Risk in the process of budgeting
Paul M. Collier* and Anthony J. Berry†
Risk in financial management has become a central pre-occupation of corporate governance(International Federation of Accountants, 1999; Institute of Chartered Accountantsin England and Wales, 1999) however little is known about how organizations construerisk in the financial management process.
留學生dissertation網(wǎng)This paper describes four exploratory and comparative case studies that were carriedout with the aim of understanding the relationship between risk and how managers
perceived and managed risk in the budgeting process.The case analysis results suggest that organizational participants socially construct risk.
Four domains of risk were identified: financial, operational, political and personal. Thesedomains of risk influenced the content and process of budgeting that were categorizedas Risk Modelled, Risk Considered or Risk Excluded. In particular, by excluding somerisks and considering others, the budget process was seen to be different to, and neededto be interpreted separately from, the content of the budget in which there was littleevidence of risk modelling or the use of probabilities. The research also showed how riskis transferred within and between organizations and how managers may contain risk.While these were exploratory case studies it was evident that risk and how it wasconstrued, understood and managed in the financial management process was a centralconcern of managers and that further research is necessary to extend this knowledge.c 2002 Elsevier Science Ltd. All rights reserved.
Key words: risk; budgeting; domains of risk; risk modelled; risk considered; risk excluded; risk
transfer; risk containment.
Introduction
Despite recent reports (International Federation of Accountants, 1999; Institute of
Chartered Accountants in England and Wales, 1999) that have emphasized the
importance of risk management, there is little evidence of research into risk in relation
to the content of budgets or the budgeting process.
Address for correspondence: Paul M. Collier. Aston Business School, Aston University, Birmingham
B4 7ET, U.K. E-mail: [email protected].
*Aston University
†Manchester Metropolitan University. E-mail: [email protected]
Received 28 March 2002; accepted 10 July 2002.
1044–5005/02/$ - see front matter c 2002 Elsevier Science Ltd. All rights reserved.
274 P.M. Collier and A. J. Berry
The purpose of the study described in this paper was to further understand
the relationship between risk and how managers perceive and manage risk in the
budgeting process. The study was an exploratory one, to establish similarities and
differences through a comparison of four organizations of different types.
The paper begins by locating the study in the literature of risk and budgeting. The#p#分頁標題#e#
second section covers the research methods used. In the third section, each of the
four cases is described and in the fourth section the paper identifies similarities and
differences by developing various categories. In the final section, some conclusions
are presented.
1. Theoretical framework
The Concise Oxford Dictionary definition of risk is ‘a chance or possibility of danger,
loss, injury, or other adverse consequences’. However, there is a general lack of
consensus (Bettis and Thomas, 1990) about the definition of risk in the management
literature where multiple definitions have been used (Baird and Thomas, 1990).
The International Federation of Accountants (1999) defined risk as uncertain future
events that could influence the achievement of strategic, operational and financial
objectives. The IFAC report contrasted the negative view of risk as hazard with a positive
interpretation of risk as opportunity. Their report was explicitly biased towards
a perspective that favoured shareholder value in which risk management ‘establishes,
calibrates and realigns the relationship between risk, growth and return’ (p. 4).
Risk was also defined by the Turnbull report as any event that might affect a listed
company’s performance, including environmental, ethical and social risks (Institute
of Chartered Accountants in England and Wales, 1999, p. 4). The guidance in the
report was based on Boards of Directors implementing ‘a risk-based approach to
establishing a sound system of internal control and reviewing its effectiveness’.
According to the report, Boards need to consider the extent to which each risk is
acceptable, the likelihood of risk materializing and the ability of the organization to
reduce the incidence and impact of the identified risk. However, reflecting the IFAC
definition, the Turnbull report acknowledged that ‘profits are, in part, the reward for
successful risk-taking in business’ so that the role of internal control was ‘to help
manage and control risk appropriately rather than to eliminate it’ (p. 5).
http://www.ttc-switch.com/thesis_sample/Australia_Thesis_sample/The often-used distinction between risk and uncertainty was derived from Knight’s
Risk, uncertainty and profit. According to Knight, risk was not knowing what future
events will happen, but having the ability to estimate the odds, while uncertainty
was not even knowing the odds (quoted in Adams, 1995).
The budgeting process is a formal method by which plans are established for future
time periods. This implies a consideration of risk and/or uncertainty. However, the
assessment of risk is problematic because it can be conceptualized by reference to
the existence of internal or external events; information about those events (i.e. their#p#分頁標題#e#
visibility); managerial perception about events and information (i.e. how they are
perceived); and how organizations establish tacit, informal (or explicit and formal)
ways of dealing with risk.
This paper proposes a definition of risk as the consideration of (a process) and
the consequences (the outcome—both fortuitous and hazardous) of unpredictable
and uncontrollable events and perceptions about those events. A consideration
Risk in the Process of Budgeting 275
of risk as much as opportunity as of hazard is consistent with the IFAC and
Turnbull definitions of risk. This definition accepts the notion of unpredictability,
while recognizing that techniques do exist (such as probabilities and flexible budgets)
to reflect a variety of outcomes. It also recognizes some requirement to represent an
assumed knowledge of future market and competitor activity, resource dependencies,
and action–outcome relationships in the content of the budget document.
The definition adopted in this paper is consistent with the observation by Berry
et al. (1995, p. 57) that ‘what one sees in a financial plan, especially one which is
projected on a spreadsheet as single-point estimates over 10 years, might be one
where the problem of uncertainty has been set aside’. While uncertainty may be set
aside in the budget, it may not be set aside in the process of budgeting.
The notion of unpredictability in relation to budgeting may be better understood
through the open/closed, rational/natural systems typology of Scott (1998, but see
his first edition in 1981). Otley et al. (1995) classified research on management control
into four categories (closed rational and natural, open rational and natural). Otley et
al. argued that research ‘needs to recognize the environment in which organizations
exist and to loosen the boundaries around the area of concern’ suggesting the need
to link rational and natural, open and closed systems perspectives.
Budgets can be thought of as management tools to portray in quantitative terms the
environmental–organizational interface. In so doing, budgets reflect environmental
constraints and opportunities and organizational strategies and goals for dealing
with the environment. As such, budgets are rational devices that assume an open
system. Using Scott’s (1998) typology, open rational systems encompass bounded
rationality, contingency and transaction cost models. In the open rational perspective,
budgeting is a descriptive model of the open natural system through input–output
modelling which assumes knowledge of the means–ends transformation process.
This paper calls this form of budgeting risk modelled. In a risk modelled budget
there is an explicit use of formal probability models to assess the effect of different
consequences over a range of different assumptions.#p#分頁標題#e#
Also in the open rational system, where organizations give attention to the
openness of the systems, the budgetary system may adopt rational risk devices
such as probability and sensitivity analysis. This form of budgeting is referred to
in this paper as risk considered. In the risk considered process, informal sensitivity
(or what-if) analysis is used to produce (for example) high, medium and low
consequences of different assumptions.
In Scott’s open natural system, organizations may conceive of the budgetary
system as a rational system and seek to close it. In essence this would be a budgetary
system designed within a protective boundary to act as some kind of quasi-rational
core. Here risk is explicitly excluded from the budgetary system and is managed
in some other domain. This paper calls this a risk excluded form of budget. The risk
excluded budget manages risk outside the budget process, and the budget relies on
a single expectation of performance.
Using Scott’s framework, managerial perceptions of risk can be seen as a function
of open natural systems while budgeting can be seen as a function of open rational
systems, with the potential for a lack of compatibility, even a tension between the
two. As such, the purpose of the exploratory case studies described in this paper was
to understand the relationship between these two perspectives and how managerial
perceptions of risk influence the process and content of budgets.
276 P.M. Collier and A. J. Berry
However, the discussions of risk in management accounting texts are most
commonly linked to closed/rational concepts and the use of probability, primarily
in the context of capital budgeting decisions, to reflect the unpredictability element. In
the latest edition of his management accounting text, Drury (2000) considered risk in
decision making through techniques such as decision trees, probabilities, standard
deviation and portfolio analysis. Similarly, research by Arnold and Hatzopoulos
(2000) suggested that financial managers’ use of financial theory and techniques
was continuing to grow. This implied a more educated managerial and financial
workforce and a greater likelihood that financial managers and accountants would
be open to more structured and analytic considerations of risk.
The literature of budgeting is wide-ranging; there have been studies of bias (Lowe
and Shaw, 1968); supervisory style and participation (Hofstede, 1968); budgetary
slack (Schiff and Lewin, 1968); budgets as behavioural tools (Buckley and McKenna,
1972); budget constrained and profit conscious styles of performance evaluation
(Hopwood, 1973); single point estimates and aggregation (Berry and Otley, 1975);
dysfunctional behaviour (Birnberg et al., 1983); the impact of environmental uncertainty
(Govindarajan, 1984); the achievability of budget targets (Merchant and#p#分頁標題#e#
Manzoni, 1989); budgets as symbol (Czarniawska-Joerges and Jacobsson, 1989); and
the power of budgets (for example, Covaleski and Dirsmith, 1988; Ezzamel, 1994;
Llewellyn, 1998). Despite this extensive literature, there is little evidence of research
into risk in relation to the process of budgeting.
A critique of budgeting in practice is provided byWallander (1999) who supported
production planning processes, but argued against the use of budgets. He argued that
businesses will either continue as usual, in which case previous experience is used
for budgeting and people continue working as before. If business does not continue
as usual thenWallander argued that there is no ability to foresee something of which
there is no prior experience.
The ‘rationalist’ model evident in much of the accounting literature has been
contrasted with a more ‘social constructionist’ one (Preston, 1995). Beck (1986, 1992
in translation) took a wider view than the individual or the organization and argued
that we live in a ‘risk society’, asserting that all risk was socially constructed.
Douglas and Wildavsky (1983, p. 14) explained risk perception as a cultural process,
commenting that ‘each culture, each set of shared values and supporting social
institutions is biased toward highlighting certain risks and downplaying others’.
The ‘cultural theory’ perspective of Adams (1995) differentiated the formal sector
of risk management, with its concern with risk reduction from the informal sector of
individuals seeking to balance risks with rewards.
In the accounting literature, the social construction perspective is represented in
work such as that by Czarniawska-Joerges and Jacobsson (1989, p. 29) who depicted
budgets as ‘a symbolic performance rather than a decision-making process; a means
of conversation rather than a means of control; and an expression of values rather
than an instrument for action’. Similarly, Brunsson (1989, pp 105, 106) saw the budget
principally in terms of talk and decisions: ‘Because of its relatively loose links with
action, budgeting provides a good instrument for conducting politics and producing
hypocrisy’.
Preston et al. (1992, p. 562) showed how ‘new patterns of language, meaning and
significance emerged through the fabrication of accounting and budgeting systems’.
Boland (1993, p. 139) concluded that the budgeting system mediated the influence
Risk in the Process of Budgeting 277
of interpretive possibilities but did not carry a ‘single, unambiguous interpretive
scheme that it imposes on the situation’.
This social constructionist approach is also reflected in the strategy literature.
March and Shapira (1987) suggested that managers were insensitive to probabilities#p#分頁標題#e#
but were focused on performance in relation to critical performance targets. March
and Shapira noted that both individual and institutionalized risk preferences were
important in understanding organizational responses to risk management.
Similarly, Bettis and Thomas (1990) argued that researchers have very little
knowledge about how managers in organizations perceive and take risks or of the
commonalities or differences between individual risk taking and risk taking by
managers in the organizational context. In contrast to the domination of the capital
assets pricing model and variance analysis in the literature, it has been argued that
researchers ‘have not captured the concepts of risk employed by managers and
investors . . . employing ex post measures that are decidedly different from managers’
ex ante risk assessments’ (Ruefli et al., 1999, p. 168). Ruefli et al. argued for ‘nextgeneration’
measures of risk that are ‘methodologically defensible, managerially
relevant and usefully generalized across settings’ (p. 179).
There have been few attempts to answer this call in the accounting literature.
An exception has been Harris (1999, 2000) who drew on psychological theories in
developing a project risk assessment framework to study risk assessment in capital
investment decision-making, in which managers used a range of analytical tools to
assess the likely risks and returns. Managers also drew upon their intuition and
influenced others involved in the decision process.
Given the concern of this paper with risk in both the content and process of
budgeting, the focus of this paper is on the social constructions of risk and how,
at the level of the individual manager, risk is perceived and incorporated into the
budgeting process. Before describing the case studies, the next section summarizes
the methodology adopted.
2. Methods
The exploratory case study (Yin, 1994) was chosen as the most suitable method for
the study because it enabled a study of the budgeting process and the perceptions
of various organizational participants as distinct from the content of budgetary
documents.
The selection of cases was intended to cover the business and not-for-profit sectors.
Within the business sector, the researchers wanted to separately consider large and
smaller organizations. Within the not-for-profit sector, the researchers wanted to
study a public sector organization and a voluntary organization. Within each of
these four groups the selection of a particular case was opportunistic in the sense
that the researchers had contact with organizations which agreed to participate.
The studies were exploratory rather than in-depth longitudinal studies and as such,
were not intended to be ‘ideal types’. Generalizability from these cases was neither#p#分頁標題#e#
intended nor expected. However, by selecting four very different organizations, the
researchers’ intent was to identify any similarities and differences in approaches
to risk in the process of budgeting across different sectors, rather than to improve
generalizability.
278 P.M. Collier and A. J. Berry
The study was conducted in each of the four organizations during 2000. These organizations
are referred to as S, T, P and Q, respectively. S was a single customer plant of
a multinational fortune 500 automotive parts supplier, an assembler and sequencer
of parts that exhibited the characteristics of a multinational. T was a manufacturing
firm, a subsidiary of an unlisted management buyout. P was a non-metropolitan
police force. Q was a voluntary sector organization that provided both direct services
to clients through funded projects and contributed to national policy debates.
These cases were not selected to be in any way representative of larger groups,
but as diverse organizations that might illustrate different approaches to risk and
budgeting. They were, however, all parts of larger national organizations. This
resulted in the (deliberate) exclusion of financial market issues from the study.
Contextual information was obtained from a variety of published documents.
In S, the researchers did not have access to financial reports but gained information
from internet searches. In T, access was given to the internal budget and monthly
performance reports and to the group annual financial statements of T’s parent. In
the case of P the sources were published annual reports, historical budget documents
and minutes of meetings. In Q data was gained from published documents.
The major form of data collection was via visits with some observation but mostly
from interviews. In S interviews were conducted with the plant manager, logistics
manager and finance manager. In T, managers with responsibility for manufacturing,
engineering and design, logistics, quality and finance were interviewed. In both S
and T, visits to and observations of the factory operations helped the researchers
to understand the explanations provided. In P, interviews were conducted with an
assistant chief constable (ACC), the finance director and finance staff at headquarters
and with a superintendent and support managers at operational divisions. In Q a
regional director was interviewed.
In these interviews, unstructured, open questions around the topic of risk were
adopted, allowing the interviewees’ own understandings of, and perceptions about
risk to emerge. The interviewees’ own perceptions of risk were used in the
researchers’ descriptions of each of the four case studies. These perceptions of risk
were not seen wholly, or in some cases even partially, from a budgeting perspective.#p#分頁標題#e#
In each organization, interviewees were asked about the process of budgeting that
existed in the organization. However, as the cases show, the process of formulating
the budget in each organization was top-down and the interviewees spoke more
about risk perception than about budgeting, which was seen in each case as
externally imposed.
The interviews were not tape recorded but detailed handwritten notes were taken.
At S and P, both researchers attended and separate notes were taken, which were
then compared. This ensured that the approach of the two researchers was consistent.
One researcher subsequently visited T while the other visited Q. In each of the four
cases, more than one visit was made to the organization. The case descriptions use
direct quotes from interviewees to represent the various perceptions of risk that were
identified during the studies.
Despite differences in the number and range of interviews and differential access
to documentary evidence (which reflected differences in the type of organization),
the access to both financial and operational managers provided an adequate
representation of the size and scale of operation of the organization for the purpose
of an exploratory study. The next section describes each of the case studies.
Risk in the Process of Budgeting 279
3. Case studies
Each of the four cases (S, T, P and Q) is reviewed in turn. For each, a brief summary of
the organization is followed by a description of operational and financial risk. In the
fourth section, the categorization of risk is considered in greater detail.
S
S was a supplier of automotive parts. The plant assembled and sequenced components
for a single customer, an adjacent vehicle assembly plant. It had been in operation
for 21/2 years, employing 240 people over three shifts, 5 days per week. Three
hundred suppliers provided over 10 000 components to produce 2500 finished parts.
There were two distinct operations in the business: assembling S’ own products; and
sequencing S’ own and other suppliers’ parts for the customer. Each contributed
about half of S’ volume. Although S was part of a Fortune 500 multinational automotive
group, it was unique within the group because it sequenced multiple products.
Operational risk
There was high variability in both operations as the customer could change—at short
notice—the build specification to bring forward special orders. In the sequencing
part of the business, S received electronic data interchange (EDI) transactions from
the customer and controlled the flow and sequence of materials from suppliers,
delivering to the customer’s production line in required sequence ready for assembly.
S faced considerable unpredictability because it did not build for stock but to order.
The plant manager commented that, ‘the Friday night MRP [manufacturing resource#p#分頁標題#e#
planning] run has been replaced with the need to get parts on site quickly’. To achieve
this, S had extensive EDI links via the internet with its major suppliers. Each day, S
received a 20-day production forecast. Orders were placed on suppliers who used
the EDI link to advise the stock they were able to ship.
The final sequence of assembly for S was determined by vehicles leaving the
assembly plant’s paint shop, which allowed 1–2 h notice to S. This applied both to
S’ own components and its sequencing of components from other suppliers. The
assembly plant was reliant on S’ sequencing to avoid loss of production time as
many parts were critical and, if unavailable, stopped the customer’s production line.
S provided 24-h customer service to solve problems identified by the customer.
The plant manager argued that S ‘was more systems dependent than technology
dependent’. Two systems were used: SAP/R3 for MRP and financials; and new
internet software developed, very recently, by a computer system supplier for the
sequencing operation. Although the sequencing technology was owned by S, they
had not incurred any investment in this system as the system’s supplier intended to
sell the system within other group companies to recover its development cost. The
computer systemsupplier was on site 24 h a day due to the dependence on the system
by S and its customer.
S carried a risk—on behalf of their single customer—that suppliers might not be
able to achieve S’ schedule, either on quantity or quality. S needed to know that
suppliers would load the components required and what stock was in transit to
ensure they were able to fulfil the sequencing required by their customer. The risk
was that the components were not available or different items to those ordered were
shipped. In particular, by using the internet technology to confirm and expedite
280 P.M. Collier and A. J. Berry
orders on suppliers, S needed to know when special action was required. Helicopters
had been used for urgent deliveries to maintain the customer’s production line.
The inbound transport and logistics had been subcontracted to a single company
that picked up parts from suppliers and delivered to S, using bar codes to track
vehicles to meet delivery windows at the rate of 4–5 per hour. About 35 ‘mega
trunkers’ were delivered to S each day. The logistics supplier’s staff was also on site
24 h each day.
Financial risk
Prices were negotiated between S and its single customer against a contract for
both components and sequencing, in which volumes were estimated with a ±3 per
cent flexibility. There was a price per vehicle for sequencing and a fee for logistics,
expediting and scheduling. Plant overheads were based on a fee per unit, but this#p#分頁標題#e#
was agreed 2–3 years in advance for the life of the customer’s model.
As the single customer specified the supplier and prices, any increases were passed
onto the customer. As logistics costs depended on the location of suppliers, costs were
also passed through to the customer. Exchange rate variations were also passed on.
As materials and freight accounted for 90 per cent of sales revenue, there was little
scope for extra profit to S.
Contract clauses minimized volume risks but S had employed labour to provide
the capacity for its customer and needed to safeguard the revenue stream if their
customer stopped production. When the customer’s production line had stopped
for 7 h, S was unable to build ahead because the customer’s paint shop dictated
sequencing. S therefore had idle labour, which could only be used on maintenance
tasks. Although there was no immediate compensation, the 3 per cent volume
flexibility resulted in some year end negotiation.
In contract negotiations and bidding for new business, price was critical in dealing
with the customer’s purchasing staff. At the time of the research visit, the customer
planned to reduce production from three to two shifts on one model due to capacity
constraints in their paint shop. S was bidding (to the same customer) for replacement
business for a different model and was attempting to minimize redundancies of staff
by winning this business. However, S was clearly feeling price pressure from its
customer.
S’ parent group’s budget targets were based on volume, but S, having a single
customer, was dependent on the volumes dictated by that customer. S was unable
to influence those volumes. However, the finance manager explained that even
after budgets were set, the customer could introduce a new model or increase the
specification for an existing model. Budget variations were therefore a consequence
of volume and design changes.
Premium freight was a significant cost to S. Even though the supply contract
was between the single customer and supplier, S undertook all the ordering and
scheduling, and so was responsible for ‘just in time’ (JIT) delivery. S was therefore
heavily reliant on its systems. The cost of supplier parts sequenced by S were paid
for by the customer direct to the supplier, so S was unable to recover premium freight
costs from suppliers, even if the fault lay with the supplier. S’ contract dictated 2
days’ stock of all components, but S was physically unable to carry this amount of
stock for all components, so the premium freight cost was often borne by S, which
eroded their margin.
Risk in the Process of Budgeting 281
Profit improvement for S came primarily from improving inventory turns through
better management of the supply chain. S was achieving an annual stock turn of#p#分頁標題#e#
between 80 and 100 but wanted to reduce stock further. The logistics manager
emphasized that ‘the level of stock impacts on finance for our own parts and space
for supplier-owned parts held for sequencing, so we aim for more frequent deliveries
of smaller quantities. The trade-off is between full loads and lower inventory’.
While cost savings on parts had to be shared with the single customer, extra profit
could come from operational improvements. The system used by S was as lean as
possible, but improvements still focused on labour utilization. Extra profit was also
possible from additional services. An example was the rationalization of the number
of batteries used and re-labelling by S’ employees.
Performance was measured by S’ parent via a plant net income after tax and
interest. The target was 5 per cent return on sales but S was achieving only about half
of that. This was considered acceptable because the business was low risk. However,
against a target return on net assets of 12 per cent, S achieved 40–50 per cent because
it had a low investment in plant. The risk of exchange rate volatility impacted S’
performance because they reported in $US.
While a variety of other risks were identified by the plant manager—health
and safety, environmental and financial—the major risk was in failing to meet the
requirements of the single customer and so cause the customer’s production line to
stop. Ultimately, the customer had the right to take control of the plant if S failed to
perform. However, S could also walk away from the contract with 6 months notice,
and the customer was obliged to take over the building lease.
T
T was a subsidiary of an unlisted automotive parts manufacturing group that had
been the subject of a management buyout in 1995, heavily financed by a number of
institutional investors. The financiers’ 3-year exit strategy had been missed as a result
of poor financial performance and this had provided the motive for change from ‘a
maintenance strategy to a development strategy’. The strategy also involved making
each part of the business visible and identifying whether each part was profitable.
Perhaps as a result of these concerns, decision making was highly centralized, with
directors having power of veto over many day-to-day decisions.
T had been established in the 1930s and had at its peak 9000 employees. At the
time of the research there were 300 employees. After the sale of a significant part of
its ‘after-market’ business some years previously, T was generating annual sales of
about £30 million from the sale of about 2000 different parts to 130 customers. It had
a stock turn of 7–8 p.a. on inventory of £3M.
T’s principal objectives were financial: to satisfy its institutional shareholders#p#分頁標題#e#
through operating profit (before interest and taxation) and operating cash flow. The
company had a bonus scheme that rewarded managers for meeting or exceeding the
targets for both measures. The company was profitable, generating a gross margin of
25 per cent, an operating margin of 15 per cent and an operating cash flow of 22 per
cent of sales.
An ongoing restructuring programme included relocation to a single site, substantial
capital expenditure on computer numerically controlled (CNC) equipment
and changes to working practices. Despite a recent reduction in the workforce,
the company had not made, and had a policy against, compulsory redundancies.
282 P.M. Collier and A. J. Berry
However, T did not inform employees about the company’s financial performance
for fear of resistance to restructuring, increased wage demands or competitor access
to information.
T’s business was largely machining, drilling and milling turned parts and assembly.
Much of the equipment used was very old (a typical machine had a 1977 plate).
The equipment was designed for a small product range with large volumes of
similar parts, although the market had changed to low volume, high variety parts.
For example, one machine required 120 h to changeover for 8 hours running. This
changeover was carried out weekly. Machines designed for 50 000 batch runs were
producing batch sizes of 2–3000 and sometimes batch sizes in hundreds. The gradual
investment in new CNC equipment had resulted in lower changeover times and T
was slowly increasing its capability for high variety and small volumes.
Operational risk
The company’s main concern was a decline in its customer base. Historically, T had
been a major supplier to large automotive assemblers, but had changed from that
market to an emphasis on after-market sales. Having disposed of that part of the
business, T was seeking to rebuild its market as a tier 1 supplier to the large assemblers.
However, despite efforts to rebuild its customer base, sales were declining.
The engineering and design manager was responsible for new products and
product changes. These were mainly original equipment manufacturer (OEM) driven
and could take 2–3 years. He said ‘the risk is that we gradually taper off what we are
selling, because our customers are strong about what they want.We also haven’t got
the resources or facilities to meet demand’. A logistics manager added ‘the risk is
getting replacement business, or else the company will continue to diminish until
the site won’t support the core costs’.
The quality manager stated the concern that had also been expressed by other managers:
‘this is a sensitive period . . . we need new business . . . money is going out but
not coming in. What is the future?’ He added that ‘the company has good knowledge#p#分頁標題#e#
and experience in engineering, but it is not a customer oriented company’.
T was not part of its customers’ JIT systems, although it did have to meet timed
delivery slots. T wanted to make to order but needed to become more flexible to do
so. Due to the large batch sizes, production inflexibility and timed deliveries, a buffer
stock was needed. Despite this, on-time delivery (OTD) performance was only about
60 per cent.
T used BPCS, a manufacturing resource planning (MRP2) system which produced
a weekly schedule, checked parts availability and sent build orders to the shop floor.
One of the logistics managers emphasized that ‘we have a shaky supplier base—a
‘you’ll get what you get’ attitude—which makes it difficult to satisfy production’. He
added, somewhat cynically: ‘we plan not to deliver . . . we use the lead times in the
system but can’t assess whether parts are available or not. We say we can do it but
then we don’t do it’.
The logistics manager added that ‘stock accuracy is crap’ with only 89 per cent
accuracy. ‘This means there is an 11 per cent chance of failure . . . we have stock losses,
most probably because we don’t transact properly, we don’t tell the system what we
are doing’.
The manufacturing manager, who was responsible for 220 of the 300 employees
described the plant as a ‘1960s or 1970s environment’. He emphasized the problems
Risk in the Process of Budgeting 283
of ‘an ageing workforce, in their 40s and 50s who have been working in the same
pattern for 30 years . . . they have been cocooned and have an old mentality and
old working practices’. T had a strong union whose working practices included
demarcation of functions. These working practices were being addressed as part of
ongoing wage negotiations.
Financial risk
The financial controller commented that the financial control emphasis had been on
material costs and labour and that T’s success had been achieved through keeping
manufacturing overheads down. He explained this using the example of overhead
approvals. ‘We use an expenditure authorization form for any expenditure over
£500 which must be signed by a director to approve expenditure from the budget.
Below £500 a low value order is used, which must be signed by a manager’. The
financial controller added that ‘the focus is now on internal reporting. It hasn’t been
on budgetary control, but it will be’.
The budget reflected sales expectations and current material costs but the difference
between actual production output, labour and overhead costs and the standard
manufacturing costs (SMC) held within the computer system were reflected in a
10–15 per cent labour and overhead under-recovery. SMCs were held by the MRP2#p#分頁標題#e#
system, although the labour and overhead cost data were at least 2 years out of
date. The result was a disparity between budget costs and the SMC for labour and
overhead, which represented almost 50 per cent of sales. By contrast, material costs,
which represented a third of the selling price, were tightly controlled through a
material price variance (MPV) reconciliation.
SMCs were ‘massaged’ where equipment was used well below capacity as a result
of pressure from the sales department, as the use of actual utilization rates led to
costs that were too high. The old plant meant that most assets were fully depreciated.
A result of new capital investment was that depreciation costs were increasing
substantially and there was pressure on finance from sales not to reflect the increased
cost in SMC as it would make products too expensive. Consequently, the old
equipment appeared more cost efficient and work was being diverted from new CNC
machines to old labour-intensive equipment as a result of the comparative SMCs.
The financial controller considered the risks facing T as ‘losing support from
investors, getting new sales, and losing people, and hence expertise and facing
customers with people who have inadequate knowledge’. In addition, many
customers were seeking selling price reductions, although in the short term it was
expensive for customers to shift suppliers, as a consequence of long approval and
testing processes for safety-critical parts.
Despite the high level of profitability being achieved, there was a concern
expressed by most managers (who were unaware of actual profitability) of the risk
of institutional investors losing patience and pulling out of the business altogether,
either by closing it down or selling it off.
P
P was a police force serving rural counties with a number of small towns. There
are 43 police forces in England and Wales, each headed by a Chief Constable with
operational autonomy, although a Police Authority for each force area and the Home
office makes up the tripartite structure of governance.
284 P.M. Collier and A. J. Berry
In 1999–2000, P employed over 3000 people, of whom 2000 were police officers.
The total employment-related expenditure (including pensions) was 88 per cent of
the revenue budget. The police force was organized into six geographically-based
divisions, with Headquarters departments providing specialist support for serious
crime, motorway policing and major incidents. Headquarters also provided the
support functions such as finance and administration, estates, communications and
information technology, personnel and training services.
Operational risk
The demand for police services and the pattern of work was amatter of the incidence
of crime and how P chose to respond to crime compared with the effort directed#p#分頁標題#e#
into non-crime, quality of life issues and crime prevention. P was, like any police
force, subject to major incidents, whether criminal or not, such as the Gloucester
multiple murders by Fred West or the Lockerbie plane disaster. The likelihood of
such incidents was believed to be low, but possible. Such events could have a major
effect on police work as they would have to be met out of current budgets or reserves.
The finance director commented that ‘we budget for the plane staying in the sky’.
Organizational performance was measured largely by a suite of non-financial
performance indicators, covering, for example, the speed of response to ‘999’ calls,
levels of crime and rates of detection, deaths and injuries in traffic accidents
and surveys of public satisfaction. P’s area was seen as a low crime area when
compared with other police forces, particularly those in urban areas, but it suffered
comparatively with the lack of resources to police a large rural area.
A key business risk was achieving the performance indicators, despite falling
numbers of police officers, itself caused by financial pressures. ‘Our operational
plan’ said a Superintendent, a divisional commander ‘is linked to resources—
budgeted posts [the number of police officers at each rank]. Each year we draw up
a programme . . . and decide what is essential. Our only real flexibility is to pay for
things through saved budgeted posts’.
There had been a shift in the emphasis of national performance measures back to
crime and detection rates. However, police forces lack the ability to control many of
the causes of crime and the ‘script’ contained in the local policing plan was not being
followed by offenders. An ACC commented: ‘In public institutions, maintaining
public trust is the biggest risk’. The ACC also highlighted the ‘risk of failure to take
measures to avoid risk’ brought about by human rights legislation and case law.
Previously, the public couldn’t sue the police for operational decisions, because of
public interest immunity. Now, ‘we don’t just have to justify a specific decision, but
the balance of resources we have applied’.
In P, strategy was concerned with using the available budget to develop new styles
of policing (intelligence led, problem solving, etc.) while needing to use available
funds to satisfy short-term targets for the number of crimes and detection rates
and other performance indicators. However, divisions found it difficult to plan
because of the uncertainty of new legislation, unpredictable demand (‘999’ calls) and
the number of incidents, over which they had little control. There was no facility
to cost policing activity, so there was no link between the financial plan and the#p#分頁標題#e#
operational plan. A divisional support services manager explained: ‘Performance
indicators are forced on us and the budget is separate . . . If the plan is to reduce
burglaries, we don’t allocate funds to this . . . there is a new emphasis on youth
Risk in the Process of Budgeting 285
offending, so we are resourcing a new style of policing, but we still need to meet
demand’.
At the individual level, personal risk existed in terms of rewards and sanctions for
the implementation of decisions. In particular, police officers could be promoted or
obtain preferred postings based on their success as measured by the organization.
The divisional support services managers also expressed concern about their own
roles if financial pressures caused the amalgamation of divisions.
Risk was seen in the effects of what the superintendent called ‘squeeze’. The impact
of less police officers and more demand was an increase in stress-related illness.
The sickness rate had increased from 10 to 14 per cent and the increase in medical
retirements then effected the pensions budget.
Financial risk
At the chief officer level, there was evidence of a merger between the financial and the
operational. The ACC explained that chief officers think in terms of risk management.
Their aim is ‘flexible response systems’ that provide the ability to change fast enough
to meet new needs. ‘We consider risks all the time but committing real money makes
them concrete’.
The national funding formula, which determined the standard spending assessment
(SSA) was the main factor in the budget allocation by the Home Office to each
police force. The result of the formula-based settlement meant that P received £96.2
million from central government, an increase of 3.6 per cent. Police authorities are
precepting authorities and the police levy (part of the council tax) determines around
one-quarter of the total funding for each police force.
P’s budget for 1999/2000 had been £116.5 million. This included a compulsory
vacancy factor of £1.7 million which had been self-imposed following central
government budget cuts in that year. The vacancy factor required budgeted posts
in each cost centre to be held vacant in order to fund the shortfall and to ensure
budget constraints were met. For 2000/2001, P required a ‘standstill’ budget of £125.7
million, an increase of 7.9 per cent to cover expected commitments while maintaining
the same vacancy factor.
The superintendent, a divisional commander with 300 staff and a budget of £7
million gave an operational perspective: ‘now everything has a cost attached and we
can say publicly for the first time that financial constraints are starting to bite. The
impact of the vacancy factor for [this division] is 10 constables—that’s equivalent to#p#分頁標題#e#
withdrawing policing from a town’.
The gap between the required budget and the central government contribution had
to be met either by further cuts to the number of police officers or through an increase
in the police levy. P provided five options to the police authority, ranging from
increasing the vacancy factor to restoring it to zero. The financial impact of the police
authority’s decision was expressed in terms of the number of police officers because,
as the finance director remarked ‘people don’t understand policing outputs, only the
number of police officers’. The police authority adopted the option to increase the
police levy to maintain the existing number of police officers ‘because it is essential
to maintain existing staffing levels in order to meet the challenging targets contained
within the Policing Plan’ (Paying for Policing: Council Tax 2000/01 leaflet).
The finance manager added that ‘part of the risk is political. The budget can
be technically superb but how will it play for politicians and in the press?’
286 P.M. Collier and A. J. Berry
These political risks were perceived to be highest at the force level, while financial
and operational risks tended to be located at divisional level.
The budget was considered low risk because, once the police levy had been
determined, predictable income arrived on known dates. Eighty per cent of the costs
were also predictable, being payroll-related, and volatility was low. The financial risk
was in providing for major operational incidents and police pensions that were part
of a nationally unfunded scheme. This risk was met by the holding of reserves, with
a target of 3 per cent of the annual budget. There was also the risk (for future years)
that the government could impose a capping limit on increases in the police levy. The
perceived political impact of P holding reserves had been addressed by P moving
general reserves into specific reserves.
P had recently instituted a process of devolution of budget management to the
commanders of the operational divisions. The principle effect of this was to allow
divisional commanders to vire between budget heads, ‘trade-in’ staff for other costs
(e.g. equipment), and vary the allocation of resources between functions to meet
local priorities. Most importantly, devolved budgets allowed operational reserves to
be carried over by each division from one year to the next. The finance manager
commented that ‘the risk of over spending or under spending by divisions was not
noticeable at a corporate level. Now one division’s under spending is not available
for someone else’s over spending’.
Q
Qwas a national voluntary sector organization and a registered charity. It was a direct
service organization providing help and support to families and children in a wide#p#分頁標題#e#
variety of settings. Q also undertook some campaigning work for a wider recognition
and understanding of the severe difficulties in which families and children can find
themselves. Like many voluntary organizations, it had a substantial professional
staff, and recruited volunteers to work in line with their values, often in acute
conditions alongside their clients.
The regional director had about 70 full-time staff, about 20 part-time staff and
approximately 250 volunteers. There were 11 project leaders managing 17 ongoing
projects. In common with much work of this type, Q delivered services in a variety
of ways, but most commonly through projects. These projects were funded by Q and
by government departments and agencies.
Operational risk
By contributing its own money and commitment, Q was able to exercise some
independence and maintain some control of project work and its significance. Q was
only willing to take on projects that were innovative and creative, which explored
and tested out new ground. Hence Q did turn away requests for some kinds of
work. By contrast, other voluntary organizations in this and similar fields had taken
on service contract work of a semi-permanent nature that was similar to facilities
management.
An important aspect of risk management was the management of reputation. The
‘issue of . . . credibility is crucial to Q’ which meant that the quality of the staff and
their work had to be high. The local authorities (LAs) gave positive regard to the
organization—their work was good, and as the regional director said ‘we are seen to
have integrity as we bargain, seen to be honest and fair in the process’. She added
Risk in the Process of Budgeting 287
that ‘we put a lot of effort into this boundary management role and in to managing
relationships with the LAs and government’.
However, difficulties can arise through press exposure of incidents both in the past
(e.g. the uncovering of abuse in children’s homes many years after it had happened)
and in the present. The fieldwork took place at the time when a U.K. national
newspaper was publishing photographs and addresses of convicted paedophiles.
The aggression and anger released by this process resulted in rumour, following
which innocent people had their houses and cars burnt. The regional director spoke
about the risk this posed to Q: ‘Can we continue to persuade the public to give us
money to do our work given some of the history of child abuse’. There were television
programmes and other publications about the export of children from the U.K. to
Australia and Canada, a process in which Q had been involved. The regional director
continued ‘we cannot be sure that what we do and what we don’t do now is that
about which, in 50 years time, people may throw up their hands in horror’.#p#分頁標題#e#
These issues give rise to personal risk to employees and volunteers. A role of the
manager in relation to risk was to ensure the coping, resourcefulness and capability
of the individuals who handled local situations. This involved providing training
and development but also enabling individuals to contain their anxiety in a culture
that valued open and honest communication, sensitive leadership at every level, the
practice of learning from experience and achieving high standards.
At a senior management seminar on risk management in Child Protection in 2000,
the regional director said the trainer ‘spoke about anxiety and containment . . . . It
helped open up the issues of anxiety of working with children at risk—there had
been five deaths of children recently [not only in Q’s care systems]’. A glimpse
of this culture was the occasional ‘primal scream’ e-mail which was accepted and
understood. ‘It gets dangerous when that doesn’t happen’ the regional director said,
‘there is a risk when staff do not respond’ to the severe issues in the field of work.
The regional director said that she worked with ‘the management procedures, but
was also responsible for and/or involved in framing boundaries and guidelines’. It
was hard to ‘get the right level of procedures and guidelines’. Sometime she ‘pushes
her luck’ but was ‘allowed to get away with things’. She said that she was ‘learning to
cope with risk’ (by implication in a natural learning or experiential learning manner).
In part, the risk was managed by ensuring that effective training took place.
Q had a range of disciplines in their staff: teaching, health, housing, secretaries,
administration and accountants. Q had a training contract with a national training
and certification body with elements of both academic and workplace training. There
was an annual training needs analysis for teams and individuals in relation to the
organizational requirements. There was in-house child protection training at two
levels. Training was given on anti-discrimination practices and there was induction
training that was mandatory for all new staff. Training also had the benefit of keeping
the quality of work up (avoiding problems) and of retaining staff.
Financial risk
The financial management of Q was based around a national planning process.
In this some 30 per cent of the total costs were met from Q’s resources (funds
raised from charitable appeals to organizations and individuals) and the other
70 per cent from grants and project funding from central or local government
departments and agencies. Q was subject to the financial reporting requirements of
288 P.M. Collier and A. J. Berry
the Charities Commission. Project finance was available for new initiatives, subject#p#分頁標題#e#
to competitive bidding against tender specifications. The projects were audited, in
terms of objectives and outcomes as well as financially.
There were many initiatives from the government in relation to social inclusion
partnerships and all projects had deliverables. To further the work of Q, the regional
director ‘never stops thinking about funding for new projects’. She gave an example
of the street teams where funding was being phased out over 3 years. Each project
had a different funding basis that carried implications for staff employment. New
government funding was only available for new work, not to continue funding
existing work. As a result, new projects had to be created in line with changing policy
objectives. Q was an organization that sought to give expression to its values. Hence
it organized its finances to break-even each year. It reported that fund raising was
becoming more difficult. It could be that the national lottery with its financing of
good causes had siphoned off some of the cash flow to charities.
The regional budget (approximately £6 million) was constructed from the funds
allocated from the centre and the expected project incomes. The financial year
matched that of government. Budgets were single point estimates, with considerable
devolution to project leaders who were expected to manage their budgets. These
managers had leeway for virement between budget headings subject to maintaining
the basis of the project.
Typically the centre required the region to work within its allocation and would not
readily provide extra funds to cope with problems. However in cases of significant
project change, the centre would offer additional financial help for restructuring
costs. Hence the central funds were not viewed as an operational contingency, but
rather as a strategic contingency.
There was a system of monthly reporting. In October there was a 6-month review
of the financial year to date in comparison with the budget and forecasts to year-end,
at which time the budgets might be adjusted. In November, papers were sent to the
LA funders to set out the expected financial projections for the next year. Sometimes
negotiations followed if the LA wanted to cut the budgets and hence reduce the
project financing. Financial risks for the projects were continuous because most
work was externally financed via LA and central government grants. These funding
bodies had become much more demanding of outcomes, management processes and
assessing the value of the work.
4. Discussion
Risk
It was observed that each of the case study organizations was facing a mini-crisis
involving risk at the time of the research visits (perhaps this is always the case?).
S was in negotiations with its single customer over the price to supply parts for
a new model given volume reductions in an existing model. Managers in T were#p#分頁標題#e#
expressing significant concerns over the viability of the business in the face of a
steady decline in sales and poor delivery performance. P had presented options to
its police authority in relation to the precepting decision that would determine the
number of police officers. Q faced press reports and public anger about paedophile
activity that influenced fund raising and public support. Each example enhanced the
Risk in the Process of Budgeting 289
interviewees’ general perceptions of risk with situation-specific illustrations of risk
as it pertained to the process and content of the budget.
The consequences of risk varied for each organization. For S, financial risk was high
for the plant studied due to its dependence on a single customer, although it shared
operational risk with its logistics and computer systems suppliers. However for S’
parent, the plant was only one among many. For T, the principal risk was the removal
of financial support by institutional investors, or operationally the degradation of
customer satisfaction with delivery such that the level of business would continue to
decline. For P, the risk given a cash-limited budget was the inability to satisfy public
demand given the ‘squeeze’ on resources and the local political impact of the tradeoff
between police numbers and an increased council tax levy. For Q, the primary
risks were to its reputation as a result of press coverage and the consequential risk to
the continuity of funding, as well as the personal risk to employees and volunteers
coping with anxiety.
Budgets
In each case, the researchers asked interviewees about the process of budgeting that
was used in their organizations. As the case studies show, it became evident in each
organization that despite managerial perceptions of risk, these perceptions were not
evident in the budget process or the content of the budget document. In S and T, topdown
budgets were established by the parent boards with no explicit regard to risk.
In P, once the budget was set by the police authority, there was little financial risk due
to the predictability of cash flows. In Q, budgeting took place within known grants
and resource allocations from the national body.
In each case, budgeting was top-down, driven either by targets or cash limits.
Despite the managerial perceptions of risk that covered wide-ranging issues, the budgeting
process was largely protected from the influence of those perceptions. Thus,
applying the typology of Scott (1998), perceptions of risk appeared to be an open natural
system, while the top-down imposition of budgets was an open rational system.
Risk construction and domains of risk
Risk was constructed by managers of S in a narrow sense, and mainly from a
technological, systems-based perspective. This was a result of the finite boundaries#p#分頁標題#e#
that had been put around the business unit: first as a single customer plant that had a
relatively short life expectancy; and second as a subsidiary of a fortune 500 company.
In S, operational and financial risk was inter-related. The single customer risk was
offset in contractual terms that minimized volume variations and passed on price
increases during the contract period. The most substantial risk of contract failure
was insignificant given S was a business unit of a multinational company. Financial
risk in terms of the expected return, while not meeting parent targets, was considered
satisfactory because there was little investment in plant and lease obligations could
be passed to the customer on 6 months notice of termination.
T’s boundaries were defined by the exit strategy of its investor and in relation to the
organization’s history, technology and product range being out of step with market
changes. This may be evidence of a prior process of misconstruction of risk. There
was evidence from the interviews that these issues of risk were being reconstructed
before, during and presumably after the time of the study. The main risk to T was the
loss of investor support and retaining sales in a very competitive market. This risk
290 P.M. Collier and A. J. Berry
was exacerbated by the poor flexibility evident in its equipment capability and the
working practices of its employees, together with difficulties that had led to poor
delivery performance that could easily result in lost business.
The construction of risk was also a continuing aspect of P’s world that combined
uncertain demand, limited resources and unclear and ambiguous relationships
between inputs and outcomes. P was continually aware of the social and political
risks that were associated with national expectations of productivity improvements
and financial constraints. In P, the research identified the notional separation of
operational and financial risk, given the absence of connection between demand
and resources. These were becoming more closely connected as the operational
implications of financial constraint were being recognized. What was particularly
evident in P was the political risk in managing relations between the police force, the
home office, the police authority and LAs over precepting decisions.
From one perspective, Q understood how funding agencies and the public saw
the issues with which they were trying to deal. From a second perspective, risk was
constructed in relation to clients and their families and in relation to the effect on
staff. There was an enhanced awareness of organizational and personal functioning
in relation to personal risk that fed into managerial processes and roles. In Q, the
operational risk was evident in the need to maintain the organization’s reputation#p#分頁標題#e#
while simultaneously being innovative. Financial risk was represented by fund
raising constraints, the outcome rationality of funding agencies and perceptions
of the public. Particularly evident was the need to balance the delivery of project
outcomes with the personal risk to its staff and how this was addressed by training
and by managers’ monitoring how well staff handled difficult situations.
In each organization the context of unique circumstances, history and technology
had led to different processes of construction of ideas about risk. In the private sector,
firms S and T, this came from a form of economic and technical rationality. In P and
Q it came from social, political and ideological change.
The cases have differently illustrated the manner in which organizations and their
actors created domains in which risk could be understood and managed. Four
domains of risk—financial, operational, political and personal—were found to exist
and could be applied to each of the four cases, albeit in different degrees. They are
shown in Table 1.
Risk was perceived in each of these domains (as an open natural system) but these
domains were isolated from the budgeting process that was dominated by target
setting or the imposition of cash limits (as open rational systems). It was perhaps
inevitable that these differing social constructions, reflected in different domains led
to differences between the process and content of budgeting.
Process and content of budgeting
In S, the budgeting process reflected a compromise between the parent company
targets and the single-customer volume and price negotiations. In T, the budgeting
process was top-down, with sales projections driving standard production costs
and profitability targets that had to meet short-term parent company targets. In P,
no representation of risk appeared in the budget, other than tacitly in the reserves
carried forward at year-end. The calculations presented to the police authority did,
however, reflect a consideration of the impact of various budgetary outcomes on the
number of police officers, and hence—albeit implicitly—on operational effectiveness.
Risk in the Process of Budgeting 291
Table 1
Domains of risk
Domain of risk S T P Q
Financial Dependent on
single customer
Investor support Pensions;
major incidents
Fund raising;
project bidding
Operational Systems
dependency
Maintaining/
increasing sales &
improving OTD
performance
Ability to cope
with demand
Absence of link
between finance
(resources) and
operations
(demand)
Incidents being
used for
blaming;
projects
unable to attract
funds; staff
unable to cope
with demands of#p#分頁標題#e#
the work
Political Difference between
customer-determined
volume and parent
company expectations
about volume and
return on sales
Internal politics to
‘massage’ SMC
Satisfying local
demand and
national priorities
simultaneously
Change in policy
and
programmes;
multi-agency
Working; future
changes in ideas
about the work
Personal None apparent Redundancy Promotion and
transfer (police)
Redundancy
(civilians)
Not coping with
issues; not
learning; not
having support
or containment
In Q, financial risk was considered in the planning process both nationally and at the
regional level, although the budgets were constructed on the basis of best estimates
of funding and expenditure.
An analysis of these cases suggests that the process of budgeting and the resulting
content of the budget may take any of the three forms identified earlier in this paper:
risk excluded, risk considered or risk modelled. Using this schema, the process of
budgeting in all four cases was characterized as risk considered, in which a top-down
budgeting process reflected negotiated targets. The content of budget documents
were risk excluded, being based on a set of single point estimates, in which all of
the significant risks were excluded from the budget itself. The one exception was P,
in which there was evidence of risk considered processes in the papers prepared for
the police authority for their decision about the precept. Risk consideration was also
implicit in the reserves maintained by P for unpredictable operational and pension
contingencies.
There was no evidence at the business unit level of any formal risk modelling and
an absence of discussion about any input–output relationships. No manager, in any
of the four cases, suggested any calculation or use of probabilities (as March and
Shapira, 1987 suggested). Some elements of risk modelling were possible—at least
implicitly—within S, but only because the nature of its single-customer business
with (±3%) volume certainty and the ability to pass most cost increases to the
customer provided the ability to predict, other than for premium freight, the financial
outcomes.
292 P.M. Collier and A. J. Berry
This suggests that budgeting as an open rational system is separate from the open
natural perspective implied by the perceptions of budget ‘participants’. There are
significant consequences for the management of risk in this separation as budgeting
did not appear, in any of the four cases, to be a tool used in risk management, nor
was risk evident in the risk excluded budgetary documents. This has implications for
where risk is held within organizations.#p#分頁標題#e#
Risk transfer
The four case studies also demonstrated how risk was shifted or transferred within
and between organizations. In S, risk shifted along the supply chain from the
customer as S assumed responsibility for ordering from suppliers and sequencing
the customer’s assembly line. However, the risk to S of volume changes and cost
increases was largely shifted back to the customer through contract negotiations
and the sharing of information. Similarly, systems and delivery risks were passed
to contractors who were located at S’ plant.
In T, risk transfer was evidenced between the sales and finance departments as the
SMC came under pressure to be ‘massaged’ so that ‘product costs’ did not impede
the important task of winning new sales. However, finance retained responsibility
for explaining variances between actual and standard cost. The investors also shifted
risk to employees as their exit strategy drove the need for restructuring and job losses,
despite strong profit performance.
In P, risk transfer was evident between government and the police force as
formula-based allocations to forces resulted in cash-limited budgets that reduced
government’s exposure, while an unfunded pensions liability was passed to the
police force. This risk was in turn passed to the police authority that in turn passed
the increased financial burden onto council taxpayers. Within P, the scheme of local
financial management passed some of the risk of meeting local policing needs to local
police commanders, while giving them the flexibility to use their devolved budget to
deliver policing in a locally relevant way.
In Qrisk transfer was evident between society, State agencies and Qas it bid for and
provided projects in line with emerging policy. One significant element of this process
was the upside value of risk in providing illuminating opportunities for learning.
However, the ‘social fantasy’ of charitable bodies being held responsible for solving
wider societal problems is an example of risk transfer from the public to social care
agencies.
Risk transfer may be an open natural response to avoid the problems inherent in the
separation of a risk-excluded budgeting content from a risk-considered budgeting
process. This response may be a reaction to the lack of participation by budget
‘participants’ in the imposition of top-down budgets and the risk perceptions held
by managers with which they were forced to cope.
Risk containment
The containment of risk was mostly carried out in the private sector organizations
by procedures of technical rationality. In S, this was achieved by close working
relationships between the assembler and sequencer, and the sequencer’s technology
and logistics partners. In T, this was achieved by closely guarding financial information#p#分頁標題#e#
from employees who may have used it to argue against downsizing and
redundancies.
Risk in the Process of Budgeting 293
In the public and voluntary sector organizations there was significant evidence of
containment by senior and middle managers. While the ability to maintain reserves
was a method by which risk was contained in P, there was an implicit reliance on the
multi-skilling and flexibility of individual police officers and their coping ability to
deal with serious crimes and major incidents that might occur. An example after the
research visit took place was the serious flooding that affected P’s area.
The evidence fromQsuggested that managers may do more than exclude, consider,
model or transfer risk. They may instead actually hold it themselves and provide
containment for the anxieties of others as they in turn deal with difficult problems
of risk and uncertainty. The act of containing risk makes a significant contribution
to an organization’s capacity to work in the face of ambiguity and uncertainty.
This raises the interesting question as to the degree to which individuals can
take this role, especially in relation to the structures and procedures that exist in
organizations. If risk is inevitable, and cannot be managed away or transferred, then
managers must find means of containing—and handling—it. Some of the evidence
from P, particularly in the coping behaviour of police managers in the face of
uncertain demand may also be viewed as that of senior managers containing risk
and working with it. There is an interesting literature of human responses to anxiety
in organizations (see, for example, Kets de Vries, 1989).
Risk containment suggests an important role for managers (as identified by
Thompson, 1967) to mediate between the rational and natural perspectives. It seems
that by ‘containing’ risk, managers may provide a buffer between the rational and
the natural. This is a kind of risk transfer from the organization to its managers who
are accountable for both budgetary performance and for risk management. It implies
a significant personal burden on managers.
Limitations of the study
There were limitations in the methods adopted although the case method was
sufficient for an exploratory study of the subject. However, this approach proved
difficult to penetrate the assumptions that had been tacitly adopted in budgeting
processes over time. In these processes, many issues of risk and uncertainty had been
handled, but these were inaccessible to the researchers.
The second limitation was due to the level of analysis being organizational units
rather than a corporate body (even in the case of P, autonomy was limited by the
influence of government). A more holistic approach would need to consider the
role of financial markets, the interface between corporate policy and business units,#p#分頁標題#e#
and the transformation processes for products and services in relationships with
suppliers and customers. In the public and not-for-profit sector, this would also mean
a deeper consideration of funding and social policy linked to the ethos, norms, values
and beliefs of workers.
Consequently, there are opportunities for further research to explore these areas
through longitudinal studies.
5. Conclusions
In this paper risk has been defined as both process and outcome. As process, risk is
the consideration of unpredictable and uncontrollable events and perceptions about
294 P.M. Collier and A. J. Berry
those events. As outcome, risk is the consequences (both fortuitous and hazardous)
of those events. While the budgeting process is concerned with the consideration of
risk, the budget document is concerned with consequences.
Despite the wide-ranging literature in relation to budgeting, there is little evidence
in the accounting literature of research into risk in the process of budgeting other than
with those technical rational processes associated with capital budgeting. There has,
however, been a call in the strategy literature for a better understanding of risk preference
(March and Shapira, 1987) and risk perception (Bettis and Thomas, 1990) in
order to develop conceptions of risk that are managerially relevant (Ruefli et al., 1999).
A social constructionist approach by which risk is socially (Beck, 1986, 1992
in translation) or culturally constructed (Douglas and Wildavsky, 1983) has been
adopted in this paper. Such an approach adds to the accounting literature that sees
budgeting as intertwined with symbol (Czarniawska-Joerges and Jacobsson, 1989),
hypocrisy (Brunsson, 1989) and providing significance and meaning (Preston et al.,
1992; Boland, 1993).
In categorizing different perspectives of risk in budgeting, this research has utilised
the open/closed, rational/natural systems typology of Scott (1998) as proposed by
Otley et al. (1995). The research was designed as an exploratory and comparative
study that involved four different organizations of widely different types. During
the course of the research, each organization became associated with a mini-crisis
that involved issues of risk.
The findings from the four case studies reveal differences based on the contexts
of unique circumstances, histories and technologies of the organizations. The four
cases illustrate how the different social constructions of participants in the budgeting
process influenced the domains—or alternative lenses—through which the process of
budgeting took place and how the content of the budget was determined. However,
the research identified similarities that can be summarized as follows:
First, four domains of risk were observed, reflecting the different social constructions#p#分頁標題#e#
of participants: financial, operational, political and personal. These domains
were important in understanding how risk was perceived in the budgeting process.
Risk was perceived in each of these domains, an open natural process, but these perceptions
were isolated from the open rational budgeting process that was dominated
by top-down targets or cash limits.
Second, the process and content of budgeting was categorized as risk modelled,
risk considered or risk excluded. There was little direct evidence of risk modelling
in the four cases and some evidence of risk consideration in one case. The process of
budgeting in all four cases was characterized as risk considered, in which a top-down
budgeting process reflected negotiated targets. The content of budget documents was
risk excluded, being based on a set of single point estimates, in which all of the
significant risks were excluded from the budget itself. The separation of budgeting
and risk management has significant consequences for the management of risk as the
process of budgeting needs to be considered separately from the content of budget
documents. This has implications for where risk is held within organizations.
Third, risk transfer took place between and within organizations. Risk transfer may
be an open natural response to avoid the problems implied by the separation of a
budget, that is risk excluded budget, from a budgeting process that is risk considered.
This response may be a reaction to the lack of ‘participation’ by budget holders and
the risk perceptions held by managers.
Risk in the Process of Budgeting 295
Fourth, managers ‘held’ risk and provided containment for the anxiety of others.
Risk containment is a particular type of risk transfer, from the organization to its managers,
which takes place in relation to residual risks, not contained within the budget.
Risk containment may be a buffer used by managers to mediate between the rational
and natural perspectives. It implies a significant personal burden on managers.
This research suggests that there is a relationship between the social construction of
budget participants at different levels of analysis that impacts the budgeting process.
In particular, the process of budgeting, by excluding some risks and considering
others, is seen to be different to, and needs to be interpreted separately from the
content of the budget. This observation is particularly relevant in observing how risk
can shift between and within organizations and how managers may themselves hold
risk on behalf of the organization.
These issues are of particular importance given the requirements of the Turnbull
report (Institute of Chartered Accountants in England andWales, 1999) for Boards of
Directors to identify, evaluate and manage significant risks in their organizations. If,#p#分頁標題#e#
as the four cases suggest, the most significant risks may be excluded from financial
reports, the requirements of governance and internal control suggest that the budget
as process needs to be better understood.
The Discussion section of this paper outlined the limitations of these exploratory
studies. Therefore, it is evident that there is a need for further studies of risk to enable
a better understanding of both the role and limitations of management accounting
and of the implications that broader conceptions of risk (including risk transfer and
containment) have for the governance of organizations.
Acknowledgements
We acknowledge the financial support provided for this research by the Chartered
Institute of Management Accountants. Earlier versions of this paper were presented
to Management Accounting Research Group conferences at the London School of
Economics in April 2000 and at Aston University in September 2000. Thanks are also
due to two anonymous referees for their suggestions. However, the paper remains
the responsibility of the authors.
References
Adams, J., 1995. Risk, London, UCL Press.
Arnold, G. and Hatzopoulos, P. D., 2000. The theory practice gap in capital budgeting:
evidence from the United Kingdom, Journal of Business Finance and Accounting, June/July.
Baird, I. S. and Thomas, H., 1990. What is risk anyway? Using and measuring risk in strategic
management, in R. A. Bettis, H. Thomas (eds), Risk, Strategy, and Management, Greenwich,
CT, JAI Press.
Beck, U., 1986. Risk Society, London, Sage (1992 in translation).
Berry, A. J., Broadbent, J. and Otley, D., 1995. Procedures for control, in A. J. Berry,
J. Broadbent, D. Otley (eds), Management Control: Theories, Issues and Practices, Basingstoke,
Hants, Macmillan.
Berry, A. J. and Otley, D. T., 1975. The aggregation of estimates in hierarchical organizations,
Journal of Management Studies, May, 175–193.
Bettis, R. A. and Thomas, H., 1990. Risk, Strategy, and Management, Greenwich, CT, JAI Press.
296 P.M. Collier and A. J. Berry
Birnberg, J. G., Turopolec, L. and Young, S. M., 1983. The organizational context of accounting,
Accounting, Organizations and Society, 8, 111–129.
Boland, R. J. J., 1993. Accounting and the interpretive act, Accounting, Organizations and Society,
18, 125–146.
Brunsson, N., 1989. The Organization of Hypocrisy: Talk, Decisions and Actions in Organizations,
London, John Wiley & Sons.
Buckley, A. and McKenna, E., 1972. Budgetary control and business behaviour, Accounting and
Business Research, Spring, 137–50.
Covaleski, M. A. and Dirsmith, M. W., 1988. The use of budgetary symbols in the political
arena: an historically informed field study, Accounting, Organizations and Society, 13, 1–24.
Czarniawska-Joerges, B. and Jacobsson, B., 1989. Budget in a cold climate, Accounting,#p#分頁標題#e#
Organizations and Society, 14, 29–39.
Douglas, M. andWildavsky, A., 1983. Risk and Culture: An Essay on the Selection of Technological
and Environmental Dangers, Berkeley, CA, University of California Press.
Drury, C., 2000. Management and Cost Accounting, 5th edition, London, Thomson Learning.
Ezzamel, M., 1994. Organizational change and accounting: understanding the budgeting
system in its organizational context, Organization Studies, 15, 213–240.
Govindarajan, V., 1984. Appropriateness of accounting data in performance evaluation: an
empirical examination of environmental uncertainty as an intervening variable, Accounting,
Organizations and Society, 125–135.
Harris, E. P., 1999. Project risk assessment: a European field study, British Accounting Review,
31, 347–371.
Harris, E. P., 2000. Strategic investment decision-making: managerial judgement on project risk
and return, Journal of Applied Accounting Research, 5, 87–110.
Hofstede, G., 1968. The Game of Budget Control, London, Tavistock.
Hopwood, A. G., 1973. An Accounting System and Managerial Behaviour, London, Saxon House.
Institute of Chartered Accountants in England and Wales, 1999. Internal Control: Guidance for
Directors on the Combined Code (Turnbull Report).
International Federation of Accountants, 1999. Enhancing ShareholderWealth by Better Managing
Business Risk (International Management Accounting Study No. 9).
Kets de Vries, M., 1989. The Neurotic Organization, San Francisco, CA, Jossey-Bass.
Llewellyn, S., 1998. Pushing budgets down the line: ascribing financial responsibility in the
U.K. social services, Accounting Auditing and Accountability Journal, 11, 292–308.
Lowe, E. A. and Shaw, R. W., 1968. An analysis of managerial biasing: evidence from
a company’s budgeting process, Journal of Management Studies, October, 304–315.
March, J. G. and Shapira, Z., 1987. Managerial perspectives on risk and risk taking, Management
Science, 33, 1404–1418.
Merchant, K. A. and Manzoni, J.-F., 1989. The achievability of budget targets in profit centers:
a field study, The Accounting Review, LXIV, 539–558.
Otley, D. T., Berry, A. J. and Broadbent, J., 1995. Research in management control: an overview
of its development, British Journal of Management, 6, 531–544.
Preston, A., 1995. Budgeting, creativity and culture, in D. Ashton, T. Hopper, R. W. Scapens
(eds), Issues in Management Accounting, 2nd edition, London, Prentice Hall.
Preston, A. M., Cooper, D. J. and Coombs, R. W., 1992. Fabricating budgets: a study
of the production of management budgeting in the national health service, Accounting,
Organizations and Society, 17, 561–593.
Ruefli, T. W., Collins, J. M. and Lacugna, J. R., 1999. Risk measures in strategic management:
Auld Lang Syne? Strategic Management Journal, 20, 167–194.#p#分頁標題#e#
Schiff, M. and Lewin, A. Y., 1968. Where traditional budgeting fails, Financial Executive, May,
57–62.
Scott, W. R., 1998. Organizations: Rational, Natural, and Open Systems, 4th edition London,
http://www.ttc-switch.com/thesis_sample/Australia_Thesis_sample/Prentice Hall International, Inc.
Risk in the Process of Budgeting 297
Thompson, J., 1967. Organizations in Action: Social Science Bases of Administrative Theory, New
York, McGraw-Hill.
Wallander, J., 1999. Budgeting—an unnecessary evil, Scandinavian Journal of Management, 15,
405–421.
Yin, R. K., 1994. Case Study Research: Design and Methods, 2nd edition, Thousand Oaks, IL, Sage
Publications.