There is a mutual influence between accounting and organisation. Horngren et al. have shown that management accounting is an instrument to influence the behavior of managers in a way that it enhances the attainment of organisational objectives (Horngren, Bhimani, Foster and Datar, 2005). So, management accounting apparently has to be customised, the need for which has been analysed in the contexts of departmental interdependence (Gerdin, 2005), supply chains (Knight and Harland 2005) and the Internet (Sheehan 2005).
Scarce sources are available, however, for accountants who look to develop new skills and techniques when their companies partner a business network (Hakansson and Lind, 2004). This paper considers from where accountants may derive guidance for this work, and how accountants’ work can support the objectives of a network. ACCOUNTING CUSTOMISATION The needs to customise are manifold, and consisting of two directions, as follows: 1.
Traditional reporting and budgeting need to be adapted to reflect varying input and output; standards of performance need to be created which measure and verify the contribution to common business processes; and, adequate process standards need to be elaborated. Further, the accountants will need to devise new costing and billing procedures that satisfy the requirements of diverse network partners, and they will also need to establish profit-sharing rules and appropriate investment incentives. 2. The methods and instruments of management accounting in a usiness network need to comply with the conditions of an environment, where respective partners have diverging business objectives, mutual trust is needed (but distrust will possibly prevail at the outset), and bargaining power will be exploited over all sorts of issues. Co-ordination will be of the utmost importance, as will openness, a willingness to share risks, and rewards. The objective is to gain competitive advantage, resulting in better business performance than would be achieved by the firms individually.
What has been enumerated here comes very close to the definition of a business network , defining elements for which include: combining diverse business objectives, gaining competitive advantage that would not be achieved individually, mutual trust and co-ordination, restricted membership; specific business objectives, and sharing risks and rewards (Rosenfeld, 1995). Neither of the two directions (above) are clearly visible in the literature, an example of where the gap between accountancy practitioners and academics is painstakingly felt (Pearce, 2004).
One reason for this may be the way that management accounting knowledge is delivered – i. e. , too “regulated”, too casuistic, and/or overly focused on reporting requirements (Spender, 2005). A different reason may be the way that research is conducted – i. e. , too case-specific, too much oriented towards output factors such as ‘value’ and ‘performance’, and less focussed on input-like methods and techniques. When input in management accounting relates to data and instruments, then it suggests that a look at traditional instruments will help to find the appropriate direction.
Traditional management accounting, it is argued, has much to offer in terms of developing methods of decision-support and of management control in business networks. This paper explores four avenues to demonstrate such connections. First, the following considers appropriate use of cost budgets, arguing that the Japanese concept of cost design (i. e. , ‘genka kikaku’ – imperfectly translated as ‘target costing’) was developed for collective and co-operative decision-making and control in a horizontal organisation.
And, furthermore, that this involves many facets of bargaining – the only, if important, difference to (open) business networks being that we mostly have diverging business objectives. The second avenue is related to collaborative planning, forecasting and replenishment. More specifically, this would be a proposition that appropriate results can only be achieved if adequate management tools were in place to verify and validate the relevant data and their sources.
Third, we connect the set-up of ‘common denominators’ in multi-stage performance-monitoring to the conception of distributed decision making – decision makers’ diverse preferences may result in more than one set of cost-value assignments; investment-oriented depreciation may differ depending on the purposes of investment, etc. In such circumstances, a network’s management accountant needs to find a denomination that is agreeable to all parties involved.
Likewise, s/he will have to find ‘common denominators’ to define the ROI (rather: ROIs) of the specified investments that are imperative to keep a network alive. Fourth, connection can be derived through introducing the notion of transaction cost to allocation and pricing. Transaction costs rise substantially when bidding and auctioning resources become the routine manner of sharing the means required by the network, and they must be accounted for. Accounting for transaction costs is one issue, but the other is for accountants to support management in attempting to reduce transaction costs.
Wherever management systems are deployed – in small firms, in large enterprises, in standalone businesses or in networks – they facilitate co-ordination, collaboration and control. Management control assures that resources are obtained and used effectively and efficiently in the accomplishment of the organisation’s objectives (Anthony and Govindarajan, 1998). When resources are contributed by different businesses and when these businesses’ objectives differ, the systems will need to be more sophisticated.
More co-ordination is required, and more costs of co-ordination arise. This, in turn, must be offset by increased effectiveness and efficiency to reach the desired performance level. One aspect of this has been dealt with by researchers and practitioners through assessing the performance of networks (however performance is defined) but, if measured at all, network performance has to date been captured at the organisational rather than the network level (Sydow and Milward, 2003).
Another aspect is that of strategies and processes (including management accounting processes) that contribute to the overall success of a network. But, again, relevant research is still in its infancy (Barringer and Harrison 2000). A third approach would be to establish what defines co-ordination in the context of management and explore which instruments enhance co-ordination. Referring to the works of Fayol (1919/49), co-ordination is encompassed in management since to manage is to forecast and plan, organise, command, co-or-dinate and control.
And, to co-ordinate is to procure an interaction among two or more individuals – including communication, information sharing, cooperation, problem solving, and negotiation (Heckert and Willson, 1963). This applies to corporate organisations as well as to business networks. In networks, four dimensions of co-ordination may be considered (Schneeweiss, 2003a), namely: 1. ‘Data integration’ describes the lowest level of integration; this requires consistency of data and, generally, of structures to secure a smooth interchange between the partners of the network. . Integrating through ‘negotiations’ in a market-like setting where the negotiations may either be reactive or they may be coordinated through a coordinating agent who has the (hierarchically) superior power to set negotiation rules and to facilitate the communication process. 3. Integration through ‘planning activities’ involving (common or mutual) anticipations and instructions where instructions will influence the decision-making through transactional changes (and thus leave the general preference attitude of the decision-makers unchanged); 4.
Integrative ‘leadership’ which aims at influencing the general decision behaviour of the parties involved by transformational changes, i. e. by achieving long-lasting shifts of the decision makers’ preference structure. The concern is towards communication and persuasive efforts. In this paper, the transformational aspects will not be explored. This leaves three dimensions: planning, negotiation and data integration. From another perspective, these three ingredients are commensurate to the essence of continuous collaboration.
That is, continuity is secured by sharing data and budgets, and by ongoing negotiation. As in any business combination, legal contracts alone do not make alliances work (Anderson and Jap, 2005), and more can be expected from ‘learning alliances’ (Khanna et al. , 1998). ‘Controlled competition’ has become a common concept in business (Gulati et al. , 2000). But, even though it is generally recognised that majority ownership does not necessarily equate to effective control, the common view of control must still shift from what is traditionally seen as ‘directive power’ in large organisations (Fligstein, 1990).
Likewise, decision support in business networks must shift from what it is like in a monolithic corporation. Decision support to be provided by the management accountant in business networks must foster multi-level optimisation, multi-level stochastic programming, collaborative production- and replenishment-planning, multiple-source-contracting and similar conjunctures of interrelated decision-making. From the point of decision theory, this type of advanced management systems can be viewed as distributed decision-making, with specifications ranging from simplistic one-person settings to antagonistic multi-person settings.
In systems of a one-person setting we will see distributed decision-making if decisions have to be made, and new information is required, over time. Decisions made at different points encapsulate asymmetric information, as do decisions made under different platforms of power. Asymmetric information may also be deemed as ‘hierarchical’. In respect of information, a hierarchical dependence is created when a relationship is non-symmetric because one party is superior to the other (Schneeweiss, 2003).
However, in multi-person settings, diverging information means additional things, namely: (1) decision rights with different grades of risk avoidance; and (2) different forms of communication between decision-making units. Hence, distributed decision-making shapes the design and co-ordination of management connectivity, and thus impacts management accounting. Asymmetric information and multi-person settings are intrinsic in bargaining, and bargaining (or trust) is a key characteristic for networks. Trust development goes through several stages (Lewicki et al. 1998) and the various grades of trust will shape the state of a network’s development – from start-up to wide market penetration, from information symmetry to antagonistic motivation . TRUST AND BARGAINING As management in a business network deals with conditions of asymmetric information (described above), asymmetric information will become a more extensive problem for management accounting, notwithstanding that problems of this kind can also be identified with single organisations. The problem can be exemplified by considering intra-company-transfers of intermediate products.
The manager of the supplying division frequently has access to private information about the cost of producing the intermediate good, whereas the manager of the buying division generally has better knowledge regarding the net revenues from selling the final product. The instrument often deployed by the accountant in this scenario – i. e. , ‘negotiated transfer pricing’, permits the divisional managers to bargain on prices and quantities, incorporating their local information into the respective decisions (Kaplan and Atkinson, 1998).
However, the outcome of this strategy may well be sub-optimal (and non-profitable), as the managers may strive for more favorable prices by overstating cost and understating revenues. Even worse, these over- and under-statings may wrongly promote capital investment. For instance, the selling division may acquire equipment up-front in order to reduce its variable production cost, or the buying division may invest in marketing activities based on false information.
Mis-investments of this type can hopefully be prevented if the management accountant of the corporation checks the investment decision on the basis of, say, standard cost. But if we have a similar relation beyond the boundaries of an organisation – i. e. , between business network partners – standard cost information or other check-and-balance provisions will not be at hand. To solve these types of problems, economic theory provides mathematical solutions, which have been developed in the context of bilateral trading (e. g. Baldenius, 2000). However, there is another way to approach this, namely through trust and openness.
H. B. Thorelli, one of the first business scholars to write on networks, defined trust as “an assumption or reliance on the part of A that, if either A or B encounters a problem in the fulfillment of his implicit or explicit transactional obligations, B may be counted on to do what A would do if B’s resources were at A’s disposal” (Thorelli, 1986, p. 24). In terms of management accounting this will allow us to dispense with the need to specify unforeseeable consequences, for it is assumed that the decision rule to be followed by A will be identical to the decision rule to be followed by B.
When there is trust, the need of pre-specifying every possible future outcome and of setting up mechanisms to prevent or correct opportunistic behavior is greatly diminished. Can trust be generated? According to theoretical contributions, two entrepreneurs wishing to invest their resources have to act on two variables (Jarillo, 1993): (1) assumptions of the owner of the resource (the other party) regarding the other entrepreneur’s motivation, and (2) the intrinsic situation as such. 1.
The first problem can be addressed by carefully choosing the partners to the different relationships, searching for people to whom the entrepreneur can relate – i. e. , those sharing similar values. Thus, identity of values and motivation will facilitate the emergence of trust. 2. With regard to the second variable, the entrepre-neur cannot expect ‘blind trust’ if this means that the members of the network must put themselves at high risk. Thus, trusting behavior can only be generated by showing that the entrepreneur would be worse off if s/he behaved opportunistically.
Management accounting has been linked to the institutionalisation of trust within an organisation (Johansson and Baldvinsdottir, 2003) as it reduces risk by providing accurate data. In this context, the phenomena of both risk and trust can be addressed by using the same models. One model, game theory, has been applied to identify the conditions under which cooperative behavior (which is equivalent to trust) could emerge when there is no central enforcement as we would (or should) find within a corporate organisation (Axelrod, 1984).
Game theory, according to its oft-cited ‘Prisoner’s Dilemma’ concept, tells us that when two players have a choice between cooperation and defection they will achieve mutual gains by co-operating, but one of them can also exploit the other by defecting. It is held that, in the long run, ‘optimum’ output is achieved by the two players if each uses the strategy to do what the other player did on the previous move. This strategy, called ‘tit-for-tat’, may look simple, but it is a useful formalisation of the basic problems associated with lack of trust.
That is, in spite of the obvious benefits of cooperation, there is always strong temptation to default and take larger benefits. An example would be joint R;D ventures. If several firms decide to take on a given project, they may think it is to their advantage to send a second-class scientist to the project, while expecting that the others will send their best. No firm wants to send its best people, fearing that the others will send their seconds and get a ‘free ride’. So, everybody sends their second-bests, and the project fails (Axelrod, 1984).
This joint R;D illustration can be used to demonstrate management accounting’s contribution towards avoiding project failure. That is, if partners institutionalise project evaluation (methods, data, accuracy and outcome instrumentality), trust is created on the foundation of hard facts. In project management, and in the context of project management capability generally decreasing, trust accumulation has been attempted by applying psychosocial project performance indicators (Tukel and Rom, 2001), but it has been shown that applying methods that relate to quantitative evaluation achieve superior results (Bryde 2005).
When business partners wish to share cost and resources (and profits and cashflows), management accounting will have four tracks by which to attempt to connect the businesses, namely: (1) target costing, (2) collaborative planning, (3) common denominators to determine profits and returns, and (4) transaction cost. All such tracks have normally been developed and deployed for use in structures that comprise central control. Each will now be considered in more detail. Target costing (‘genka kikaku’)
The nature of ‘target costing’ can be elucidated on the basis of the Japanese denomination, ‘genka kikaku’ (Kato, 1993). Genka kikaku may be aligned with ‘genka kaizen’ and ‘genka keisan’, whereby ‘genka’ in all three terms is equivalent to ‘consensually design, plan and report’. ‘Kikaku’ refers to project(s), ‘kaizen’ refers to (production)-improvement, and ‘keisan’ to cost. From the outset, target costing is much more than a simple procedure for arriving at allowable cost on the basis of market price minus planned profit (i. e. market-into-company’). The main elements are: common and consensual decisions concerning how to reach and control allowable cost; collaborative action concerning how to close the gap between these allowable costs; and the standard costs that had been established before the action (‘drifting cost’). The background here is the Japanese philosophy of ‘pull-method’ of production where a subsequent stage of the production process shapes the previous stage by communicating down the line precisely what objects are required at any given moment of need.
In Western organisations, by contrast, we might be more familiar with the ‘push method’, whereby the previous-stage inflexibly supplies the following one with its processed goods. This amounts to a collective two-way-management that is not comparable to feedback in the push method. Therefore, target costing is closely related to ‘just-in-time’ (JIT), and should better be translated as ‘cost-design’. The interaction between cost accountants and engi-neers, the relationship between feedback and feed forward, and the process of cost reduction at design stage are all common features of both JIT and target costing.
The objective, then, is to discover and remedy problems quickly. Another dimension to ‘market-into-company’ is the close link between allowable cost and customer expectations. In the process of de-composing allowable cost down to its parts, functions and components of a product or service, the contribution to customer value is determined (i. e. , ‘value control card’). Then, a consensus must be found on how much cost is allowed (i. e. , how much cost reduction has to be achieved) for which part, function or component.
The less customer value produced by one constituent, the lower its cost must be. An illustration of value creation in networks would be Japanese car making. All stages and steps of target costing are distributed over a quantity of partners in the supply chain of, e. g. , Toyota. The same would apply to the business network of Benetton’s or the partners of a construction business network (Nicolini et al. , 2000). The partners partake in target costing such that prices, volumes, works systems, the levels of quality etc. are controlled by what might be termed ‘participative budgeting’.
Many Western networks are also moving away from mere arms-length contracting into a stable relationship where long term objectives are collectively set (Jarillo, 1993). This is where management accounting can contribute adequate procedures, data and performance indicators. From another perspective, collective target setting in a complex supply chain reflects the complexity of distributed decision making problems (DDM). Let us consider the design of a medium-term contract between a supplier and a producer. If the contract offers a bonus for a JIT delivery, his will induce the supplier to ensure correct delivery. Simultaneously, however, to enable the supplier to deliver correctly, in most cases capacities must be adjusted. Thus, planning for the operational level has a direct consequence on the medium-term level of capacity adaptation. Altogether, there are three interwoven DDM problems (Schneeweiss, 2003), namely: (1) between the tactical contract level and the operational level; (2) between medium-term and short-term decisions; and, (3) between the producer and supplier.
Participative budgeting, in essence, is about coordinating the allocation of management effort to interrelated productive activities. Obviously, planning cannot prescribe detailed courses of action to the parties involved because they need to maintain the power of decision. Planning here deals with rough, aggregate results of managerial action and determines budgets that define the performance goals to be met. In a network, what has to be defined are aggregate output requirements and input restrictions for the various domains of individual responsibility.
Solutions for this type of decision are offered by mathematical programming. In management accounting practice, mathematical programming often plays only a minor role, while, on the other hand, accountants experience all sorts of difficulties with activity-based planning. But mathematical programming is not only based on an activity analytic model of the operations to be planned, both approaches are closely akin. Practice, in merely dealing with activity-based planning, tends to consider the results of only a few (comparable) alternatives.
On the contrary, an optimisation model would consider a large spectrum of alternatives, represented by inequalities, and trying to select the alternative that scores highest on a prescribed performance measure (Luhmer, 2000). It is not the objects of budgeting in a network that differ from budgeting in a centrally controlled environment; rather, it is their characteristics. Here and there we have products, services, processes, transactions and their cost; here and there we have structural elements that cause costs of steering, investment-oriented costs, costs of coordination and life cycle costs.
However, when boundaries vary, and when membership, preferences and shareholder interest vary, the composition of products, services, processes and transactions will vary, as will their life cycles. An appropriate comparison is that of a railway network, whereby nodes of that network (i. e. , its tracks, rolling stock, switches and depots) are scattered across a geographic area. The configuration of these nodes, in turn, determine where, when, to whom, how much and how quickly goods can be transported (Gottinger, 2003).
Furthermore, it is these ‘how’s’ that convey a constantly varying composition. Hence a proper de-composition of costs, prices and capital expenditures must be provided in order to assign each component of cost, price and capital expenditure to the network member who contributes the component and its physical counterpart. Management accounting’s foremost task is that of establishing this structure of components – a structure that must be transparent, flexible and responsive to the information members require.
The notions of ‘direct’ and ‘indirect’, ‘standard’ and ‘accrued’ will shift in meaning with network-accounting: Indirect cost allocated to a component contributed by one partner of a network will be converted into direct cost (‘partner specific cost’) when this component enters a new stage of processing with in the network. Compare this to transfer pricing within corporations. The cost elements of the price in their origin may have been ‘direct’ and ‘indirect’, to the recipient of the transferred good or service – the transfer price is variable cost at all times.
But, partner-specific cost must also be determined for the purpose of determining a partner’s profit share. The notion of ‘specificity’ is innate – i. e. , a concept that specific investment (and specific cost) induces closer partner-relations. This may be an important feature of supply networks. In respect of standard costing, in a stand-alone enterprise, this tool aims that objects that are standardised will become streamlined and will eventually help to reduce costs.
But, in a network, cost reduction requires collective efforts and will mostly be achieved by streamlining not the goods or services that are produced but by optimising the processes which produce them. Standards will be set for parameters, indicators of performance, input levels and output levels, and the reference point will be value-added. So, standards and cost variance analysis still have a role to play. But it is more important to interweave cost or profit variance in the planned value beforehand and to take measures to get rid of variance before actual performance.
With target costing, and especially with target costing in networks, management accounting has developed from a feedback to a feed-forward control system. Target costing is cost design (genka kikaku), and is a feed-forward perspective, a collaborative perspective, the same as cost reduction which requires planning and collective efforts. Networking comprehends the new paradigms of multi-skilling, collective responsibility, cooperation, two-way information, horizontal organisation and proactive management – areas, it is argued, to which management accounting ought to contribute solutions.
It may not be wrong to suggest that Japanese management accounting has found these solutions by establishing a set of new paradigms that are at least sufficient in supply chain networks (Nishimura, 2003). Another important focus of network accounting centers on intra- and inter-organisational activities and processes. Activity-based-costing (ABC) is linked to target costing by a common feature – i. e. , market perspective. One aspect relates to entangling product strategy and feed forward control, while the other relates to allocating cost from the perspective of business strategy.
Activities consume strategic resources which network partners contribute, and the purpose is to deliver products or services based on activities with high value added. So, ABC (in a network) connects resource allocation to strategic pricing policy, whereby focus is on resources which convey activities that foster products or services with high value added. The network partners’ interest is in bringing about the highest possible value added. One partner may just wish to bring capacities to use in a network which otherwise might be idle and where the activities attached to these capacities are not state-of-the-art.
Other partners will have to carefully watch over the business processes to be re-engineered if they do not meet the targets for value-added. This monitoring is undertaken through the network’s management accounting. Activity-based management is challenged at its highest when activities are intertwined amongst partners who are free to leave a network almost any time they want. Non-permanent membership to a network is inborn in any relationship, and more so if it is based on access through the Internet.
One has the freedom to enter and exit a virtual group without having to be responsible or accountable in most circumstances. This is a genetic handicap compared to physical organisations which have evolved an established set of laws, rules and contracts. Still, some crucial standards, authority, reference and creditability to command respect and order must be created and accepted in virtual alliances. Day-to-day operation is only possible by continuously linking activities into processes which produce the output sought after by the network partners.
Even if subject to change, relationships, bonds and obligations between its members must be defined, as well as (joint) ownership of its products and processes. One solution is to have modular processes which can be detached and re-attached when a partner leaves or enters the network. This format is very often chosen by R;D networks. Collaborative Planning, Forecasting, and Replenishment In the preceding section, orientation to value added processes was outlined to be vital for a network’s optimisation. It was also shown how this is connected to strategic planning and customer focus.
An even stronger emphasis on customer focus and, especially, consumer response, is inducted by the concept of Collaborative Planning, Forecasting, and Replenishment (CPFR). CPFR is a registered trademark of the Voluntary Inter-Industry Commerce Standards (VICS) Association. The main emphasis is on making retail and manufacturing work together in orienting the supply chain towards customer needs and, thus, two main parts emerge: (1) supply chain management (SCM), and (2) category management (CM), concentrated on store assortment, product promotion and product introduction.
The characteristic is comprehensive solutions and optimum exchange of data involving all members of a supply chain partnership, eliminating all inefficiencies that occur through uncoordinated sequences and idle capacity. In CPFR, we re-encounter the notion of the ‘push-principle’. Production and distribution in the supply relationship are synchronised not on the basis of what is ‘pushed into the pipeline’ but on the basis of information from retail outlets. We may generalise as follows. In every value chain – producers to retailers, joint R;D, consultants’ networks, knowledge networks etc. there is a sequence of relationships between customer/consumer and supplier of activities. These inter-lockings are domains of management accounting. The accountant of a network and/or the accountants of network partners will (jointly) determine the parameters, the performance indicators, and input and output levels; also, how things will be measured, how often, and by whom. And, if deviations occur, all network members will need to intervene. Many highly-sophisticated tools have been developed for SCM systems, and accounting has had to set up appropriate instruments that provide the data required by these systems.
But, many SCM system applications are designed for a configuration in which one dominant partner imposes. In its beginning, ‘category management’ originated in the supply chains of the U. S. textile and garment industry, and addressed the requirements of a multi-partner setting. Groundwork for SCM was laid down by Kurt Salmon in the textile industry (Seifert 2002); the most divulged initiatives on collaborative planning, replenishment and forecasting started at large-scale units like Wal-Mart and Warner-Lambert.
And, even though ‘win-win’ situations had been advocated, sometimes it might have appeared more like “we win – you figure out how to win” (Ireland and Bruce, 2000, p. 90). When all network partners have equal power, there should be a ‘win-win-…-win’ situation for all. However, all partners will need to equally endow resources, and engage in joint efforts to establish the network’s strategy and business plans, and to shape the configuration of value chain(s).
Partnerships of equally distributed rights and duties have been fostered and promoted, much more than large-scale logistical endeavors, by the opportunities afforded by the Internet and the B2B marketplace. The latter enable partnerships to better control the entire process. Control that had been centralised is now shifted towards distributed control, and management accounting is also practiced beyond the boundaries of just one organisation. But, this is not necessarily new – it has happened before, when order-processing between buyer and seller in a purchase relationship crossed organisational borders.
So, in this sense, management accounting needs now only to adapt and enlarge such ideas. Here, we see an ‘avenue’ for management accountants to move into – i. e. , from a field in which s/he is competent within a centrally controlled organisation, to a broader and more complex field, with wider range of applications. Cost structures of the various value adding activities/processes/operations need to be analysed in detail, constraints and individual restrictions need to be accounted for, and inputs contributed by the network’s members need to be ascertained.
Here, there is a sense by which this means expansion of, in this instance, Porter’s ‘value chain analysis’ (Porter, 1985) beyond the boundaries of a single organisation. It is argued that Porter’s concept of competitive advantage does not fit network-collaboration because network partners will always contest in some way (Jarillo, 1993). But it is only one step from analysing business unit interrelationships to inter-organisational relationships (VCAP, 2005). The objects of study are identical in intra- or inter-organisational relationships – activities, cost behaviour, and cost-differentiation.
With activities crossing boundaries, competencies and resources need to be been determined and measured jointly by the network partners, and collaboration points must be established. Management accounting must thus map these points to the network partner’s assets and costs, and must assist in establishing the need for information that is to be shared at the collaboration points. Management accounting will also have to participate when IT defines the means and formats of data exchange. A further set of tasks relates to how orders for goods and services will be committed, processed and delivered.
Many networks employ a ‘broker’ whose job is to create an order forecast, and to commit the partners’ resources for each incoming order. This broker will also need accounting information to make decisions on how to manage regular orders, and how to handle exceptions – i. e. , job cost, input prices, capacity utilisation, processing time, etc. A large quantity of such information will be needed during the early phase when the partners’ collaborative processes are being integrated for the first time and interfaces are being established.
Modern IT technology such as SAP’s Enterprise Application Integration (EAI) will assist in connecting network partners’ IT systems, but there is always a need to homogenise the contents, and this must be done by the management accountants involved (SAP, 2004). When a network emerges, there will already be a need for change, and during its lifetime, there will always be further need for change. Even if in a first phase the basic processes contributed by the network partners can be interfaced without much redefinition, they might soon need to be reinforced o make them permanent, and additional processes will have to be adapted when they enter the network. The new ‘to be’ processes will have to be developed from the current ‘as is’ processes, based upon requirements derived from performance metrics and measurements – again, an accountant’s domain. Common denominators to ascertain ROIs and profit shares When resources partake in a network that pertain to different network members, the decisions that have assigned those resources will mostly have been made on the basis of the members’ individual capital budgeting and accounting systems.
There will probably be disparity amongst, for example, decision-makers’ preferences and aspirations, accountants’ methods and ability to budget cost and benefits, sensitivity analyses, and check-and-balance schemes. There will be differences in principle, in depth, as well as industry-specific differences. Many networks combine activities that originate from companies with very diverse backgrounds. Take an e-bookstore where storing and replenishment, physical handling and the settling of payments are executed by network partners who might not only determine their ROIs in very different ways but whose aspirations of ROIs are also disparate.
Another example is a utility network where producers, transmitters, broker-dealers and distributors of energy put together a portfolio of manifold resources. In a system that was ‘unbundled’ from a conglomerate into separate operators and ‘re-bundled’ through intermediaries and through contracting and/or acting on specific product and financial markets, investment decisions and investment control will lie with each individual operator or intermediary (Fernando and Kleindorfer, 1997).
On the other hand, all must agree on data reliability. Long, medium and short-term functions and decisions will be premised on forecasts that are established jointly, and then physical and financial flows will occur. Pricing decisions and decisions to charge cost will influence the returns on assets and services, so the decision-makers and the accountants must have a clear picture of which asset or service is related to which financial flow.
The flows may be highly intertwined, and the way in which the components are linked together may change depending on which orders are processed. So one task, in the investment decision, is to trace from asset (physical and intangible) to activity and, from there, to the benefits and costs. A second task, for the profit share decision, is tracing backward from the benefits and costs to activities and assets. The solution is not really a new set of algorithms. First, any approach must be sensitive to the limitations of the underlying accounting information.
Second, the valuation of investment, in particular when it comes to new technologies and rapid technological change, need to be looked at with a different lens – discounted cash flows would need to consider which new options can be created by the project through ‘technological learning’ (Aggerwal, 1997). Third, decision-makers and accountants will need to change the manner in which they conceive and measure economic performance. Put bluntly, one might say that the ‘engineering-economics’ mentality (Degarmo et al. 1984), developed in large businesses, used cost/benefit measurement and capital budgeting to halt technological change – i. e. , employed to justify increases in existing capacities over new ones. The revenue-requirement capital budgeting model cannot easily be applied to networks since they create a set of benefits that may differ considerably from the traditional direct benefits usually examined in the conventional model. We can again consider the example of a modern utility network, whose characteristics can be generalised to cover almost all networks in many other industries.
The assets employed will not be active at all times and there is little distinction between their state of being ‘in use’ versus a state of being ‘idle’. They are more capital-intensive than expense-intensive and we frequently see sunk and stranded costs. The technologies employed do not deplete over time (i. e. , they have infinite durability) though they may become functionally obsolete when processing specific type of orders. The value of a used asset has more to do with technological change than with physical depreciation.
From the viewpoint of generating revenue, although the asset combination does not alter, the availability, quality and delivery options may be sufficiently altered to produce benefit categories which have previously not been conceptualised or understood. This type of technology management is most common in the electric power markets (Fernando and Kleindorfer, 1997). Together with this flexibility and optimisation, the cost/benefit attributes become indivisible. The network bundles operating cost, product throughput, gradations of product quality and flexible or reversible delivery options, nd the benefit stream cannot be easily delineated in terms of the network partners’ individual resource input. As outlined above, tracing benefits and costs to activities becomes essential, as does the issue of looking at shifting costs rather than cost saving. If we define the value-added attributes of an activity or a combination of activities as those which increase the availability or reliability of output in the most cost effective manner over time, we need to trace the origins of adding-value back to that combination.
We therefore should define clusters of combinations, relate them to the respective asset input, and then derive the respective return on investment. Thus, we will arrive at a basis for investment decisions. And, since this also works in hindsight, we have a basis for determining the profit attributable to the partner (or partners) who made such investment(s). Transaction costs Transaction cost, in its broadest sense, is the cost of conducting business with someone else (Axelsson and Easton, 1992). They are costs borne in order to allow an exchange between two parties.
From the viewpoint of activity analysis, this exchange equates to one party (A) asking the other party (B) to perform an activity (aB) because the cost of aB plus the cost of the business transaction between A and B (tc) is lower than if the activity would be performed by A (aA). Thus, from aB + tc ; aA, we can infer that the exchange will yield a benefit for A and B if the expected capitalised value of the joint undertaking (the ‘value of cooperation’, VC) is higher than the aggregate of the expected capitalised values of separate undertakings by A(VA) and B(VB).
In order to maximise VC, the partnership of A and B must minimize aA, aB and tc. This may sound trivial, but cost theory and cost accounting have neglected tc and VC for many decades. The concept of transaction cost was first introduced Coase (1937) and further developed by Williamson (2005), although accounting for transaction costs is still in its infancy (Van der Meer-Kooistra and Vosselmann, 2000). One reason for this slow development may be that Williamson’s conjectures aim at explaining the governance structures of inter-organisational relationships, nd referrals to management practice were not really sought nor intended (Dekker, 2000). Another (rather paradoxical) explanation might be the fact that numerous outsourcing decisions have been deemed to bring about ‘transaction benefits’ (but, when asked for the underlying accounting data, management are often in denial – Love and Roper, 2005) . Management accounting might furnish a substantial contribution to improving a network’s performance by demonstrating the impact of transaction cost reductions.
For this, we can refer to the four main reasons that Williamson cites for the existence of transaction cost, namely: (1) the inability of people to analyse everything in advance (Simon’s ‘bounded rationality’ (1976); (2) business uncertainty; (3) the presence of few buyers and sellers in any given kind of transaction; and, (4) ‘opportunism’ – i. e. , where individuals might try to take advantage of others. For instance, one form of opportunistic behaviour might be when an exchange partner increases the dependency on an asset that cannot be used outside a focal node of exchange.
Recalling our earlier discussion of ‘asset specificity’, we would certainly be in a position (as accountants) to measure the degree to which an asset loses productive value. However, when risks can be identified, there are also instruments to safeguard, and they will assist making a clear connection between risk behaviour and cost. Another (probably more operative) approach to measure and reduce transaction costs, would be to draw from the categorisation of Picot et al. (1996).
They argue that in all phases of relationships (preparation, handling, controlling and winding-up relationships), there should be such measurable items as: 1. search costs (measured by the expenses incurred in the search at special organisations or institutions, or for the use of telecommunication, online services, publications or consultants); 2. information costs (measured by expenses incurred when dealing with problems that disturb the exchange of information); 3. decision costs (expenses related to arriving at shared agreements amongst partners.
Decision costs may also be caused by contracts that were not fulfilled in the way they were negotiated, or by contracts that were not closed in their intended form); 4. bargaining costs (caused via negotiations – e. g. , lawyers fees, consultants, etc. ); 5. control costs (from the adaptation and supervision of transaction results – e. g. payments control or costs of arranging technical standards or quality); 6. handling costs (from the management of operations); and, 7. adjustment costs (e. g. , from the implementation of new laws or new IT-standards). A list of additional items could be attached.
However, reports with numerical expressions of these easily quantifiable items are rare in practice, as in extant research. If management accounting is intended to produce reliable information for contractual or regulatory arrangements, all the above variables (and more) must be quantified, and the price of producing the information must be in line with the benefits derived therefrom. Having said that, the extent to which transactional cost-related information should be established, its degree of exactness, and the frequency of reports will depend upon specific network situations.
For example, permanent networks require less-frequent reporting than those where partners enter and leave more often; dominant partners may demand more detailed analysis, and when risks are higher there may be more emphasis on accountability. COLLATING THE FOUR AVENUES FOR MANAGEMENT ACCOUNTANTS It has been considered how business network accounting can be nurtured by target costing, collaborative planning, quantifying transaction costs, and finding common denominators for returns on investment.
The aim is to develop notions that might enhance and support efficiency and effectiveness in inter-organisational activities. Interestingly, a review of the inter-organisational management accounting literature (Caker, 2001) suggests that more emphasis has been placed on how control of business relations can be achieved. Of case study research to date, most either deals with ‘dominant partners’, e. g. NISSAN and its role in the buyer-supplier network (Carr and Ng, 1995), or with situations where trust has not yet been built up between network partners (Thrane and Mouritsen, 2002).
It has also been demonstrated (Van der Meer-Kooistra and Vosselmann, 2004) that one might distinguish between accounting for the interest-alignment, accounting for the commitment-alignment, and accounting for the coordination of activities. Important differences may exist, such as ‘ex ante’ versus ‘ex post’-information, the extent of power/legitimacy base for the use of accounting, and the objectives of accounting – ranging from prevention of strategic opportunism to coordinating activities in the operational realm.
In this context, the following exhibits a grid (classification) of the four avenues set against the characteristics of bargaining and trust: Accounting for target costsAccounting for collabora- tive planning, forecasting and replenishmentAccounting for common denominators of ROIAccounting for transaction cost Bargaininghigh, unilateralhigh, multilaterallow, multilaterallow, unilateral Trusthigh, multilateralhigh, unilateralhigh, multilateralHigh, multilateral Exhibit 1: A management accountant’s ‘four avenues’, mapped against bargaining and trust.
Where bargaining prevails over trust, target costs are set through power (unilaterally) but high efforts will be made by the dominant partner to impose its calculus. All partners will establish their own budgets and enter into strong negotiations because they do not trust each other. Each partner will rely on its own denominators of ROI, and attempts to reach communal indicators will be low. Transaction costs will be high but there will be low effort to account for them and, if done, it will be a separate exercise and the outcomes will not be communicated.
However, where trust prevails within a network, the focus changes, and management accounting can contribute towards building trust. There is an argument that uncertainty (linked to an absence of trust) can be absorbed “either by building higher levels of trust or by building more extensive control mechanisms with the associated increase in information” (Tomkins, 2001, p. 180). However one views this, trust will cause transaction costs to fall.
Nevertheless, when network partners act upon trust rather than bargaining power, management accounting may be even more inclined to ascertain the magnitude of transaction costs and how they could be further reduced. Target costs and common denominators for ROI will be elaborated in a joint effort. And, in collaborative planning, forecasting and replenishment, the data provided by the pull-partners will be widely acknowledged and left to their discretion. Exhibit 2 (below) captures the implications for management accountants’ methods at each stage of a network’s development:
Start-upMarket penetrationMature networkRe-launch Bargaining, antagonistic motivationAnalyses of (given) target costsAccounting for collabora- tive planning, forecasting and replenishmentIndividual maximisation of profit share Sub-optimal minimisation of (some) transaction costs Trust, information symmetryConsensual accounting of target costsAnalyses of (given) data for collaborative planning, forecasting, replenishmentAccounting for common denominators of ROIConsensual accounting for transaction cost Exhibit 2: Management accountants’ methods at different stages of a business network’s development
As depicted above, there is ‘life-cycle-costing’ within networks. If the exhibit were three-dimensional, we could also visualise that in a network’s life-cycle there could be shifts from antagonistic motivation to homogeneous motivation, and vice-versa. The management accountant for his/her part will work out which instruments need to be chosen for particular stages of a network’s life-cycle. Management accounting, it is argued, should follow business structures, just like business structures follow business strategy.
This begs the question: does management accounting really do this? DISCUSSION With connectivity becoming one of the foremost constituent elements of competitive advantage, and with new features of an entrepreneurial mindset contributing to knowledge economy (McGrath and Macmillan, 2000), has the mindset of management accounting followed suit? Why, when networking strategies have led to new business structures and new forms of organisation, do we know very little about accountants’ business models (Sheehan, 2005) and even less about how they improve profitability?
It may be argued that, since management accountants have not yet completely found their role within business networks, they are reluctant to share their experiences. Otherwise, while a small number of researchers have used the ‘role’ concept in their studies of inter-organisational networks (Grandori and Soda 1995; Johansson and Borell 1999; Nunan 1999; Snow et al. , 1992), the explicit application of role theory (e. g. Montgomery, 1998) is rare. And, as far as the role of management accountants is concerned, it is practically non-existent (Knight and Harland 2005).
It is not suggested here which roles management accountants ought to assume. Rather, this paper offered a contribution towards a ‘survival kit’ for those accountants who work for a business network, and to help them to integrate into this new environment. When reviewing the literature on changes in accountants’ role(s) (Burns and Baldvinsdottir, 2005), it is easy to become reluctant to accept that the request for more ‘business-integration’ (IFAC, 2002) has been successful. But, it is a matter of urgency to close the gap between theories of knowing what needs to be known and knowing how to do what needs to be done (Pfeffer and Sutton, 1999).
Theoretical knowledge needs to be translated into practice and, even though the management guru way of solving problems through issuing practical guidelines for action sometimes deserves criticism (Norreklit et al. , 2005), ‘survival kits’ can be helpful. The recurrent enumeration of the components of business-integration (Jarvenpaa, 2001) still appear to lack applicability towards business networks. Rather, in business networks, the accountant has to cope with business-disintegration – i. e. with abrupt structural changes such as decentralisation of power, diversification of investment objectives, and varying numbers of network participants