ثبات و تغییر در حسابداری مدیریت در طول زمان؛ شواهد از یک قرن یا بیشتر از گینس

In recent years, much has been written on the nature of management accounting change, and indeed stability. Many researchers have used concepts such as rules and routines to interpret this change and/or stability. Recent research has provided an increasingly clear picture of what rules and routines are, as well as contributing to our understanding of the processes of change and stability in management accounting. Management accounting research has mainly presented rules and routines as related phenomena, but some conceptual work has suggested they are separable and can (and possibly should) be considered independently when studying processes of change/stability within management accounting. However, empirical support for such work has been scarce to date. This paper uses data from the archival records of the Guinness company in an effort to establish whether rules and routines, at least in management accounting research, are best considered separable concepts or not. The archival records are artefacts of rules and routines and thus can be used to trace the interactions of rules and routines over time. Support for the notion that rules and routines should be considered separately is presented. The findings also portray the stable, but changing, nature of management accounting routines over time; a point worthy of further research.

Burns and Scapens wrote “whether management accounting has not changed, has changed, or should change have all been discussed” (2000, p. 3). Since their seminal work, much research has been undertaken on refining the meaning of these phenomena, and to a lesser extent on teasing out the interactions of rules and routines as presented by Burns and Scapens (2000) (see for example, Quinn, 2011, Lukka, 2007, Ribeiro and Scapens, 2006, Spraakman, 2006, Hassan, 2005, Siti-Nabiha and Scapens, 2005, Dillard et al., 2004 and Soin et al., 2002). These concepts, rules and routines, were used by Burns and Scapens (2000) to understand the processes of management accounting change (and stability) over time. They presented what is now a largely accepted conceptual framework, which details how the interactions of rules and routines can explain how management accounting remains relatively stable over time, or can change (Burns and Scapens, 2000, p. 10). One of the key tenets of the work of Burns and Scapens (2000) is that there is a process (encoding, enacting, reproduction of rules and routines) by which management accounting may evolve, change, stabilise and re-evolve over time. This process, in a holistic sense, is well accepted, but in recent times some key concepts underlying the process set out by Burns and Scapens (2000) have been explored in more detail. Quinn (2011, p. 338) addressed “some issues of definitional clarity” around the concept of management accounting routines in particular, but by association, rules as set out by Burns and Scapens (2000). Briefly here, Quinn (2011) draws on the work of Feldman and Pentland (2003) to view management accounting routines as having two dimensions; namely, ostensive and performative. Quinn (2011, p. 344) also proposes rules are formal and written or represented in some physical way, and are in fact artefacts of routines. More detail on rules, routines and artefacts is given later, but a proposal by Quinn (2011) is that ontological clarity on the nature of rules and routines will assist management accounting researchers to gain a more in-depth understanding of how rules and routines interact. This understanding of the interactions of rules and routines is potentially important as Quinn (2011) suggests management accounting routines, or rules, may be more prevalent in certain types of organisation. However, the work of Quinn (2011) could be criticised for being conceptual – no empirical data is given to support the assertions on the interactions of management accounting rules and routines. Briefly, a key differentiation between Burns and Scapens (2000) and Quinn (2011) is that the former presented rules and routines as bound together in the process of management accounting change, whereas Quinn (2011) portrays them as separable and distinct concepts, and proposes that rules need not exist. This study seeks some empirical grounding (or otherwise) for these two key assertions of Quinn (2011). This study uses archival records – namely the archives of the St. James's Gate Brewery of the Guinness company – to study management accounting rules and routines over an extended timeframe and explore Quinn's (2011) propositions. While more detail on the methods used is given in the next section, it was envisaged at the outset of this research that archival records would be suited for two reasons: (1) more records were formally written in the past, thus increasing the likelihood of written rules being present and (2) the extended research timeframe achievable through archival research provides greater scope for studying the interactions of management accounting rules and routines over time and how these interactions promote stability or bring about change. The remainder of the paper is structured as follows. Section 2, briefly reviews some of the organisational, institutional economics and management accounting literature to date, focusing mainly on the concepts of, and interactions of, rules and routines. Then, Section 3 sets out the methods used to obtain and analyse the archival data, as well as briefly exploring the potential benefits of studying change over longer timeframes. Section 4 introduces the archival data from Guinness and outlines a story of the relative stability of management accounting in one area of the company over an extended timeframe. It also provides several examples of management accounting change, the factors which were potentially driving change, and describes the interactions of rules and routines over time for each example given. Section 5 completes the paper, with some concluding remarks, limitations of the research and suggestions for future research.

The previous section described examples of change and stability in management accounting at the Guinness Cooperage, as conveyed through the interactions of rules and routines. At the outset, the aims of this study were to provide some empirical grounding for the key proposals of Quinn (2011) – namely that rules and routines can be unbundled and that rules need not exist. The unbundling is in particular contrast to Burns and Scapens (2000), who treat rules and routines as a combined unit of analysis in their framework – indeed they note their positioning of rules and routines in their framework as somewhat arbitrary (2000, p. 10). If we reflect on the management accounting rules and routines of the Guinness Cooperage as shown in Table 3, we can see that in most cases written rules (as defined by Quinn, 2011) were formed to bring about change in the existing management accounting practices at the Cooperage – namely examples 1, 2A, 3 and 4. We can also see from the examples presented in Table 3 that sources of change to rules and/or routines can be both internal and external to the organisation (Burns and Scapens, 2000). Reflecting on the research objectives set out earlier, let us first consider Quinn's (2011) proposal that rules need not exist. In Table 3, example 2B (quarterly Cooperage accounts) supports this element of Quinn (2011) – that is, that rules are not necessarily always part of the process of management accounting change as originally set out by Burns and Scapens (2000) (see also Fig. 1). In example 2B, a substantial change to management accounting occurred in that the Cooperage assumed responsibility for the preparation of departmental accounts. As noted, this responsibility “passed” from the Audit Office, but despite many volumes of detailed Cooperage records spanning from 1867, there is no evidence that the Audit Office actually prepared departmental accounts for the Cooperage in years prior to 1909. Nor was any written evidence of the expected or required content of the Cooperage departmental accounts found i.e. there was no rule as at 1909 for the Cooperage department to follow. Thus, without a rule, staff at the Cooperage used their ostensive understanding to prepare departmental accounts according to their needs and the needs of the Board. As this task was repeated many times, we can term it a routine (see also Section 2 and Table 3) and the ostensive routine guided the performative routine. Thus, as noted by Quinn (2011, pp. 345–346), rules are not necessarily part of the process of management accounting change (as in example 2B) and when rules are not present, the ostensive routine is drawn upon to guide actors ( Pentland and Feldman, 2005, Pentland and Feldman, 2008 and Feldman and Pentland, 2003). However, all other examples set out in Table 3 contradict Quinn's (2011) notion that rules need not exist. Indeed, example 1, 3 and 4 not only seem to portray change based on rules, but new rules. Thus, based on the evidence presented here, and the case on which Quinn's (2011) work is based (see Quinn, 2010), it would seem that rules or routines may be respectively more important in the processes of change in different organisations, and potentially should be considered separately by researchers. This leads on to the second research objective, which is to explore whether or not rules and routines can be unbundled as suggested by Quinn (2011), i.e. that the concepts are best considered ontologically and empirically separate. Section 2 has, in effect, made this assumption in the methods used to determine how rules and routines can be interpreted from artefacts. The data conveyed in Table 3 suggests there is more than one ‘starting point’ in the interactions of rules and routines which eventually bring about management accounting practices. That is, rules and routines are separable and distinct concepts, and at a point in time are ontologically distinct. While this seems to contradict Burns and Scapens (2000) and support Quinn (2011), it is possible that over time, rules and routines will become similar and indistinguishable, and for empirical research purposes can be bundled as envisioned by Burns and Scapens (2000). However, unbundling rules and routines may carry some import for empirical management accounting research. Due to the archival nature of the present research, it is not possible to determine with absolute certainty that the starting points of rules/routines shown in Table 3 are correct. Despite this, it is proposed that understanding the starting point in rule and routine interactions will assist the interpretation of processes of management accounting change (and stability) as originally set out by Burns and Scapens (2000). For example, the simplification of the Cooperage accounts in advance of the opening of the London brewery began with a new rule. This rule carried with it the power of the Board and was as such more likely to be implemented and eventually become a routine. Similarly, if a routine has been enacted and stable for quite some time, it is more likely to remain stable over time as it becomes taken-for-granted – as in the case of the controlling of cask/kegs – and eventually encompassed by rules. An obvious question for future research is what are the factors which cause rules or routines to be relatively more important in organisations, and thus in their processes of change. For example, the Guinness company may be historically considered a more formalised organisation, at least in an Irish context, thus implying the organisation functioned along clear lines of authority and power structures and would be more likely to be rules-based. In addition to the examples of change to management accounting at the Cooperage cited in Section 4, the story of stability of management accounting practices around returnable casks/kegs is worthy of further mention. The management accounting associated with controlling casks/kegs – recording movements and inventory – has not in essence changed from the 1870s to the present day. It could be argued that this routine has changed in that, for example, (1) it is now computerised, not in handwritten ledgers and (2) the level of detail has been reduced from tracing each cask by customer to tracing in a pooled fashion (see Section 4.3). It could also be argued that this routine has not changed and been quite stable over time. Using the four defining characteristics of a routine as outlined by Pentland (2011) (see Section 2), the routine meets these characteristics whether we consider it to have changed or remained stable over time. It can be argued that the patterns of action underlying the routine (see also Pentland et al., 2010) have changed over time, but to be a routine, only a general pattern is required. Thus, here it is proposed that management accounting routines around controlling cask/kegs have not changed in a century or more, as the general pattern described by the Sales and Logistics staff today is similar to those in the past. To relate this argument to the work of both Quinn (2011) and Burns and Scapens (2000), their work does not address the degree to which rules and routines may be become institutionalised i.e. how taken-for-granted they are. Arguably, without the extended timeframe permitted by the archival analysis here, this steadfastness of routines may have gone unnoticed. Based on the evidence at Guinness, it would appear that some management accounting routines within the Cooperage are more ‘sticky’ over time, or potentially closer to the institutional realm as denoted by Quinn (2011) and Burns and Scapens (2000). Although the actions undertaken by actors (performative routine) for the controlling/recording of casks/kegs has varied over time i.e. moved along the action realm of Quinn (2011)/Burns & Scapens’ (2000) framework over time, arguably the ostensive routine remained in a ‘time warp’ along the institutional realm i.e. it did not change over time. To put this another way, there is both stability and change in the same routines over time. It could also be argued that the ostensive routines for controlling cask/kegs at Guinness (i.e. an understanding of what should be done) have not changed, but the performative routines have (i.e. how we do it). Recent work by Vromen (2011) suggests routines can be construed as multi-level mechanisms. Briefly here, this strand of research suggests routines can be construed as both a whole and constituent parts (Vromen, 2011, p. 183). Thus, for example the ‘whole’ controlling of casks/kegs has remained stable over time, but the ‘parts’ of the routine (i.e. the various patterns of action) may have changed over time (see also Pentland et al., 2010 and Pentland et al., 2012). The issue of management accounting routines in particular being both stable and changing over time is one which is not fully addressed here, or by Burns and Scapens (2000)25 or Quinn (2011), and is worthy of further research. There are a number of limitations to the research presented here. First, the work is based on the records of a single organisation and thus findings here may not be generalisable. In particular, more studies of the interactions of management accounting rules and routines are needed to support or challenge the findings presented here. In particular, research in understanding why rules or routines in management accounting take on more importance in organisations would be welcome. Recent literature on routines in particular as mentioned here will aid researchers identify routines more easily. Rules, on the other hand, have received less attention in the literature. Second, as noted by Johansson and Siverbo “institutions and routines cannot be observed, only the artefacts” (2009, p. 150). While every effort has been made here to extract the rules and routines from the artefacts (see Section 3), this research is not a definitive substitute for actual observations of behaviour or questioning organisational members. The interpretations of the artefacts presented here cannot be absolutely certain without input from actors who performed the routines. Third, although the archival records have been excellently catalogued, well maintained and meticulously examined, it is not possible to be absolutely certain that all elements in the processes of change depicted here have been recorded over time. Fourth, it could be argued that archival research does not sit well with researching phenomena such as rules and routines as they are better studied through observation and direct enquiry. However, as noted by D’Adderio (2011), artefacts are a central component of routines and are often utilised in the study of routines (see also, Pentland and Feldman, 2008). Finally, although the data from the Guinness Cooperage provides a rich story of management accounting change over an extended time period, the period adopted is reflective of a time when many records were hand-written or typed. While it is interesting that – see example 2B, Table 3 – formal (and written) rules were not always present at the Cooperage, in today's business environment many management accounting tasks are captured, processed and stored by information technology as opposed to written or typed by hand. This research does not address the impacts of technology on rules and routines in any way. For example, Volkoff et al. (2007, p. 839) propose the concept of material routines which are “organisational routines embedded in the ES [Enterprise System] in the form of system-executed transactions – sets of explicitly defined steps that require specific data inputs to automatically generate specific outcomes”. Following the logic of Quinn (2011), these material routines would in fact be rules (see also, Oliveira and Quinn, 2012). More research in a contemporary environment may be fruitful in teasing out how technology encapsulates and/or interacts with management accounting rules and routines, and ultimately affects the processes of change/stability in management accounting practices as set out by Burns and Scapens (2000) and Quinn (2011). To sum up, this paper has provided some evidence in support of the work of Quinn (2011), but also contradicts some of his findings. It has been shown that rules are not necessarily a component of management accounting change/stability; whereas routines appear pervasive in this same process. However, rules also proved to be the starting point of what eventually became routines at Guinness. Thus, rules and routines are separable concepts, and both separately may help us interpret management accounting change. By unbundling rules and routines, and considering the latter in more micro-components as set out by Feldman and Pentland (2003) (see also, van der Steen, 2011), it may be easier for researchers provide a richer and more detailed understanding of the processes of management accounting change than originally envisaged by Burns and Scapens (2000). Understanding too, through future research, why rules and/or routines may be relatively more important in the processes of management accounting change may further underpin the overall picture of change painted by Burns and Scapens (2000). However, one potentially interesting factor is not present in the current research, nor in Quinn (2011) or Burns and Scapens (2000) – namely, information technology. Research on the interactions of rules and routines within and without technology is likely to prove very fruitful and add to the present and extant research on understanding management accounting change and stability which adopts concepts such as rules and routines.