Legitimacy and parliamentary oversight in Australia

The rise and fall of two public accounts committees

The Authors

Kerry Jacobs, Australian National University, Canberra, Australia

Kate Jones, Australian National University, Canberra, Australia

Acknowledgements

The authors wish to acknowledge the contribution of seminar participants at University of Technology Sydney and the contribution of the two anonymous reviewers.

Abstract

Purpose – The purpose of this paper is to address the question of whether two early Australian public accounts committees were established for the purpose of legitimating governments of the time.

Design/methodology/approach – The paper addressed these issues through a study of the establishment, early work and abolition in the 1930s of the Victorian Committee of Public Accounts (VCPA) and the Joint Committee of Public Accounts (JCPA).

Findings – Clear evidence is found that the Joint Committee of Public Accounts (JCPA) had been copied from the VCPA and that the VCPA had been copied from the UK House of Commons Committee of Public Accounts, which was established in 1861. This would indicate that the primary objective in the establishment of both these committees was legitimation rather than control. It was found that the subsequent work of both the VCPA and the JCPA showed a drift away from an accounting focus towards a policy focus. This is similar to the JCPA experience described by Degeling et al. in relation to the JCPA, which also supports the legitimation argument. It was also found that both committees could be disestablished with relative ease because their legitimating purpose was no longer strong enough to demand their continuation and that, in fact, their abolition became the factor that served a legitimating purpose for governments.

Originality/value – The paper suggests that the ideas of legitimation and mimetic isomorphism provide a more convincing explanation for the nature and work of these two public accounts committees than the idea of accounting colonisation.

Article Type:

Research paper

Keyword(s):

Public sector accounting; Australia; Regulation; Modern history.

Journal:

Accounting, Auditing & Accountability Journal

Volume:

Number:

Year:

2009

pp:

13-34

Copyright ©

Emerald Group Publishing Limited

ISSN:

0951-3574

Introduction

Power (1997) claimed that institutions of oversight and audit have come to play an increasingly important role in the creation of legitimacy and social stability in modern society. He highlighted changes within the public sector and drew a link between processes of audit, trust and risk within structures of governance and regulation. Hood et al. (2001) also highlighted these issues of governance and regulation within a process of the evolution of a risk society described by Ulrich Beck. However, these institutions have long histories and interesting origins.

The research question addressed in this paper is whether the ideas of isomorphism and policy transfer provide an explanation for the establishment and early work of the Victorian and the federal Australian Public Accounts Committees (PAC). The PAC is a key example of an institution of audit, governance and regulation within the Westminster parliamentary system. Although it might not be considered a classical example of accounting practice, it has been required to exercise an accounting role through the control and oversight of parliamentary spending. Therefore, the PAC has become a central element of the process of parliamentary and democratic accountability. Our focus on this committee can be understood as part of the institutional turn to public administration, politics and accounting. Hopwood (1984) argued that accounting language and practices have become intertwined in issues of public policy, both reflecting and influencing the public debate through the creation of selective patterns of economic visibility and disciplining performance so that accountability can be demanded, policed and enforced. It is these patterns and practices, which make up the structures or institutions of accountability and oversight.

The paradox of any institutional analysis is that the institutions being studied are the product of events and influences from the past and yet function and interact with contemporary settings. Therefore, in order to adequately grasp the nature and operation of these processes of oversight, governance and accountability that Power (1997) claims characterise modern society, we must consider the origins and early operations of these social institutions and practices.

Within the accounting literature, Degeling et al. (1996) provide one of the few studies of the work of a parliamentary audit and oversight committee – the Australian federal Joint Committee of Public Accounts (JCPA). Degeling et al. (1996) proposed that accounting and economics tend to colonise broader concerns regarding equality, access and quality within the work of the committee. However, they found that was not the case and that the initial “accounting” focus of the reports produced by the JCPA was replaced by a policy and political focus over time. The limitation of this work is that while Degeling et al. (1996) provided a valuable description of the changing accountability focus of the work of the JCPA, they failed to fully explore the origin and establishment of this committee. They also did not consider that the JCPA was suspended in 1932 and not re-established until 1952. These gaps provide the motivation and focus for our paper. We note the direct links between the establishment of the JCPA and the earlier work of the Victorian Committee of Public Accounts (VCPA). It can be established from parliamentary debates and the legislation that the JCPA was copied from the VCPA, which predated the JCPA by nearly 20 years. The JCPA and the VCPA were also abolished at a similar time. Therefore, these institutions need to be understood as a pair in both their establishment and their disestablishment. We pay particular attention to the establishment of the Victorian committee (VCPA) because it predated the JCPA.

We found evidence that the JCPA had been copied from the VCPA and that the VCPA had been copied from the Westminster Committee of Public Accounts, which was established in 1861, in the UK House of Commons. This indicates that the primary objective for the establishment of both these committees was legitimation rather than control, which would be assumed if accounting served a functional role. Within the institutional framework, this process of copying can be described as mimetic isomorphism (Powell and DiMaggio, 1991; Scott, 2001). The concepts of mimetic, coercive and normative isomorphism assume that institutional practices serve to create legitimacy rather than fulfil a functional role. We explore this possibility further by contrasting the work of the VCPA with the analysis presented by Degeling et al. (1996). If the role of the VCPA were a functional one, it would be reasonable to expect that the VCPA would achieve the objectives and purpose established at its foundation.

We addressed these issues through a study of the establishment and work of the VCPA between 1895 and 1931. We found evidence for the case of legitimation because the subsequent work of the VCPA drifted from an accounting to a policy focus in a manner similar to that of the JCPA, as described by Degeling et al. (1996). However, we suggest that the ideas of isomorphism and policy transfer provide a more convincing explanation for the nature and work of these two committees.

Public accounts committees and legitimacy

McGee (2002, p. 9), in a review of the work of PACs and Auditors-General across the Commonwealth, argued that democracy entails accountability for the exercise of power and that the PAC is one organisational form in which Parliament ensures the accountability of government, particularly the process of public financial accountability. Normanton (1971, p. 312) suggested that public accountability was the chief hallmark of western democracy, which called for “openly declared facts and open debate of them by laymen and their elected representatives”. This perspective adopts a strongly functionalist approach towards these social institutions, presenting a regulative and possibly a normative argument for the nature and function of these institutions. In that sense, the PAC can be seen as a vital pillar to the legitimacy of political institutions, as argued by Jurgen Habermas (1976). Offe (1996) followed the same argument with the idea that the elements of legislative representative bodies such as parliamentary committees play an important role in securing public legitimacy for the political systems.

Within the Westminster model of parliamentary systems, elements of accounting oversight were introduced as early as 1861. The first PAC was established in the UK Westminster parliament in 1861 by the then Chancellor of the Exchequer, William Gladstone (McGee, 2002; Jacobs et al., 2007). The responsibility and role of the Westminster PAC was to examine the accounts showing the appropriation of sums granted by Parliament to meet the public expenditure. While the members nominated were selected to give satisfaction to “both sides of the house”, they were primarily intended to bring as much talent, knowledge and experience as possible to the examination of accounts (Hansard, Vol. 162, 19 April 1861, col. 773). Gladstone claimed that it was the members who were best qualified (in the examination of public accounts) that were selected for a duty which was seen as being “of a dry and repulsive kind” (Hansard, Vol. 162, 19 April 1861, col. 774; Stent, 2004). PACs were established as guardians of the public finances, with a responsibility for holding the Government to account for its use of public money and resources (McGee, 2002). Within this system, accounting was presented, if not understood, as being an objective, neutral and apolitical technology. Gladstone himself took the view that “Economy is the first and great article (economy as I understand it) in my financial creed” (Morley, 1903, p. 62). However his passion for thrift was combined with a desire to establish a regulated and orderly system of financial control dominated not by politics but by the legal and constitutional provisions. He was not alone in this. Chubb (1952) described the attitude of members of parliament towards public spending in the following terms:

… they approached it rather as financial and constitutional experts than as politicians, and the system they designed was intended primarily to deal with the facts and the law. The facts, recorded in the accounts, were clear; the law, embodied in the Appropriation Acts and Treasury regulations, was also clear. It remained for the Accounts Committee to give a decision. Thus the experts combined the predilections of the age with the knowledge that came of investigation and study, and their conclusions were not only welcome, but incontestable (Chubb, 1952, p. 35).

However, the link between legitimacy, democracy and accountability as embodied by the PAC is complex. In general, the argument is that contemporary implementations of NPM/NPFM reform have altered accountability; there is an increasing emphasis on managerial and managerialist forms of accountability and correspondingly less on political, parliamentary and public accountability (Hughes, 1998; Jacobs, 1998; Jacobs et al., 2007). This point was clearly made by Gray and Jenkins (1993) in their review of the place of accountability within the UK public sector. Pusey (1991) and Guthrie and Parker (1998) have made a similar argument in the context of Australia.

Degeling et al. (1996) analysed the work of the Australian federal Joint Committee of Public Accounts (JCPA) and compared the foundational statements about the committee's claimed accountability function with the performance of accountability as contained in its reports. Degeling et al.'s (1996) thesis was that the JCPA would adhere to its foundational focus on matters of financial accountability and possibly narrow its sphere of interests, abstaining from any matters with a non-financial element. From a functionalist perspective, it would be reasonable to assume that the JCPA would follow its foundational mandate in the work conducted. However, from a legitimacy perspective this is not a valid assumption as the committee itself would have either been copied or imposed to improve the perceived legitimacy of the Commonwealth Government. Degeling et al. (1996) found that the JCPA did not function as a colonising influence and that while the early work of the JCPA reflected its financial accountability mandate, this was not true in the later years, when it was concerned with effectiveness and policy advice instead.

Degeling et al.'s (1996) conclusion that the financial accountability mandate was colonised by the political is understandable as they focused on the early work of the JCPA which predates NPM. What is not clear from Degeling et al. (1996) is the place accounting played in the formation of these institutions. We will look in more detail at the establishment of these committees, to consider both the JCPA and the Victorian Committee of Public Accounts (VCPA) and provide an alternative explanation, drawing from an institutional theoretical understanding. Because the VCPA predated the JCPA by nearly 20 years (Davey, 1960, p. 3) and the structures mandate and practices of the JCPA were copied from the VCPA it would be reasonable to describe the initial structure of the JCPA as an example of isomorphism or policy transfer and argue that the primary function and role of the JCPA was legitimacy. If this was true then the JCPA's abandonment of its financial accountability focus would be unsurprising. Therefore it is necessary to study the earlier VCPA and determine the nature and origin of its mandate within that jurisdiction and to understand what influenced its establishment. This also raises the possibility that aspects of accountability attributed to NPM reform may have a longer history as they may be reflected in the institutional structures developed in the 1860s and 1870s.

New institutional theory (see, Powell and DiMaggio, 1991; Meyer and Rowan, 1977) has been extensively used within the accounting literature to describe change over time (e.g. Fogarty, 1996; Carpenter and Feroz, 2001; Fogarty and Rogers, 2005). The argument is that institutions need legitimacy in order to function and to be perceived as rational. Scott (2001, p. 58) provides a brief definition of legitimacy with the statement that “… in addition to material resources and technical information organisations require social acceptability and credibility in order to survive and thrive in their social environments”. Scott turns to the work of Suchman (1995, p. 574) to provide a fuller definition of legitimacy as “… a generalised perception or assumption that the actions of an entity are desirable, proper, or appropriate within a socially constructed system of norms, values, beliefs and definitions”. The argument is that institutions comply with these social norms and value and tend to adopt similar structures or practices in order to secure that legitimacy.

The way that organisations adopt similar structures or practices is described by Meyer and Rowan (1977, p. 352) as institutional isomorphism. Powell and DiMaggio (1991, p. 67) provided a three-part typology of the social rules and pressures, which lead to institutional isomorphism: coercive isomorphism, mimetic isomorphism and normative isomorphism. Powell and DiMaggio (1991, p. 67) identified the common legal environment and state regulation as examples of coercive isomorphism. They also stated that expectations and persuasion experienced as force constitute this kind of isomorphism and that these pressures can take the form of budget standardisation, performance measures, reporting systems or even the effects of monopolistic service infrastructures such as transport and telecommunications. However, where organisational technologies are poorly understood and goals are ambiguous, uncertainty provides a powerful force that encourages imitation. Powell and DiMaggio (1991, p. 69) described this as mimetic isomorphism. This process is also described in the politics literature as policy transfer or policy learning (Dolowitz and Marsh, 1996; Jacobs and Barnett, 2000). Powell and DiMaggio (1991, p. 71) suggested that normative isomorphism stems primarily from professionalisation and the expression of a set of values and expectations associated with university education and cross-organisational professional networks. It is the commonality of values associated with professional orientation and the selective recruitment of staff from certain educational institutions (Powell and DiMaggio, 1991, p. 72) or certain social and economic settings, which leads to similarity in organisational structures and practices (Jacobs, 2003). Scott (2001, p. 56) further developed the idea of normative isomorphism by arguing that common values also include rules, routines and procedures. Scott (2001, p. 57) went so far as to include cultural and cognitive values as examples of normative isomorphism – in effect, the shared conceptions associated with how actors attribute meaning to social experience and construct social reality.

Isomorphism would explain the objective for the establishment of the early VCPA in Victoria as strengthening the legitimacy of the colonial parliament. It could reflect a coercive pressure from the UK to adopt structures and practices similar to those of the Westminster parliament. It could also reflect institutional or policy transfer and copying; the Westminster parliament was seen as an example of good government and democracy, so it became desirable that the Victorian model copy it in every way. The next part of this paper explores the establishment of the VCPA. We also briefly revisit the establishment of the federal JCPA described by Degeling et al. (1996) to explore the legitimacy aspects.

Establishing the Victorian Committee of Public Accounts

The VCPA was established in 1895, but the creation of a PAC in the Victorian parliament was first proposed in 1870 by the Select Committee on Financial Arrangements. Prior to its successful creation, there had been two failed attempts in 1886 and 1891. There was no evidence of any effort on the part of the UK based colonial authority to encourage or require the establishment of a PAC and therefore there is no direct evidence of coercive isomorphism. There is evidence for mimetic isomorphism as the first attempt proposed a committee that was based exactly on the British House of Commons committee. The second attempt was more radical, arguing for a committee that would be a cross between the British and the French systems of parliamentary scrutiny, with power to examine the estimates as well as the appropriations (VPD, Legislative Assembly 2 December 1891, p. 2805). The French system involved committees in both houses. The 33-member Commission des Comptes had similar responsibilities to the House of Commons committee, but the other committees dealt with appropriations. Overall, the parliamentary committees in the French system oversaw government finance at all stages. The initiative for the second attempt, in the form of a Public Accounts Committee Bill, came from William Shiels, Attorney-General and Minister of Railways, seconded by the premier, James Munro. Shiels described the objective of this bill in his second reading speech:

The house had special duties to perform, and among them was that of putting down extravagance. What was wanted in the public interest was a means of arriving at the truth with respect to financial matters (VPD, Legislative Assembly 2 December 1891, p. 2807).

Shiel's speech indicated that, from the early proposals, the role of the VCPA was to provide an objective and politically independent analysis of financial activity and that it was understood accounting could provide that. Jones (2006, pp. 198-199) argued that the opposition to this bill was because it was proposed as legislation rather than as a standing order (the then basis for the committee at Westminster) and that it diverged from the practice of the Westminster House of Commons with the proposed additional estimates review powers. In the debate there was clear resistance and opposition to the Victorian Parliament going beyond the British model in any way (Jones, 2006, p. 199) and while copying off the UK model was a powerful source of legitimacy, copying off the French model was not.

In 1894, the conservative premier, James Patterson, established a royal commission into constitutional reform with 12 members representing all political positions. The royal commission not only recommended that Victoria establish a PAC but also presented a draft standing order for the committee. The government changed several months after the royal commission had reported and Patterson was replaced by a Liberal, George Turner, but the move to establish a PAC still went ahead. The motion that a standing order to establish the committee be prepared was moved by Robert Best, who had chaired the royal commission. The background to, and bipartisan support for, the establishment of the VCPA suggest that it was not the result of party politics. Best was by this time a minister in the Liberal government and his motion was seconded by the conservative Godfrey Carter, who had moved the 1886 motion. The motion of Best and Carter clearly specified the financial accountability role of the committee and the notion of its neutral and apolitical nature:

Financial matters should, in my opinion, be completely apart from party politics; they should be dealt with on their sole merits. The very object of a neutral committee of this kind should be to recommend as to the most economical management, financial and administrative, of public affairs … (VPD, Legislative Assembly, 1894, p. 1212).

Like the 1891 proposal, the royal commission's proposed standing order also included the power to examine the estimates. However, when Best put the motion to parliament, he omitted the proposal to deal with estimates “to eliminate from the motion all controversial matter … so that we may be exactly on all-fours with the English precedent” (VPD, Legislative Assembly 19 December 1894, p. 1210). This move was correct because in the subsequent debate, any deviation from the English system and any evidence of a move towards the French system were staunchly resisted (Jones, 2006, p. 200). The resulting standing order provided the following introduction and specified the following duties for the committee:

At the commencement of every session of Parliament the Legislative Assembly, according to the practice of Parliament with reference to the appointment of select committees, shall appoint a select committee of seven members, to be called the Committee of Public Accounts, with powers to send for persons, papers, and records. The duties of the Committee of Public Accounts shall be as follows:

To examine the accounts of the receipts and expenditure of the colony, and to bring under the notice of the Legislative Assembly any items in those accounts, or any circumstances connected to them, to which it may consider the attention of the Legislative Assembly should be directed.
To report to the House any alteration which may appear to the committee desirable to be introduced in the form of or method of keeping the public accounts, or in the mode of receipt, control, issue or payment of the public money.
To inquire into and report upon any question which may have arisen in connexion with the public accounts.
To inquire into and report to the Legislative Assembly upon the investment of and dealings with the funds of the Commissioners of Savings Banks.
To deal with any special references that may be made to them by the Legislative Assembly (VPD, Legislative Assembly, 1895, pp. 2093-2094).

The introduction clause (a) of these standing orders covered the entire UK House of Commons standing order while adding the additional duties of examining the receipts (in addition to the expenditures) and reporting back to the Legislative Assembly (parliament). Although the obligation to report any desirable change to the accounts and accounting system (b) was beyond the formal UK mandate, this role was performed in practice. The wide-ranging power reflected in clause (c) to inquire into and report on any question connected with the public accounts was not reflected in UK mandate, although Jones (2006, p. 205) claimed that the UK PAC may have taken this power on itself. The focus (d) on the funds of the commissioner of savings banks was peculiar to the Victorian committee and reflected a concern at the time with the activities of the Victorian savings banks (Jones, 2006, p. 205). Therefore, this specific focus can be seen as a reflection of the fact that the Government had been investigating the structure and lending policies of Victorian savings banks (Jones, 2006, p. 205). This banking focus was not retained in the mandate transferred to the federal JCPA in 1914. Finally, clause (e) indicated that the VCPA was authorised to deal with any specific reference from the Legislative Assembly. This role also reflected the actual work and role of the UK PAC. The Victorian VCPA had taken 25 years to establish, and the stated objective was to provide a neutral, objective and apolitical form of accounting control and measurement. While early initiatives had failed, they succeeded with an “exact copy” of the UK committee.

There is an obvious question of why it took so long to establish the VCPA. We would argue that there are a number of reasons for that. Jones (2006, p. 205) claimed that the mandate of the VCPA indicates that it was not simply a copy of the mandate of the UK PAC at its inception. However, the extended mandate of the VCPA reflected the powers that had been granted to, or claimed by, the UK PAC by 1895. The concern with the oversight of the commissioners of savings banks was the only acceptable deviation from the UK model and, therefore, the only response to local Australian conditions. On the whole, the process of the establishment of the VCPA would provide a strong argument for mimetic isomorphism. It was important for legitimacy reasons that the VCPA was seen as being an exact copy of the current practices of the UK PAC. Any attempt to introduce ideas from the French was quickly rejected.

Scott (2001) indicates that the three institutional types of isomorphism are rarely separate but often function together in the generation of legitimacy. Therefore, in addition to the mimetic driver were there also elements of normative or coercive isomorphism? While a strong normative element to these changes is unlikely, Jones (2006, p. 208) does note that many of the key protagonists in the attempts to implement a Victorian PAC had visited England and would have been influenced by the parliamentary practices and systems they observed there. While there was strong evidence of the desire to copy the UK model and therefore mimetic isomorphism, there may still have been some of the ongoing cultural norms described by Scott (2001) and therefore also be elements of normative isomorphism.

Writers on Australian politics have generally rejected a coercive argument in the formation of the Government institutions. Jones (2006, p. 194) quotes the work of AF Madden (1979), who suggested that at the time responsible government was being established in Australia and New Zealand, there was little interest in exporting the British model of Westminster to these colonies. Although most states had a period of direct government from the UK, they were granted the power of self-government in the 1850s (Weller and Flemming, 2003 p. 14). New South Wales developed its own constitution from 1850 (Smith, 2003, p. 44) and was quite independent before that time given that the governor was generally appointed from within the colony rather than from London. Therefore while the institutions and practices may have been modelled on Westminster they were seen as not being imposed by Westminster. This also seems to have been the case with the other states and clearly the case with the establishment of the Commonwealth in 1901. Carroll (2006) indicated that until the 1850s there was a measure of voluntary policy and institutional transfer from Britain to its colonies. Between the 1850s and 1901 there was increasing local innovation and a growing transfer within Australia after 1901. If there was clear coercive pressure from the UK it is likely to have been more evident in the establishment of the institutions in the state of Victoria in the mid 1800s than in the Commonwealth institutions in the 1900s. There was no evidence of direct coercion in this case.

It is possible that the arguments against UK influence may have oversimplified the situation. One explanation for the successful establishment of the VCPA in 1895 was the crises of public confidence in the then Victorian government. The 1880s had been prosperous years for the Victorian economy, with high prices being paid for wheat and wool, high levels of British investment and a land boom fuelled by speculation. By 1894, however, the economy had collapsed, unemployment had risen, government revenue had fallen and a number of banks had collapsed. Some colonial politicians, including Munro, had been involved in scandals associated with land speculation and railways. Chua and Poullaos note that of the 48 members of the Victorian parliament in the late 1880s and early 1890s, 30 were land speculators (Chua and Poullaos, 1998, p. 172).

In 1890-1891, for the first time, government expenditure exceeded revenue and economy in government became a major concern. It was necessary to quickly and clearly restore public confidence and legitimacy. The establishment of a public accounts committee modelled on the Westminster model was an obvious and timely solution. While not exactly coercive it should be noted that during the 1890s that Victoria was heavily dependent on British capital (Chua and Poullaos, 1993, p. 699; Chua and Poullaos, 1998, p. 162). Therefore economic collapse, political scandals and the debts of the Victorian government meant that considerable pressure was applied to restore the legitimacy of both the market and the parliament (Chua and Poullaos, 1998, p. 173). The establishment of the VCPA in 1895 can be seen as a direct response to that desperate need to restore legitimacy.

In their discussion of the federal JCPA, Degeling et al. (1996, p. 34) highlight a similar need to strengthen public legitimacy and confidence. The Fisher Labor government had lost the 1913 election amidst criticism of its financial and economic policies and following the loss of a referendum to increase the powers of the Commonwealth. The Liberal Cook government that succeeded it established the JCPA with a stated responsibility to scrutinise and control public expenditure. Degeling at al.'s (1996) finding that the work of the JCPA drifted from its initial mandate seems consistent with our understanding that the establishment of the JCPA was an attempt to garner legitimacy of part of the Government. When the Treasurer (Sir John Forrest) introduced the bill for the JCPA, he made it clear that while the establishment of the committee was desirable, the duties and roles of the committee would be left for the parliament to define (APD (Australian Parliamentary Debates), 1913, p. 3277). It is clear from Forrest's introduction that the existence of the committee was deemed to be important rather than its having a particular mandate or function. This would challenge Degeling et al.'s (1996) focus on the mandate and the functionalist assumption that the committee was established to provide financial oversight while lending some support to a legitimacy argument.

In discussing the JCPA bill, Forrest also cited the standing order that had established the Victorian committee and set out its powers and duties together with similar orders from Westminster as a guide to be followed by the Commonwealth (APD (Australian Parliamentary Debates), 1913, p. 3278). The fact that the JCPA was a joint upper and lower house committee was a substantial structural change from both the Victorian and the British precedents. Nonetheless, in 1915, the committee requested from the Victorian committee “copies of State records to assist and [sic] Commonwealth Committee's operations” (NAA: A12840, 2004/1). This is not surprising. The Committee of Public Accounts Act that was finally passed specified responsibilities for the JCPA almost identical to those of Victoria's VCPA. From Forrest's statement, the evidence for the use of the Victorian standing order as a model and the similarity between the Victorian standing order and the Commonwealth legislation, it seems unquestionable that the JCPA proposal was copied from the VCPA practice. Within the institutional theory framework, this is consistent with mimetic isomorphism.

The establishment of the JCPA as a joint committee of both the upper house (Senate) and the lower house (House of Representatives) in an anomaly for an argument of isomorphism. In this case the structure and practices were not a copy from the Victorian model or from Westminster practice. However, it is clear that genuine innovations are both practical and desirable. In that sense the establishment of the JCPA as a joint committee can be argued as a functional response to the concern of both houses to exercise financial oversight. In essence the joint approach prevented the wasteful duplication of resources and the confusing duplication of role. However, it is also possible to present a legitimacy argument. While the seats in the House of Representatives are allocated to reflect the population and are therefore dominated by the largest states (Victoria and New South Wales), the positions in the Senate are structured to reflect the concern that the interests of the smaller states would be protected and therefore the Senate is sometimes described as the “states house”. It was this very concern that almost resulted in some of the smaller states not joining the federation. Therefore the inclusion of representatives from both the Senate and the House in the process of financial oversight can be seen as a savvy political move, which provided further legitimacy to the actions of the JCPA.

We argue that the establishment of both the VCPA and the subsequent JCPA was driven by the need for legitimacy. However, although an institution may be established as a tool of legitimation under the influence of mimetic isomorphism, it has the potential to change and evolve to better meet local needs and conditions as reflected in the joint JCPA (Carroll, 2006). This provides a powerful institutional explanation for the drift away from the initial JCPA mandate observed by Degeling et al. (1996). In the next part of this paper, we explore whether a similar drift or evolution is also evident in the work of the VCPA.

The work of the VCPA

Following the Degeling at al. (1996) template, we analysed the reports of the VCPA from its establishment until its disestablishment in 1932. There were profound differences in the political environments of the VCPA and the JCPA at their establishment. In 1895, although groups inside parliament were described as parties and described themselves as such, they did not have the organisational structure of modern political parties; there was no extra-parliamentary infrastructure and no party discipline. Although members of parliament could be broadly divided between liberals and conservatives, they could, and did, form loose alliances and factions with each other according to personal relationships, common interests and short-term advantage. By the time the JCPA was established in 1914, political parties had become firmly entrenched in Australian politics (Starr et al., 1978), the Australian labour movement was in a position to contest the first federal election in 1901 as a party, the first Liberal Party had been established in 1910 and the Country Party, the predecessor of the National Party, had been created by state farmers' organisations around the time of World War I.

Between the establishment of the VCPA in 1895 and 1932, there were 15 committees. The fifteenth committee, appointed in 1931, was the last until 1955 as no members were appointed to it in 1932. Between 1895 and 1932, the VCPA published 57 reports, of which 55 were reports of its inquiries. The remaining two, in 1897 and 1901, reported only on the activities of the committee1. They were entitled Report on Activities and have been excluded from the analysis. There were also some clear differences in the activity levels of some VCPAs, with the third (1900-1902) and the eleventh (1920-1921) producing no reports at all while the eighth (1911-1914) produced eight reports and the ninth (1914-1917) and the twelfth (1921-1924) both produced seven reports each.

Initially the VCPA adhered to the financial focus laid down for it by the standing order. The first seven committees, which covered the 17 years from 1895 to 1911, produced reports mostly with a financial focus. The third (1900-1902) committee only produced a report on its activities (1901), which noted that no questions had been referred to it by parliament or government and that “the Annual Report of the Audit Commissioners still continues to be presented to the House at such a late period as to render it impossible for the Committee of Public Accounts to take it into consideration during the Session in which it is received” (Committee of Public Accounts, 1901). In other words, if neither government nor parliament referred questions to it and the audit commissioners' report was not available, the committee was left without work.

The eighth committee (1911-1914) began the move away from a financial oversight focus and produced eight inquiry reports. The ninth committee spanned most of the period of World War I and addressed a number of financial issues, such as the financial position of the state coal mine at Wonthaggi (Committee of Public Accounts, 1916b), the access of parliament to information included in the budget statement and the Auditor-General's annual report (Committee of Public Accounts, 1916a; Committee of Public Accounts, 1917a). However, it also showed a clear willingness to address more political issues, such as the management of the Government cool stores, the disposal of funds on government research scholarships (Committee of Public Works, 1915; Committee of Public Works, 1916) and the need for improvements in the management of public sector organisations (Committee of Public Accounts, 1915a; Committee of Public Accounts, 1917). Subsequent committees followed this movement into more political themes, although financial issues were occasionally addressed. Reports ranged over issues of substance related to land settlement, the Country Roads Board and government-owned businesses. In effect, the work of the VCPA generally followed the transition observed by Degeling et al. (1996) from a financial focus towards a more political one. However, it needs to be noted that there were also substantial periods when the work of the VCPA was intentionally or unintentionally blocked.

Both the VCPA and the JCPA had been established to bolster the legitimacy of the Governments in their respective jurisdictions at a time when those governments needed and wanted to assert their own economic and financial wisdom. They had continued to exist regardless of how far their reports strayed from the mandates laid down for them. They also played a legitimating role for the respective parliaments in that their existence was evidence of how parliaments could oversee government spending. This was significant, as the power of parties increased throughout the early years of the twentieth century. Therefore, the power of legitimacy seemed to be derived from the existence of the VCPA committee rather than the activity and did not seem to be challenged by the periods of relative or enforced inactivity.

What remains unanswered is why the VCPA and the JCPA were abolished in 1932. This seems to provide a major challenge to the legitimacy thesis. The explanation given retrospectively and anecdotally in the case of the VCPA and at the time of abolition in the case of the JCPA was “economy”. This was the height of the 1930s Depression, and money would be saved if the committees ceased to operate. If the VCPA and the JCPA were created to bolster the public legitimacy of the parliamentary systems, why did they suddenly become unnecessary? The next section of this paper explores the issues and events associated with their suspension. It explores the final reports produced by the VCPA and JCPA on the grounds that the committee suspension might have been a political response to these reports rather than an economy measure as reported.

The demise of the VCPA

By the late 1920s, the VCPA's reports were principally concerned with politics and policy rather than finance. Of the six reports produced by the fifteenth committee (1929-1932), only one (on infringements of the Public Account Advances Act) focused on financial control.

The operation of the Maffra Beet Sugar Factory was an important and political issue, which generated considerable interest. The fourteenth VCPA had already reported on it comprehensively in 1929 (Committee of Public Accounts, 1929). Sugar beets had been grown in a small way in Victoria since 1866. The Victorian Government had become involved in the 1890s when it advanced £63,000 to a company building a beet sugar factory at Maffra. The company defaulted on the interest payments and the Government took possession of the factory. It was idle for ten years, reopened in 1910 and was still operating when the VCPA began its inquiries. Some years after the factory reopened in 1910, the Government decided to remodel it to increase its capacity. Between 1917 and 1924, the Director of Agriculture and the manager of the factory had visited the USA in pursuit of best practice. Subsequently, two American beet sugar engineers had been brought to Victoria to report on the remodelling proposals. One of them remained to supervise the work. The VCPA suggested that there was no apparent benefit from this. It also reported that public servants who had been called before the committee did not provide the material evidence requested (Committee of Public Accounts, 1929, p. 5). The report of the fifteenth committee (Committee of Public Accounts, 1930a) had a similar agenda, suggesting ways of improving the management and organisation of the Maffra factory and increasing productivity. In effect, the committee made itself responsible for reviewing the economic and financial results of a government business through highlighting issues of efficiency and effectiveness in the management of the organisation. This would have been seen to challenge the legitimacy of the then government by questioning its management of the sugar beet factory.

In 1930 and 1932, the committee also issued two reports on the Closer Settlement Board (Committee of Public Accounts, 1930; Committee of Public Accounts, 1931a). The term “closer settlement” referred to a longstanding government policy of increasing the population in rural areas by granting land to prospective small farmers (McDonald, 1978; Dingle, 1984). The effectiveness and success of the policy were important political issues, and both reports went beyond financial matters to address questions of its successes and failures. The first report was, for the most part, a severe criticism of the accounting procedures of the Closer Settlement Board. However, it also supported the view of a Commonwealth Royal Commission into Soldier Land Settlement that the soldier settlers had failed in their settlements because the areas granted were too small, they were undercapitalised, they were poorly trained to be farmers and there had been a drop in the value of primary products. The second report in 1931 reviewed the administration of the Lands Department and the Closer Settlement Board. The VCPA argued that the land settlement policy had become “a heavy drain on the revenues of the State” and would have to be revised (Committee of Public Accounts, 1931a, p. 4). These reports would also have been seen to challenge government policy and to publicly question the ability of the Government to effectively manage state resources. It seems likely that this would have undermined the public legitimacy of the Government.

The fifteenth committee issued five other reports on government businesses or utilities. In 1930, it examined the Victoria Dock Cool Stores and in the Country Roads Board. The report into the Victoria Dock Cool Stores (Committee of Public Accounts, 1930b) followed a letter to the Melbourne newspaper the Argus from the proprietor of another cool store arguing that the Government was not equipped to conduct such a commercial operation. The VCPA examined the operations of the Victoria Dock Cool Store, supported its management and made some recommendations about its improvement. In its report on the Country Roads Board (Committee of Public Accounts, 1930), the committee made recommendations that related to not only roads and the management of the board but also registration fees and compulsory third party insurance for motor vehicles. Such an active and public policy formation role would have done little to build public confidence in the policies of the Government. While it is clear that the fifteenth VCPA committee had moved substantially away from an initial financial focus and was now engaged with the criticism and formation of policy, this was not a particularly new activity. The move to disestablish the committee could have been a response on the part of the Government to the criticisms and the associated loss of public legitimacy. However, the then Labor government (Hogan 1929-1932) took no direct action against the committee, possibly because of the weakness of his government and the level of political instability at that time. The decision not to establish a new VCPA was taken by the subsequent conservative (Argyle 1932-1935) government.

While there were political motivations for the Argyle Government not to re-establish the VCPA they may have also been broader social and political factors. Like the rest of Australia, Victoria was hit hard by the Depression. Like the Commonwealth and other state governments, the Victorian Government struggled to find economic and social policies to combat it. At a conference of Commonwealth and state ministers held in Melbourne between 18 and 21 August 1930, Sir Otto Niemeyer, the Bank of England's emissary, presented his prescription for Australia's economic and financial policy (Schedvin, 1970, pp. 179-184). In 1931, the Commonwealth Labor government, led by James Scullin, introduced the Premiers' Plan. This was a series of financial measures to combat the Depression broadly built on Niemeyer's prescription of spending cuts and financial stringency. This plan led to a major split within the Labor Party. Hogan, as the Victorian premier, supported it, but it was opposed by the union movement and much of the ALP. In early 1932, the Victorian central executive instructed Labor parliamentarians not to support the plan. In April, the Hogan government fell when three independent members who supported the Premiers' Plan joined the opposition in a vote of no confidence (Louis, 1968, p. 193). In the May 1932 election, the ALP was defeated, its numbers in the Legislative Assembly being reduced from 30 to 16, and replaced by a coalition government led by Sir Stanley Argyle. Hogan and several others were expelled from the ALP in July. The Argyle government did not appoint members to the Committee of Public Accounts. There was no announcement or foreshadowing, although later it was reported that the committee had not been reappointed for reasons of economy (ABS (Australian Bureau of Statistics), 1980, p. 93).

It was clear that during the final part of the Hogan government the factional fighting within the Labor Party meant that the premier retained little control over the work of the VCPA. This could partially explain the open criticism expressed in some of the late reports. At this time, party discipline and cohesion were weak, coalitions were common, governments with a parliamentary majority were uncommon and the lifespan of governments was often short (Strangio and Costar, 2006, p. 6). This does not explain why a VCPA committee was not appointed under the Argyle government. Clearly, one fear was that the open criticism of government policy seen under the Hogan VCPA might be continued. The fact that the new government would avoid a powerful source of public criticism of its policies and actions in failing to establish the VCPA would have been attractive. However, the failure to reappoint a VCPA also provided a socially acceptable response to the radical cost-cutting agenda promoted by the Premiers' Plan and the Niemeyer agenda. Cutting parliamentary spending rather than public services would have led to public support and legitimacy, not least because it was a visible way for the parliament to be seen to be subject to the same austerity measures experienced by the public.

The demise of the JCPA

Like the VCPA, the JCPA had also moved away from a focus on financial performance to an engagement with politics and policy. In the last session of its existence, it completed six reports. The first, tabled in December 1929, was about agricultural and pastoral leases in what was then the Federal Capital Territory (Joint Committee of Public Accounts, 1929). The inquiry had come about because of a deputation from the Rural Lessees' Association in the Federal Capital Territory to the Minister for Home Affairs which expressed concerns about the shortness of tenure, cost of improvements and size of holdings in the territory. The JCPA made specific policy recommendations, such as the establishment of a land advisory board and financial assistance for extending leases and purchasing fertilisers.

The second report, in April 1930, related to a claim against the War Service Homes Commissioner (Joint Committee of Public Accounts, 1930) by Henry Dean and Son. While Henry Dean and Son initially won their court case, they lost on appeal. They then contacted the Prime Minister, James Scullin, who referred this to the chair of the JCPA, requesting that the committee consider the company's “moral claim for compensation” (Joint Committee of Public Accounts, 1930, p. 5). The committee did not support the compensation claim. Nonetheless, this report seemed unrelated to the normal work of the JCPA, and there was no provision in the legislation for the Prime Minister to refer matters to the committee.

Between 1930 and 1931, there were a number of inquiries and reports produced dealing with financial problems associated within and between the states. The first was an inquiry into Tasmania's financial disabilities requested by the Prime Minister in 1929 after consideration by Cabinet. The committee visited Tasmania, conducted an inquiry and reported in August 1930. In September, the Acting Prime Minister, James Fenton, wrote to the committee requesting it to inquire into the financial position of South Australia (Joint Committee of Public Accounts, 1931, p. 5). The committee made several visits to South Australia. On 14 March 1931, the Prime Minister informed the committee that the Government had decided to change the terms of reference. He asked the committee to simultaneously review requests for “financial assistance to mitigate the effects of Federation” from South Australia, Tasmania and Western Australia. The JCPA was being used as a buffer to distance the Government from the need to address these deeply controversial resource allocation issues. While the JCPA agreed to carry out the wishes of the Prime Minister, it did note that the task was virtually impossible (Joint Committee of Public Accounts, 1931, p. 5). However, they did not object to the fact that the request had originated from the Prime Minister.

The sixth report, tabled on 8 August 1930, was not a report of the committee but was the report of a royal commission into allegations that members of the committee had been bribed (Royal Commission, 1930). The Royal Commissioner, George Dethbridge, rejected the allegations that the JCPA chairman, Percy Coleman, or any other committee member had been paid a bribe. However, it is difficult to underestimate the damage that these allegations would have made to the perceived credibility of the JCPA as “honest brokers”. They would have substantially undermined the usefulness of the JCPA as a tool for public legitimacy. In addition, it was clear that the committee, while not openly challenging government policy, as in the case of the VCPA, had substantially departed from a focus on financial probity to a direct engagement with policy and politics. The tasks given to the JCPA by the Prime Minister led to public criticisms. The committee had come under public attack for wastefulness when it visited Hobart in 1930 (NAA: A12840, 2004/2002), and there were some within parliament who argued that committees such as the JCPA were extravagances that could and should be curbed.

After the agreement on the Premiers' Plan in July 1931, the Federal Parliament debated the budget estimates and looked at ways in which parliamentary expenditure might be reduced (APD (Australian Parliamentary Debates), 1913, pp. 4573-4593). Members proposed reductions in expenditure on the parliamentary refreshment rooms, the heating of Parliament House and the suspension of the Public Works Committee and the Committee of Public Accounts. Parliamentary salaries and allowances were already being reduced as part of the general reduction in government salaries (Jones, 2007). Discernible in the debate was the feeling that parliament should be seen to be contributing to the general sacrifice; if old age pensions are to be reduced, so must parliament's trappings of office (APD (Australian Parliamentary Debates), 1913, pp. 4573-4593). Nonetheless, the central focus of the debate was on the Public Works Committee and the JCPA. The foremost opponents of these committees were Robert Archdale Parkhill, the Nationalist member for the Sydney electorate of Warringah, and Joel Gabb, a former ALP member. Both these men were strongly committed to cutting budget spending and both argued that the JCPA should be suspended. Other members and senators did not support the committee's suspension but agreed that its expenditure should be reduced by reducing the number of members or cutting staff

In October 1931, following the estimates debate, Scullin introduced two bills to reduce the number of members of the Committee of Public Accounts and the Public Works Committee from nine to seven (Holmes, 1966, pp. 35-36). Parkhill supported the abolition of both committees, which he described as “relics of the antiquated system of ‘perks’ for parliamentarians” (APD, 1931a, p. 658). Even a former member of the JCPA, Senator Foll, complained that it had “roamed practically all over the Commonwealth, investigating matters which ought not to have been referred to such a body” (APD, 1931b, p. 837). The Scullin Government fell at the end of 1931 (November 1931) and the bill lapsed.

At the December 1931 election, the Scullin Labor government was replaced by the United Australia Party led by Joseph Lyons, which included former Nationalist, independent and Labor members. The new parliament, which first met in February 1932, did not appoint any members to the Committee of Public Accounts. The Governor-General, in his speech at the opening of parliament on 17 February 1932, announced that the Government would suspend the operations of both the Committee of Public Accounts and the Public Works Committees (APD, 1932, p. 9). It introduced legislation to do so in October 1932. Parkhill became a minister and was a powerful parliamentary and political figure within the United Australia Party. Gabb sat as an independent, although he supported the Government and maintained a reputation for his opposition to perceived extravagance.

At this time, what little legitimacy the JCPA provided was weakened by accusations of bribery and waste. Even those involved acknowledged that the work the JCPA was being asked to do had little resemblance to classical functions of a public accounts committee and that the ongoing use of the committee by the Prime Minister for political ends had undermined it. In this context, the increasing perception of the JCPA as a political perk and Parkhill's arguments for its abolition made the suspension virtually inevitable. Given the financial austerity of the Depression, it is possible that the disestablishment of the JCPA and the public works committee resulted in generating legitimacy for the incoming Lyons government.

Conclusion

In this paper we have adopted an institutional approach to the study of parliamentary oversight and governance. While authors have highlighted the need to adopt an institutional approach to the understanding of audit, regulation and governance, little work has been done that takes seriously the concept that these institutions are established for legitimacy rather than functionalist purposes. Both Hood (1991, 1995) and Power (1997) have argued that these institutions of oversight and audit have come to play an increasingly powerful role in contemporary society. We have argued that in order to understand this role, we need to more fully understand the driver for their establishment.

We adopted the work of Degeling et al. (1996) as a starting point for our own study. Degeling et al. (1996) studied the work of the Australian federal parliamentary public accounts committee (JCPA) from its establishment in 1913 until its suspension in 1932. Degeling et al. (1996) argued that it was likely that the financial focus of this committee would colonise other aspects but found that, although the initial mandate was focused on financial accountability, the work of the committee shifted to a focus on issues of policy and politics. Instead of Degeling et al.'s (1996) focus on accountability and colonising, we adopted an institutional theoretical focus more consistent with the activities and changing nature of these committees. In particular, we believe that this enables us to more fully explain the establishment of the JCPA and the changing nature of its work. By way of contrast, we discussed the Victorian Committee of Public Accounts (VCPA), which pre-existed the JCPA by nearly 20 years and was openly acknowledged in parliamentary documents as the template and example for the JCPA mandate and structure.

By more fully exploring the establishment of the earlier VCPA and the subsequent JCPA, we found that the primary objective for the establishment of both these committees was legitimation. Therefore, the early functional mandate was not of critical importance and the subsequent deviation from it to focus on broader issues of policy and politics is more understandable. The central element of the legitimating power was that both these institutions were seen as an accurate copy of existing British practice as reflected in the Westminster parliament. However, from records and documents it seems clear that the Victorian practice, as reflected in the VCPA, presented a more proximate mimetic source for the establishment of the JCPA.

The work of both these committees diverged over time from the mandates given to them in their founding documents as they drifted from a focus on accounting to a focus on policy. Degeling et al. (1996) have argued that this represents a colonisation of accounting by politics. Implicit in their analysis is the belief that the establishment of the JCPA was driven by a concern for the function of the committee – that is, the content of its reports and the recommendations within them. We have argued that the purpose of both the JCPA and the VCPA was to provide legitimacy for the Governments at the time of their establishment. The VCPA was established in 1895 at a period of policy transfer from Britain to its colonies and the JCPA in 1913 when policy transfer was beginning to occur within Australia (Carroll, 2006).

Following from Degeling et al.'s (1996) methodology, we analysed the work of the VCPA between 1895 and 1931 and established that it had followed a similar path, with a transition from a focus on financial matters to a focus on politics and policy. We also established that the content of the reports appeared to have little impact on the existence or status of the committees. In other words, the Governments that had facilitated the establishment of the committees appeared not to care that they were ignoring their designed purpose. To further follow up this phenomenon, we analysed the events surrounding the demise of the two committees, including the last reports they produced. We found that the VCPA was openly critical of government and was actively involved in policy formation. The JCPA was also directly involved in policy formation, although this was in cooperation with the Government rather than in criticism of it. However, the JCPA also faced two other major challenges. Its members were accused of taking bribes and it was publicly criticised for profligacy. In addition, it was directed by the Prime Minister to undertake major economic inquiries which were both beyond its expertise and inappropriate in terms of its mandate.

In the political crisis that emerged from the Depression, both the VCPA and the JCPA became perceived as unnecessary extravagances. Political capital and corresponding public legitimacy could be generated by abolishing them. In both cases, they were abolished by conservative governments that replaced Labor governments that had failed to control the financial crisis and that had been torn apart by internal dissension. In the abolition of the VCPA and the JCPA the new governments could be seen to exercise the same economies that they proposed for the public. Therefore in 1932, the acquisition of legitimacy for the Governments (and to some extent the parliaments) lay in the abolition of the VCPA and the JCPA, just as in earlier years it had resided in their creation.

Degeling et al. (1996) conclude that the work of the JCPA shows little evidence of instrumental purpose or violation. But rather than such organisations and structures are inevitably a product of social and political processes. While we agree with this conclusion we would argue that the idea of social and political legitimacy provide a powerful way to explain and understand these particular social and political processes. the case of the VCPA and JCPA provides an important insight into processes of institutional change and legitimacy. While the establishment of these committees delivered a level of public legitimacy, particularly due to the fact that they were visibly copied from the UK model, the disestablishment of the same committees also delivered legitimacy. The absence of an established functional role did not seem to be particularly problematic. It is important to recognise that while mimetic and other forms of isomorphism can deliver a level of legitimacy when an institution is established, this legitimacy is not necessarily sustained over time. In this case, the broader social focus on austerity associated with the Depression and the urgency to cut, and be seen to cut, expenditure meant that the disestablishment of these committees generated legitimacy rather than their ongoing existence or work.

We suggest that these insights provide a radically different perspective on the contemporary process of financial oversight, governance and the audit society. It is important to separate the functional perspective on these institutions from an institutional/legitimacy one. We cannot take for granted, as has been in much of the NPM literature, that institutions such as public accountants committees play a functional role. Clearly the advocates for a public accounts committee in Victoria understood that the function of the VCPA was to provide conventional financial oversight. Indeed, these ideals are still expressed by those who describe how these committees should operate now (McGee, 2002). While these norms are clearly desirable, it is also evident that over time the nature of the work of the VCPA (and the JCPA) more clearly reflected political aims and influences. It showed that in practice it was not possible to isolate the practice of financial oversight and accounting measurement from the intensely political environment in which it functioned and the need to generate legitimacy in that setting. In that sense, Hopwood's (1984) point that accounting language and practices have become intertwined in substantive issues of policy may have been true for a much longer time than suggested by Hopwood's (1984) analysis and that drawing a neat distinction between politics and accounting within this setting becomes meaningless. While it is not necessary for accounting to dominate politics, it is interesting that the perception of an accounting focus provided such an important source of legitimacy. It may prove that the growth of practices of governmentality and oversight described by Power (1997) are not principally associated with functional activities but are primarily driven by the need to create and sustain the legitimacy of broader institutions and structures, such as parliament and democracy and that this may have been happening for a far longer time than authors such as Power (1997) and Hopwood (1984) have previously indicated.

From a contemporary perspective, this narrative provides an important insight into the power of mimetic isomorphism and the legitimacy power of the British Westminster models and practices. It also seems clear that once a PAC has been institutionalised, it will rarely be challenged to justify its current actions and functions. Recent reviews of the Australian PACs illustrate that considerable diversity is evident in practice (KPMG, 2006; Jones and Jacobs, 2006). After all, these institutions functioned for 26 and 18 years respectively. It is important to note that these committees were subsequently re-established in the 1950s after calls for their re-establishment from both inside and outside parliament.

References

ABS (Australian Bureau of Statistics) (1980), Victorian Year Book, Government Printer, Melbourne, .