Journal of Human Resource Costing & Accounting

Accountability in local government: a principal-agent perspective

Caroline Nyman Fredrik Nilsson Birger Rapp
Article information:
To cite this document:
Caroline Nyman Fredrik Nilsson Birger Rapp, (2005),"Accountability in local government: a principal-agent
perspective", Journal of Human Resource Costing & Accounting, Vol. 9 Iss 2 pp. 123 - 137
Permanent link to this document:
http://dx.doi.org/10.1108/14013380510645397
Downloaded on: 13 May 2015, At: 02:16 (PT)
Downloaded by FREIE UNIVERSITAT BERLIN At 02:16 13 May 2015 (PT)

References: this document contains references to 21 other documents.

To copy this document: [email protected]
The fulltext of this document has been downloaded 1119 times since 2006*
Users who downloaded this article also downloaded:
Louise Kloot, (1999),"Performance measurement and accountability in Victorian local
government", International Journal of Public Sector Management, Vol. 12 Iss 7 pp. 565-584 http://
dx.doi.org/10.1108/09513559910308039
Ioannis A. Kaskarelis, (2010),"The principal-agent problem in economics and in politics", Humanomics, Vol.
26 Iss 4 pp. 259-263 http://dx.doi.org/10.1108/08288661011090866
Kevin Baird, Herbert Schoch, Qi (James) Chen, (2012),"Performance management system effectiveness
in Australian local government", Pacific Accounting Review, Vol. 24 Iss 2 pp. 161-185 http://
dx.doi.org/10.1108/01140581211258461

Access to this document was granted through an Emerald subscription provided by 514603 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for
Authors service information about how to choose which publication to write for and submission guidelines
are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as
providing an extensive range of online products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee
on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive
preservation.

*Related content and download information correct at time of download.

 The Emerald Research Register for this journal is available at The current issue and full text archive of this journal is available at
www.emeraldinsight.com/researchregister www.emeraldinsight.com/1401-338X.htm

Accountability in
Accountability in local local government
government: a principal-agent
perspective
123
Caroline Nyman, Fredrik Nilsson and Birger Rapp
Department of Computer Science, Linköping University, Linköping, Sweden

Abstract
Purpose – In Sweden there is an ongoing debate on ways to enhance accountability and
Downloaded by FREIE UNIVERSITAT BERLIN At 02:16 13 May 2015 (PT)

transparency in order to avoid future scandals both in private companies and local government. Until
now the debate has had clear political overtones, and there are no generally accepted methods for
analysis of this issue. The purpose of the article is to survey and analyze the chain of accountability to
be reviewed by the auditors and through the use of principal-agent (PA) theory to identify the
problems that may arise in holding persons accountable.
Design/methodology/approach – The paper shows how PA theory can be used to structure and
analyse the intricate situation in Swedish local government. The model is particularly suitable in
complex systems of accountability. Therefore, the paper explores two cases where the outcomes were
discharge from liability. Based on the two case studies, the paper extends the empirical investigation
to a list of 63 qualified audit reports in local government for 2002 and 2003.
Findings – PA theory facilitated the understanding and structuring of the complex accountability
situation and helped to identify important weaknesses in the system. The analysis underscored that
neither better accounting nor better auditing systems alone will resolve the dilemma in improving
accountability. The link between auditors and elected officials and civil servants needs to be
investigated and clarified. Among other things, this means that the focus of today’s debate should shift
toward the lack of clear responsibilities within local governments.
Research limitations/implications – This paper explores an area of research where few studies
have previously been conducted – therefore, the paper is to some extent exploratory. In the future the
number of empirical examples could be increased, and the different problems in the system of
accountability in local government could be discussed in more detail.
Practical implications – This research paper is based on an empirical investigation – as far as we
know no such study has previously been performed in Sweden The principle of public access to
information in Sweden provides unique opportunities for openly studying the work of auditors in the
system.
Originality/value – The research shows that PA theory may be used to structure complex issues of
accountability. In addition, the study recommends a new focus for the debate on accountability.
Keywords Management accountability, Local government, Sweden
Paper type Research paper

Introduction
In recent years, we have read about many companies and other organizations that have
been creative, sometimes even manipulative, in accounting for their operations. For
example, the Skandia affair, widely publicized in Sweden, has highlighted the need for Journal of Human Resource Costing &
greater control and transparency in business. For public authorities as well, the Accounting
Vol. 9 No. 2, 2005
importance of providing more control and transparency has been brought to the fore. pp. 123-137
q Emerald Group Publishing Limited
Olsson (2003) has raised the appropriate question of whether better accounting 1401-338X
systems and more sophisticated auditing could prevent future scandals. DOI 10.1108/14013380510645397
 JHRC In Sweden, a government study has proposed a code of corporate management, and
a similar code is also being drafted in the public sector. Together with well-functioning
9,2 audits, such codes are important instruments for increasing transparency. In major
companies – as in municipal, county, and regional government – there are several
links in the chain of delegation, and when so-called “soft” values are reviewed, there is
a substantial risk that information will be concealed, both intentionally and
124 unintentionally (Bontis, 2003). In Sweden, local government auditors are charged with
examining not only finances but also performance in regard to the “soft” values of
appropriateness, effectiveness and productivity. The study of local government
auditing and its role in the review of such values is an interesting area of research.
The point of departure for this article is that local government auditing, like
traditional corporate auditing, is intended to reduce the asymmetry of information
between principal and agent. In this sense, the audit can be considered a signal to the
Downloaded by FREIE UNIVERSITAT BERLIN At 02:16 13 May 2015 (PT)

principal in the principal-agent chain (Yandell, 1988). For this signal from the auditors to
be reliable, a thorough understanding of the system of accountability and of the flaws in
the system is required. The purpose of the article is to survey and analyze the chain of
accountability to be reviewed by the auditors and through the use of principal-agent
(PA) theory to identify the problems that may arise in holding persons accountable.
While the analysis concerns conditions in county council districts and municipalities in
Sweden, in our view the findings are also applicable to the corporate world.

Previous research
The use of principal-agent (PA) theory has previously proved fruitful in studies of
auditing in private firms (see, for example, Power, 1997; Baiman et al., 1987; Yandell,
1988). Several of these studies use that theory in analyzing the role of the auditor as an
agent in relation to a principal. Some international groups of researchers have
highlighted how auditing in a public-sector environment can help to reduce the
asymmetry between principal and agent in regard to information (see, for example,
Laughlin, 1996; Payne and Jensen, 2002). Other researchers have used PA theory to
analyze accountability and delegation in central government (see, for example, Strøm,
2000; Müller, 2000; Lupia and McCubbins, 2000; Venton, 1997). These scholars,
however, have not studied the role of auditing in central government as a source of
information in the system of accountability, but have focused on issues of delegation
and democracy. They also provide valuable contributions concerning the importance
of the contract and mechanisms of selection in principal-agent relationships.
Yandell (1988) has used PA theory to examine auditing as one element in the system
of signals to principals. In his article, he assumes a relationship involving one principal
and one agent. In our article, the audit is also regarded as a signal to the principal, but
we assume a situation where there are several principal-agent relationships. This
complexity in the chain of accountability makes it possible to conceal information from
the principal. The audit can therefore help to improve transparency and to provide the
principal with valuable information on the agent’s performance.
In summary, there are few studies – to our knowledge – where PA theory is applied
to auditing in municipalities and county council districts. Swedish research in the field
of local government auditing is not very extensive and is devoted primarily to issues of
democracy and law (see Ahlbäck, 1999; Lundin, 1999; Sveriges Kommuner och
Landsting, 2005). Few groups of researchers studying local government auditing have
focused on auditing as a tool for providing information. Thus, there is presently a
shortage of empirical studies and analyses of the effects of information asymmetries in
an environment like that of municipalities and county council districts. Moreover, there
 is little knowledge about what auditing could contribute to the identification of Accountability in
problems that may arise in a complex accountability relationship (with many
principal-agent relationships). For example, what does it mean to have both local government
management by civil servants and management by elected officials? The contribution
of this article is that it identifies and analyzes these types of complex accountability
relationships and the problems that they cause.
125
Local government auditing and corporate auditing
In Sweden, local government auditing is based on a system where auditors are
politically appointed by the municipal or county council for a period of four years.
These appointed auditors issue an audit report to the council, which thereafter decides
on discharging boards and committees from liability (Local Government Act, chapters
5:25a-b and 5:31). As the council’s instrument of review, the auditors’ independence in
Downloaded by FREIE UNIVERSITAT BERLIN At 02:16 13 May 2015 (PT)

relation to boards, committees and drafting committees is critical. Local government

auditing is similar in design to the organization of municipalities, county council
districts and regions, in that elected officials are responsible for operations and have
civil servants to assist them. Unlike elected members of boards and committees,
however, auditors are empowered by law to choose the experts that will assist them
(Local Government Act, chapters 9:8).
The assignment of appointed auditors in local government is broader than that of
the certified accountants who audit corporations. Thus, auditing of local government
companies also involves non-certified auditors (often appointed from among local
government auditors), who review operations from the standpoint of the owners’
directives and of the mission given to the company by the local government. The duties
of local government auditors are set forth in the Local Government Act (chapter 9:9),
which provides among other things that their examination is to be conducted in
accordance with Good Auditing Practice. In this respect, the situation of local
government auditors is the same as that of the certified accountants who audit
corporations. But there are also many differences. The Local Government Act provides
that the auditors are to examine whether operations are conducted in accordance with
their purpose, whether they are conducted in a financially satisfactory manner,
whether the accounts provide a fair financial picture and whether the internal control
for which the boards are responsible is adequate.
The duties of certified accountants are governed primarily by the Companies Act.
This Act provides that the auditors are to review the administration of the managing
director and the board of directors and to determine whether the accounts provide a
fair financial picture of the company. These provisions permit broad interpretation of
the appropriate scope of the administrative audit. In practice, certified accountants
focus their review more on financial accounting than do the appointed auditors in local
government. One possible reason is the difference in the aims of the organizations to be
examined. To put it simply, the aim of corporations is to earn money and to provide a
return to the shareholders, and the purpose of local government operations is to use tax
revenue in an optimal manner in conducting operations for the public good. In regard
to the financial audit, local government auditing can gain useful knowledge from
corporate auditing; in regard to examination on the basis of “soft” values, corporate
auditing can learn much of value from local government auditing.

The system of accountability

In a municipality, the citizens elect councillors to represent them between elections: this
is a principal-agent relationship where the council is the agent of the citizens, who are
 JHRC the principals. In the next stage, the council becomes a principal when it appoints
members of boards and committees, who thereby become agents of the council. Under
9,2 the Local Government Act, the council is obligated to hold accountable those required
to report to them, i.e. the municipal executive board and the various other boards.
These are to be held accountable for what they have done or failed to do on the basis of
the mission and objectives that they have received from the council. The council is
126 obligated to formulate the mission of the boards in statements of general aims and
plans. In addition, the council prepares guidelines and delegates power for its term of
office – at the same time, the council is fully free to deny discharge from liability and to
dismiss any person or persons who have not performed their duties satisfactorily. The
council is responsible to the citizens for requiring an accounting for operations as well
as for finances from those obligated to render such an accounting (Svenska
Kommunförbundet, 2004).
Downloaded by FREIE UNIVERSITAT BERLIN At 02:16 13 May 2015 (PT)

The preparation and carrying out of decisions is performed to a substantial extent

by civil servants, although the final word formally rests with elected officials. If
accounting for operations and finances in a democratic system is to function properly,
those in power/decision makers must be sure that their decisions are based on
comprehensive and accurate documentation and are carried out so that the intended
effects are achieved. In this connection, it should be noted that civil servants in local
government, even at a very senior level, are not required to render an accounting and
are therefore not answerable in the audit on accountability.
The local government auditors are charged on behalf of the local government
council with reviewing and evaluating the liability of those who are accountable. The
auditors thus fulfil their mission from the council, which finances the audit and issues
the rules that govern it. The system is based on the auditors’ lack of power to hold any
agents accountable; rather, their duty is to present their opinion to the council on the
question of accountability. The decision on accountability is then taken by the
members of the council, who may not always be impartial in such decisions. Since
council members frequently sit on boards and committees as well, there is a danger
that their loyalty to members of the same political party may outweigh their loyalty to
the citizens. The citizens, in turn, hold officials accountable in general elections every
four years. In these elections, however, these officials are not accountable individually,
as votes are totalled at the party level. On the other hand, improper behaviour by one or
two elected officials may lead to fewer votes for their party as a whole, which will then
win fewer seats on the local government council.
In the opinion of most people, an agent who fails in the performance of her/his
mission should also be held accountable. In a corporation, the decision on
accountability is taken by general meeting of shareholders, based on the auditors’
recommendation; in local government, too, the council decides on accountability after
obtaining the opinion of the auditors on this question. The difference is that in a
corporation the decision makers have a direct interest in the determination of
accountability, whereas in local government council members have an indirect interest
in the matter – it is the citizens who have a direct interest. Since the political parties in
power form the majority on the council, the determination of the issue of accountability
will naturally be the subject of partisan political discussion (Nyman et al., 2005).
According to chapter 3 of the Local Government Act, local governments must have
a council and an executive board (kommunstyrelse), a board of elections (valnämnd ), as
well as a supervisory board (myndighetsnämnd ), which, however, may be a committee
of the executive board. Often local governments are organized with one or more boards
appointed by the council. According to the Local Government Act, local governments
 may also include drafting committees (beredningar) in their organization. These bodies Accountability in
are often charged with analyzing external developments and formulating goals for the
future. The function of drafting committees may be most accurately regarded as
local government
relieving the executive board and other boards of some of their workload and enabling
the council to be more active in municipality or county government. All of the boards,
including the executive board, answer directly to the local government council.
According to the Local Government Act (chapter 6:1), the local government executive 127
board is charged with managing and co-ordinating local government affairs and
supervising the operations of the other boards. As a practical matter, the executive
board may serve as a general body that governs the entire municipality or county
while also leading/co-ordinating the work of the other boards for the council. The chain
of accountability is illustrated in Figure 1, where the black arrows indicate by whom
(the principal) and to whom (the agent) the mission is given, whereas the dashed line
Downloaded by FREIE UNIVERSITAT BERLIN At 02:16 13 May 2015 (PT)

shows to whom the agent is responsible. As shown in Figure 1, there are several
principals and several agents in the system of accountability, and thus multiple
principal-agent relationships. The basic assumption, in accordance with the Local
Government Act, is that there is a primary principal (the citizen) and a primary agent
(the council) in all local governments. The double arrow between the executive board
and the other boards/committees indicates the co-ordinating function of the executive
board and its obligation to supervise the other boards. As a consequence of this
obligation to provide co-ordination and supervision, the executive board can serve as a
superior board in local government. Thus, in certain local governments the executive
board may in practice function as a principal in relation to the other boards, which in
turn function as agents. According to the Local Government Act, the executive board is
not superior to the other boards, but as a practical matter it often holds a stronger
position relative to them.

The chain of accountability

In a situation of only one principal and one agent, the principal can control the agent’s
behaviour through actions, measurements and follow-up of the assignment (Laughlin,
1996). This means that by their behaviour both actors may influence each other so that
the assignment is performed satisfactorily. Social interplay and control would thus

Figure 1.
The system of
accountability
 JHRC function as a mechanism of review in itself. The basic values and the conscience of the
agent, as well as the relationship to the principal, would then serve as a control
9,2 mechanism assuring that the assignment was being performed satisfactorily. Thus,
through the relationship the principal has “committed” the agent to acting in a desired
manner. This would be particularly clear in situations where the actors have recurring
contact over longer periods – in other words, in a situation where “repurchasing” is
128 important, openness is also important. The business relationship then begins to
resemble a relationship between friends (Behn, 2001). In situations with complex
accountability systems, where there are numerous links involving principals and
agents, the relationship alone does not suffice as a control mechanism. The reason for
this is that it becomes too difficult to obtain an overview of the system of
accountability, thus making it possible to conceal information from the principal in
furtherance of the agent’s interests (Laughlin, 1996; Tarschys, 2002).
Downloaded by FREIE UNIVERSITAT BERLIN At 02:16 13 May 2015 (PT)

As shown in Figure 1, the system of accountability in local government is multiple.

The system of signals in this delegated chain of accountability is therefore complex.
The duty of the auditors on behalf of the primary principal is to examine, clarify and
expose the flaws in this chain of accountability. It should be obvious that doing so is no
easy matter, and that the risks of error are considerable. When several principals and
agents are involved, there is also a danger that shortcomings in the system of signals
will lead to serious consequences for the organization (Rapp and Thorstensson, 1994).
These consequences are described in greater detail later in this article. The chain of
accountability which the auditors are charged with reviewing consists of a number of
agency relationships as well as relationships involving several principals. Some of
these are described in Figure 2.
The chain of accountability in local government is divided into the organization of
civil servants (P4/A3a and A3b/A4 in Figure 2) and the organization of elected officials
(P2/A1, P3a/A2a and P3b/A2b in Figure 2. As mentioned previously, civil servants are
not subject to audit and are thus outside the scope of the auditor’s review of
accountability, but they nevertheless affect what happens in the chain of accountability
by carrying out and preparing political decisions for elected officials and by serving
those officials. Figure 2 illustrates this situation with a dashed line; the organization of
civil servants is below the line and the organization of elected officials is above it. As
noted earlier, the auditor’s opinion on discharge from liability applies only to local

Figure 2.
The system of
accountability from a
principal-agent
perspective
 government officials who are required to present accounts. This situation leads to Accountability in
interesting relationships in the chain of accountability.
The office of local government often serves as the highest level of the civil servant local government
organization. The municipal, county or regional director is then the highest-ranking
civil servant, who prepares decisions and acts as the link from the executive board and
the council to operations. Sometimes this director also occupies a position as the
superior of the other administrative directors. This means that the local government 129
administration is delegated additional power from the executive board and is the agent
(A3a) in this link. Thereafter, it functions as a principal (P4) in that it delegates
assignments further to the administrative staff of various other boards (A3b). The
administrative staff of a board is subordinate to the board and is thus its agent (A4). In
total, this means that the administrative staff of a board can have two principals and
potentially two conflicting sets of objectives to consider.
Downloaded by FREIE UNIVERSITAT BERLIN At 02:16 13 May 2015 (PT)

The local government leadership may have made a decision that the administrative
director must follow as a subordinate of the director of local government, who in turn is
subordinate to the local government executive board. A board may then reach a
decision that is inconsistent with that of the local government executive board. The
administrative director of the first board then faces a conflict between the decision of
the local government administration and the political decision of her/his board.
Another situation may be one where the local government executive board, for
example, prepares the budget and changes the budget proposals of the other boards
before they are submitted to the council for approval.

The auditors and the system of accountability

The traditional audit review is focused on the financial accounts and their conformity
to law. We customarily refer to accounts that provide a fair financial picture. In a
municipality or county council district, which does not operate for a profit, there is an
additional component: namely, how operations are conducted, i.e. their effectiveness
and appropriateness to their purpose. The former may be termed “tangible” values and
the latter “intangible” values.
It is a classic truth that what is measured in an operation is also what is performed.
This means that the system of rewards is of central significance. If local government is
to be operated and reviewed in accordance with democratic principles, the system of
accountability as a whole must also be designed to permit control and transparency.
Problems arise when the agent does not choose to act in the service of both the
principal’s and the agent’s own interests. This can happen in a situation where
information is asymmetric and there is little risk of detection. It might also occur in
systems that are so complex that it is difficult to determine who is the primary
principal and what is the purpose of the operation for the primary principal. When the
intentions of the council with regard to the operation are not clearly specified or are
defined too broadly, there is a danger than the agent will formulate her/his own
mission, which may not necessarily be the one intended by the primary principal.
There is also a risk that the agent may act defensively by performing what will be
rewarded and measured. In such a case, an agent further down the chain of
accountability may act in a manner incompatible with the best interests of the whole.
In our material, we can see tendencies toward such behaviour, where audit reports
state that boards have consistently neglected to manage their operations within the
budget limitations set by the council. In these cases the boards have probably judged
that they can run their operations at a deficit and still get by. In our material, there is
much to indicate that this judgment is correct, as the council very rarely denies
 JHRC discharge from liability for failure to conduct operations according to the objectives
and guidelines set by the council. For 2002, two boards were required to resign for such
9,2 reasons, but in these cases the election had produced a new majority. That same year a
local board of engineering was denied discharge, but at this time that board was
abolished. Also, in 2003, discharge was denied to a board that ceased to exist, and in
yet another case that year a board failed to observe the rules of delegation. Another
reason could be that a board has formulated its own objectives or that the objectives of
130 the council have been in focus but have been allowed to take precedence over the
budget targets of the council. A more thorough study of the relationship between
council and board is needed to verify hypotheses of this kind.

Controlling the agent

Basically, there are two alternative ways for a principal to control an agent. One is to
Downloaded by FREIE UNIVERSITAT BERLIN At 02:16 13 May 2015 (PT)

invest in an information system to study the behaviour of the agent. The other is to
reach a result-oriented agreement with the agent, i.e. to make the agent’s compensation
dependent on the results of the operation. However, with a result-oriented contract, the
cost of control to the principal may be very high. If the principal designs a system that
does not affect the agent, and if the agent does something that the principal had not
anticipated, a situation that could benefit both parties will not be achieved. Achieving
that situation is often rendered more difficult by the fact that the agent has received the
assignment because of her/his unique competence. Thus, the agent has knowledge that
the principal lacks. Such a relationship leads to asymmetry of information that can
allow the agent to deceive the principal. The principal must therefore design the
system so that the common objective is fulfilled. In local government, this means that
the council must take responsibility for providing clear objectives and guidelines for
operations. If the council does not do this, there is a risk that the mission will be unclear
and that the board itself or the executive board will take over the role of the council. In
a relationship, costs of control arise because the parties do not have perfect information
about each other – in other words, there is asymmetry of information (Rapp and
Thorstensson, 1994).
If systems of internal control are lacking, and if the council does not formulate clear
missions for boards and committees, both the costs of incentives and the costs of
discrepancies are likely to increase. If the council does not assume its civic
responsibility to formulate objectives and missions for boards and committees, some
other body will take over this duty. The natural consequence will then be that the local
government executive board, which lacks such a legal mandate, assumes the duty.
Moreover, if the council “abdicates”, it will be difficult to impose accountability.
Basically, it is essential that the mission is clear and that properly functioning control
and management are present in local government, and that there are established
procedures for follow-up and control of operations. The system of internal control
should function both directly and indirectly as a signal to the principal concerning the
behaviour of the agent. In municipalities and county council districts, the system of
internal control should include procedures for attestation, rules for delegation, systems
for reporting discrepancies, etc. According to the Local Government Act, the auditors
have a duty to perform a special examination of the system of internal control.

Method
From the discussion so far, it is apparent that the complex system of accountability
results in a lack of clarity that may create problems in the determination of
accountability. One of the main reasons is that information can be concealed and that it
 is often not clear who the principal really is. When the local government council Accountability in
neglects its responsibility to set clear objectives and missions for boards and
committees, the determination of accountability becomes more difficult. This is also the local government
case when council members have other loyalties to consider besides loyalty to the
primary principal – the citizen.
We studied the determination of accountability in two cases, one from 2002 and one
from 2003 (see also Nyman et al., 2005). The cases were chosen because they led to a
recommendation by the auditors against granting discharge from liability, with the
131
council in both cases following that recommendation. In Swedish municipalities,
county council districts and regions, discharge for 2002 was denied in three different
cases: one of these is our first case (a county council district). For 2003, discharge was
also denied in three different cases, one of which is our second case (a municipality).
We interviewed the council chairman, the chairman and vice chairman of the auditors
Downloaded by FREIE UNIVERSITAT BERLIN At 02:16 13 May 2015 (PT)

at the time, and the chief auditor in the county council district. We interviewed the
council chairman, one council vice chairman, the chairman and vice chairman of the
auditors, and the chief auditor in the municipality. In addition, we read and analyzed
minutes of meetings, memoranda, notes and other relevant documentation. Both cases
involved an extended period, and thus relevant material covering a considerable time
in the past was also studied for the purpose of providing an understanding of what had
happened.
As a result of the case studies, we decided to conduct a broader study, where we
read and analyzed “unclean” audit opinions for 2002 and 2003. We used information
from the “accountability bank” of the Swedish Association of Local Authorities
(Sveriges Kommuner och Landsting) on the internet (see www.skl.se/revision), where
all “unclean” audit opinions are stored. An “unclean” audit opinion is one where the
auditors present criticism and/or recommend denial of discharge from liability. When
reviewing the “accountability bank”, we discovered additional cases that were not
included in the bank. Thus, our empirical material is richer than the material in the
“accountability bank”. However, since we have not read the audit reports for all 290
municipalities, 18 county council districts and two regions for the two years, we do not
know whether we succeeded in capturing all “unclean” audit reports.
Our review is based on audit reports, other reports on audits conducted and on the
minutes of council meetings. Good Auditing Practice, according to the Swedish
Association of Local Authorities (Svenska Kommunförbundet, 2003) stipulates nine
criteria[1] for recommending denial of discharge or for presenting criticism. We
determined which criterion or criteria were used most often in 2002 and 2003. We also
compared the boards/committees concerned, the outcome in the council, the role of civil
servants and whether there was clear partisan political influence. Since the auditors
have not always clearly indicated the criterion for recommending denial of discharge,
we interpreted the text to permit classification.

Case studies[2]
In one larger mid-Swedish municipality, the council denied discharge for 2003 to the
executive leadership of the arts and leisure board. The reason was that the board had
decided on improvements for a public bathing facility and that the cost control for this
project had been unsatisfactory. The auditors criticized the board for totally
inadequate internal control and management control of this investment. The board had
gone ahead with the investment during the year without any financing at all and in
conflict with the strict standards of budgetary control set by the council. No reports
were available from the purchasing representatives or the project manager. By
 JHRC granting delegation to the chairman with no limitation on amount, the board had also
violated the Local Government Act and the general rules of the municipality for its
9,2 boards. Moreover, invoices had been attested after budgeted appropriations had
already been used up. The executive leadership of the board was considered to bear a
greater responsibility than the other members of the board since the former received
annual compensation, which is associated with more extensive responsibility. The
132 council was unanimous in denying discharge to the executive leadership of the board.
In this case, the executive leadership can be considered to have violated the operating
guidelines set by the council.
In the county council district, the council was denied discharge for 2002. In the
council, the review on the question of accountability was the subject of a long and
intense debate, and a divided council denied discharge. The reasons were the
continually growing deficits and the passivity of the county council in regard to
Downloaded by FREIE UNIVERSITAT BERLIN At 02:16 13 May 2015 (PT)

management control. These shortcomings were allowed to persist without any action
being taken to remedy them, and the deficits grew. In the opinion of the auditors, the
council’s preparation of budgets was not correct, as deficits were present from the
outset and the equalization subsidy was reported separately. This increased the
financial deficits and decreased the equity capital of the county council district. For
several years the auditors had been criticizing the management control of the county
council district, and the auditors unanimously recommended that discharge from
liability be denied for 2002. The 2002 election gave the county council district a new
political majority, and the departing board was not granted discharge for 2002.

“Unclean” audit opinions for 2002 and 2003

In 2002, the auditors’ opinion (see Table I) included criticism and/or recommended
discharge from liability in at least 35 Swedish municipalities and county council
districts/regions; in 2003, this was the case in at least 28 municipalities and county
council districts. We examined and analyzed these cases more closely. It should be
noted that auditors could present criticism and recommend denial of discharge for
several boards/the executive board in the same municipality/county council district.
It should be noted in this connection that 2002 was an election year. Whether that
situation affected the number of criticisms, or “cases”, is difficult to determine. It is
clear, though, that the number of cases where the auditors recommended denial of
discharge and the council followed the auditors’ recommendation was fairly constant
in the two years. In all cases where the audit opinion was “unclean”, we closely
examined the auditors’ reasons. Our purpose was to identify any weak links in the
chain of accountability, to the extent that this was possible. We can see that in 2002 the
election affected the outcome in two cases where discharge was an issue, with the new
majority denying discharge to the old majority following the recommendation of the
auditors in the audit report.

2002 2003

Criticism 50 40
Table I. Auditors’ recommendation that discharge from
Opinions in audit reports liability be denied 10 12
for 2002 and 2003 Council denies discharge from liability 3 3
 Good Auditing Practice stipulates nine criteria (see the “Method” section) for denying Accountability in
discharge or providing criticism. In our study, we more closely examined which local government
criterion or criteria were used, and how often, for 2002 and 2003. The most frequent
reason for criticism was “Deficient management, follow-up and control”, but other
criteria were also used, alone or in combination. The other criteria used were
“Inadequate goal achievement, failure to observe the objectives and guidelines set by
the council or in regulations”, “Previously criticized shortcomings”, “Damage to the 133
public trust or other intangible injury”, “Unauthorized decision-making” and “Deficient
accounting”.
We analyzed more closely those cases where the auditors recommended that the
council deny discharge. The findings are shown in Table II, distributed by the criteria
stipulated in Good Auditing Practice, by the board concerned, and by whether the
Downloaded by FREIE UNIVERSITAT BERLIN At 02:16 13 May 2015 (PT)

objectives and guidelines of the council were followed. We note that in surprisingly
many cases, the auditors indicated that the objectives and guidelines of the council had
not been followed. However, the auditors did not always specify this shortcoming as a
reason for their decision to recommend denial of discharge.
In those cases where the auditors chose to recommend denial of discharge, the
objectives and guidelines of the council had not been followed in six cases in 2002 and
in eight cases in 2003. For 2002, the council denied discharge to three boards: an
engineering board that subsequently ceased to exist and two other boards (one
executive board of a county council district and one executive board of a region). In the
latter two cases, the departing board was denied discharge, and a change in political
majority had occurred at the latest election. For 2003, discharge was denied for three
boards:
(1) one executive leadership of a board of arts and leisure;
(2) one engineering board; and
(3) one social welfare board.

Criterion according to Good Auditing Practice 2002 2003

Deficient management, follow-up and control 10 11

Inadequate goal achievement, failure to observe the objectives and guidelines
set by the council or in regulations 1 3
Previously criticized shortcomings 2 2
Deficient accounting 1 0
Damage to the public trust or other intangible injury 0 1

Which board? 2002 2003

Social welfare, health and other care 3 4
Executive board of municipality, county council district or region 4 1
Engineering, environment and service 2 2
Child care, schools and education 1 2
Arts and leisure 0 2 Table II.
Boards for which
Have the objectives and guidelines of the council been followed? 2002 2003 auditors recommended
No 36 23 denial of discharge,
reasons for
Note: aAll cases in the study recommendation
 JHRC In two of these three cases, the actions of civil servants were considered in connection
with the issue of accountability. In one case, the board was criticized for not having
9,2 questioned the documentation provided by administrative staff to support the
assessments and reports that had been made. In the other case, civil servants and
elected officials had exceeded the delegation limits and guidelines set by the council.
From analysis of the complete material for the two years, it is clear that there is
134 considerable debate in local government councils on decisions regarding discharge. In
several cases, partisan political differences are quite apparent, while in other cases
such differences have probably influenced decisions – i.e. council members have not
voted against members of their own political party. A further reflection is that on the
issue of accountability, council members may have decided to vote for discharge rather
than against it on the ground that the mission was viewed as unclear.
Downloaded by FREIE UNIVERSITAT BERLIN At 02:16 13 May 2015 (PT)

Discussion
The ultimate purpose of the auditors in local government is to provide an annual
review of accountability and an auditors’ opinion on discharge preparatory to the
determination and decision by the council on that question (Sveriges Kommuner och
Landsting, 2005). In local government, the auditors’ opinion usually recommends that
discharge be granted, but in some cases the auditors present a critical opinion and/or
recommend that the council deny discharge. Where there are problems with a
board/committee, the auditors most often choose to criticize rather to recommend
denial of discharge. A critical opinion is not subject to decision by any
board/committee; the auditors themselves make the decision. Consequently, criticism
arouses less turbulence and often has beneficial and intended effects. A
recommendation that discharge be denied is usually made by the auditors when
previous criticism has had no effect and the board/committee in question has obviously
ignored the criticism. A government study (SOU 2004:107, 2004) has proposed that
local government councils also take a position on criticism and that in the
determination of accountability that the council states the reasons for any decision at
variance with the auditors’ opinion on accountability. The aim of this stricter provision
is to counteract any tendency toward partisan politics on the issue of accountability.
For the accounting year 2002, a number of audit groups chose to include in their
audit opinion a recommendation against discharge, the reason being deficient control.
In the case of the “county council district”, the council also followed the auditors’
recommendation in denying discharge to the county executive board. In cases where
discharge is denied on the ground of deficient control, the consequences are unclear.
Our empirical material suggests that it is politically difficult to deny discharge for
reasons of an intangible nature, such as deficient control or management, or damage to
the public trust. Decisions to deny discharge have also been based on other grounds. In
two of the cases that we studied, the denial of discharge accompanied a change in
political majority, and the issue was coloured by partisan politics. In other cases, the
board had obviously not observed the guidelines and delegation rules of the council,
with strongly negative effects on the finances of the board. Except for purely criminal
cases, the decision on consequences is left to the council. According to chapter 5 of the
Local Government Act, the council may decide to relieve elected or appointed persons
of their assignment when discharge is denied. Such a step should be considered
extremely drastic and has rarely been taken. Therefore, denial of discharge can also be
a “black mark” on a person’s political career, an unfortunate consequence for the
individual, but in some cases also for a political party as whole.
 The system of accountability is complex Accountability in
The complexity of the system of accountability makes it difficult to review and
determine accountability, and also to hold persons and governing bodies accountable.
local government
In this article we have discussed the system and chains of accountability in municipal
and county government. We have also provided an overall picture of the relationship
between principal and agent. In the empirical study, we can see signs that the council
(the principal) has not had a clear view of objectives to guide boards and committees 135
(the agents). At the same time, auditors have cited failure to follow council decisions. In
several cases we have also seen indications that the objectives and guidelines set by the
council have not been observed. In the study, we have focused only on the mission and
role of the auditors, and on the traces of various views of objectives that they have
identified. Through study of the different views of mission objectives held by principal
and agent, it is possible to identify how the audit can focus its review as well as how
Downloaded by FREIE UNIVERSITAT BERLIN At 02:16 13 May 2015 (PT)

the system of signals can be designed.

Costs resulting from asymmetric information

Various costs can arise from asymmetric information owing to the complex nature of
the system of accountability. The costs that arise and that may even be transferred to
the citizens are a result of this complexity. If the budget is overrun because of deficient
control and follow-up of operations by a board, tax revenue is probably not being used
in a way that is effective and appropriate in the long run. This may mean that the
operation provided for the citizens is a good one in the short run, but that it will need to
be limited in the long run.

The audit can be viewed as a signal

If we regard the auditors’ review as a signal to the council, it must be clear and precise
if it is to be credible. In several cases, auditors have identified and criticized a board or
committee for a number of years before deciding to recommend denial of discharge.
The auditors are required to perform a balancing act between not being too hasty in
recommending denial of discharge, on the one hand, and not simply continuing to
criticize and possibly being ignored as boringly repetitive on the other. Yandell (1988)
has shown that an audit can be viewed as a signal to the principal, reducing the risk
that the principal will be deceived. This is only true, however, if the audit successfully
identifies undesirable behaviour by the agent or shortcomings in the system of
accountability. The signal from the auditors is based on their analysis of risk and
materiality, and on their assessments. Improvement in the planning phase, and
identification there of the weaknesses in the chain of accountability, is of central
importance if the auditors are to improve the transparency of operations and provide a
clear signal of weaknesses in the chain.
An audit conducted with and based on a good analysis of risk and materiality may
provide a clear signal to the principal, but this signal will be pointless if the principal
ignores the auditors’ assessment. It is therefore essential that the principal, i.e. the
council, gives the reviewing body due recognition and in case of an “unclean” opinion
decides in accordance with the auditors’ recommendation more frequently than today.
One possible reflection based on the empirical material is whether the current injection
of more politics into the review of accountability by the council may be due to the
observation that many council members occupy “more than one chair” and may
therefore be reluctant to decide against members of their own party. If a council
consists substantially of persons with accountability subject to review, the review
process will naturally be weakened. For this reason, the proposal in the government
 JHRC study that the council state the reasons for any decisions inconsistent with the
auditors’ recommendations is a positive one. It will probably tend to strengthen the
9,2 process of reviewing accountability.

Reviewing accountability on the basis of ”soft” values

136 In our study we clearly see that when auditors have cited “soft” values, i.e. deficient
management, control or follow-up, or damage to the public trust or other intangible
injury, as reasons for recommending denial of discharge, councils have nevertheless
chosen to grant discharge. On the other hand, when auditors have cited “hard” values
as reasons to deny discharge, it is easier for councils to follow their recommendation. In
the determination of accountability, there is thus a built-in conflict in denying
discharge for shortcomings that cannot be calculated and confirmed quantitatively in
Downloaded by FREIE UNIVERSITAT BERLIN At 02:16 13 May 2015 (PT)

some form. There may be several reasons for this, one being that it is politically
impossible to deny discharge; another would be the fear that elected officials would be
reluctant to accept a position where they would risk criticism for errors of judgment.
Auditors and councils must therefore agree on the content and strength of the criteria
used in the review of accountability. Moreover, persons whose accountability is
examined need to know the criteria on which they are judged.

Conclusions
On the basis of our study, we have shown that neither better accounting systems nor
better auditing methods will make it easier to determine accountability. The system of
determining accountability is flawed in other, more fundamental respects. The unclear
and complex system of accountability makes it hard to achieve radical improvement of
auditing methods or resort to professional accountants as means of correcting the
problem. Using the framework of PA theory has facilitated the analysis of the chain of
accountability. The theory has proved so useful that in our opinion it will be more
frequently employed in analyzing local government auditing in the future. In our view,
our study and its conclusions are also applicable, in appropriate parts, to the business
world. In multinational firms, there is complexity similar to what we have studied in
local government. The advantage of studying the phenomenon in Swedish local
government, rather than in large companies, is that in local government the Swedish
principle of free access to information makes it possible to obtain all necessary
information in an organization. The broader scope of the mission of appointed auditors
in local government, compared to that of certified accountants, also makes it possible to
study the audit review and determination of accountability on the basis of “soft”
values.

Notes
1. These nine criteria are: “Inadequate goal achievement, failure to observe the objectives and
guidelines set by the council or in regulations”, “Deficient management, follow-up and
control”, “Damage to the public trust or other intangible injury”, “Financial injury”,
“Unauthorized decision-making”, “Operations not conforming to law, criminal conduct”,
“Insufficient preparation of decisions”, “Deficient accounting” and “Previously criticized
shortcomings”.
2. The material is based on a previous study by the authors (see Nyman et al., 2005).
 References Accountability in
Ahlbäck, S. (1999), Att kontrollera staten (Controlling the Government), Uppsala University local government
Library, Uppsala.
Baiman, S., Evans, J.H. III and Noel, J. (1987), “Optimal contracts with a utility maximizing
auditor”, Journal of Accounting Research, Vol. 25 No. 2, pp. 217-44.
Behn, R.D. (2001), Rethinking Democratic Accountability, Brookings Institution Press,
Washington, DC. 137
Bontis, N. (2003), “Intellectual capital disclosure in Canadian corporations”, Journal of Human
Resource Costing & Accounting, Vol. 7 Nos 1/2, pp. 9-20.
Laughlin, R. (1996), “Principals and higher principals: accounting for accountability in the caring
professions”, in Munro, R. and Mouritsen, J. (Eds), Accountability: Power, Ethos and the
Technologies of Managing, International Thompson Press, London.
Lundin, O. (1999), Kommunal revision – en rättslig analys (Local Government Auditing – a Legal
Downloaded by FREIE UNIVERSITAT BERLIN At 02:16 13 May 2015 (PT)

Analysis), Iustus förlag, Uppsala.

Lupia, A. and McCubbins, M. (2000), “Representation or abdication? How citizens use institutions
to help delegation succeed”, European Journal of Political Research, Vol. 37, pp. 291-307.
Müller, W. (2000), “Political parties in parliamentary democracies: making delegation and
accountability work”, European Journal of Political Research, Vol. 37, pp. 309-33.
Nyman, C., Nilsson, F. and Rapp, B. (2005), “Ansvarsprövning med förhinder” (“Obstacles in the
determination of accountability”), forthcoming.
Olsson, B. (2003), “Editorial”, Journal of Human Resource Costing & Accounting, Vol. 7 No. 3,
pp. 5-6.
Payne, J.L. and Jensen, K.L. (2002), “An examination of municipal audit delay”, Journal of
Accounting and Public Policy, Vol. 21, pp. 1-29.
Power, M. (1997), The Audit Society: Rituals of Verification, Oxford University Press, New York,
NY.
Rapp, B. and Thorstensson, A. (1994), Vem ska ta risken? (Who is to Bear the Risk?),
Studentlitteratur, Lund.
SOU 2004:107 (2004), Att granska och pröva ansvar i kommuner och landsting (Examining and
Determining Accountability in Local Governments), Fritzes förlag, Stockholm.
Strøm, K. (2000), “Delegation and accountability in parliamentary democracies”, European
Journal of Political Research, Vol. 37, pp. 261-89.
Svenska Kommunförbundet (2003), God revisionssed i kommunal verksamhet 2002 (Generally
Accepted Auditing Standards), Ordförrådet AB, Stockholm.
Svenska Kommunförbundet (2004), Fullmäktige och revision, 2:a upplagan, Kommentus förlag,
Stockholm.
Sveriges Kommuner och Landsting (2005), Ett granskande samhälle: kommunerna och revisionen
(An Auditing Society: Audits in Local Government), Edita, Stockholm.
Tarschys, D. (2002), “Huru skall statsverket granskas? – om riksdagen och den demokratiska
kontrollen” (“How should government be audited? – on parliament and democratic
control”), ESO report DS 2002:58, ESO, Stockholm.
Venton, J.P. (1997), “A general theory of delegation, accountability and empowerment”, The
Canadian Journal of Program Evaluation, Vol. 12 No. 2, pp. 163-88.
Yandell, D. (1988), “Audit information and incentives for efficiency”, American Economist, Vol. 32
No. 1, pp. 49-58.
 This article has been cited by:

1. Nidal Rashid Sabri, Rania Yaser Jaber. 2007. Managerial performance of Palestinian local authorities.
Transforming Government: People, Process and Policy 1:4, 350-363. [Abstract] [Full Text] [PDF]
Downloaded by FREIE UNIVERSITAT BERLIN At 02:16 13 May 2015 (PT)