off-the-shelf software packages may be efficient drivers for MA change (Granlund and Malmi, 2002; Hyvönen et al., 2006), the available technology to meet the distinctive accounting needs in NPD may not exist, at least as standardised software products, or the new centralised MA system can be considered as an external factor typically causing resistance to change (Granlund, 2011; Hyvönen et al., 2006), in which case the technological element is closely tied to the human element of non-change.

²²Based on Eisenhardt and Graebner (2007), theory-building research using cases typically answers research questions that address ‘how’ and ‘why’ in unexplored fields particularly well. They also claimed that with multiple cases, the authors set an appropriate level of abstraction and a multi-organisation study also enable broader exploration of research questions and theoretical elaboration.

²³As regards the empirical data, all of the individuals and case companies appearing in this study are presented anonymously by mutual agreement to ensure both internal and external confidentiality. Key informants reviewed the case reports to assure their validity.

²⁴STPs were prepared for 13 months, the rolled plans verified and frozen for six months ahead, after which the STP figures were revised. The PLP processes were tightly connected with both the long-range planning (3–5 years) based on various roadmaps and the STP process (1–1,5 years), as well as the Demand Supply Balancing process to match supply with sales volume. However, the PLP calculations were even more closely connected to the latest estimate process, which took place at least quarterly and at each milestone of an NPD program.

²⁵The company was involved in NPD activities mainly as it developed its customers' ICT systems and software. In Case C, the company's own service product development and productisation were limited.

²⁶Cf. Chanegrih (2008) and Sulaiman and Mitchell (2005), who found that the elimination of parts of the MA system without replacement did not occur in their responding companies, and who suggested that MA elimination without replacement may be indicative of a reluctance to dispose of underused MA elements and consequently of the potential existence of redundancy within parts of the MA systems of many organisations.

²⁷For example, in NPD, the focus area was typically ‘innovation and renewal’, but during times when the original equipment manufacturers posed a great number of requirements for features and functionalities for game development, ‘customer perspective’ was selected as a highly important area.