WPS3613 Scrutinizing Public Expenditures Assessing the Performance of Public Accounts Committees By Rick Stapenhurst1 Vinod Sahgal2 William Woodley3 and Riccardo Pelizzo4 World Bank Policy Research Working Paper 3613, May 2005 The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the view of the World Bank, its Executive Directors, or the countries they represent. Policy Research Working Papers are available online at http://econ.worldbank.org. 1Senior Public Sector Management Specialist, World Bank Institute 2Principal Financial Management Specialist, World Bank 3Senior Consultant, World Bank Institute 4Assistant Professor, Singapore Management University Abstract In 2002, David McGee wrote a sent to 51 national and state/ provincial comprehensive report on two important parliaments in Commonwealth countries elements in the system of public financial in Asia and Australasia and Canada and accountability, namely the office of the the United Kingdom. We find that the Auditor General and the parliamentary institutional factors which most account oversight committee commonly referred to for the success of the PACs are the focus as the Public Accounts Committee (PAC). on government's financial activity rather The purpose of the present paper is to than its policies, the power to investigate deepen McGee's analysis of PACs. In all past and present government expenses, particular, we define PAC success and the power to follow up on government identify those factors that impact on PAC action in response to its recommendations, peformance. We use data that were and its relationship with the Auditors collected by the World Bank Institute in General. 2002, when a survey questionnaire was Contents 1. Introduction............................................................................................................. 1 2. Section 1: Legislatures and Public Financial Accountability ................................. 2 3. Section 2: Survey Results ­ Overview.................................................................... 4 5. Section 3: Success Factors for PACs...................................................................... 7 6. Section 4: Benchmarks for Measuring PAC Performance?.................................. 21 7. Conclusions :......................................................................................................... 24 Annexes 1. Annex 1 : Main Conclusions from McGee book, The Overseers......................... 26 2. Annex 3 : Survey Instrument................................................................................ 28 3. Annex 3: Full Listing of Success Factors ............................................................ 33 4. Annex 4: List of Respondents to the Survey ........................................................ 35 5. References............................................................................................................. 36 Introduction In 2002, David McGee wrote a comprehensive report on two important elements in the system of public financial accountability, namely the office of the Auditor General and the parliamentary oversight committee commonly referred to as the Public Accounts Committee (PAC)5. Based on the deliberations of a Study Group organized by the Commonwealth Parliamentary Association (CPA)6, he examined current practice across the Commonwealth regarding the roles and functions of Auditors General and PACs and the interaction between these two institutions. The purpose of the Study Group was "to assess how PACs are working in practice and whether they are fulfilling expectations as important guarantors of good governance. The group ... tried to distil the practices ... into checklists of considerations for Parliaments, always recognizing that social, economic and political factors within each country mean that there is no one organizational form that a PAC will take. The group... also discussed and tried to define a modern rationale for PACs ...[which] is necessitated by and is based on the international dimension that results from a greater inter-dependability of nations today ... [which] ... has brought in its wake greater challenges to good (and honest) government."7 McGee sought to identify possible courses of action to improve outcomes through more effective use of PACs. Three main priorities were identified: · Capacity building ­ the need to improve the ability of Parliaments and their PACs to carry out their functions by being provided with adequate resources, training and access to relevant expertise · Independence ­ that they be free from political or legal constraints that could inhibit them from carrying out their duties diligently8 · Information exchange ­ that PACs have the means to exchange information and ideas so as to keep them up-to-date with important developments, changing standards and best practices as they emerge9. The individual conclusions and recommendations of the Study Group are summarized in Annex 1. The purpose of the present paper is to deepen McGee's analysis of PACs. In particular, we seek to attempt to define successful PAC performance and to identify those factors which facilitate or hinder such successful performance. We use data that were collected by the World Bank Institute (WBI)10 in 2002, when a survey questionnaire was sent to 52 national and state/provincial parliaments in Commonwealth countries in Asia and Australasia and 5McGee, David : The Overseers ­ Public Accounts Committees and Public Spending, Pluto Press, London 2002. 6and supported by WBI 7op cit, p.4-5; from the Study Group organized by the Commonwealth Parliamentary Association 8this priority was noted to be particularly important for Auditors General 9op. cit, p. 6 10In collaboration with the World Bank's South Asia Financial Management group 1 Canada and the United Kingdom. Thirty-five responses were received of which 33 were usable, a response rate of 67 percent. Supplementary interviews (both in person and on the telephone), were conducted to seek further insight and clarification to the survey results. In the first section of this paper, after providing a summary of the role of legislatures in financial oversight, we present a general concept of public financial accountability. We underline the critical role that PACs play, noting their origin in the 19th century and their widespread use throughout the Commonwealth and elsewhere. Through the data collected by the WBI, we are able to deepen our knowledge about how PACs work and, in particular, what ­ in the views of the chairs of these committees ­ constitutes successful performance and what factors influence such performance. The second section of the paper presents the survey instrument and summarizes the broad findings of the survey. Section three investigates the success factors (and constraints) for PAC effectiveness, based on the survey results. In section four, we identify possible benchmarks for measuring the effectiveness of PACs, and discuss how developing countries may be able to think about strengthening their PACs. In the fifth and final section of the paper, we draw some conclusions and propose what an ideal PAC might be like. Section 1: Legislatures and Public Financial Accountability i) Role of legislatures Legislatures perform three functions - representative, legislative, and oversight. They perform a representation function in that they represent the will of the people, which is the legitimate source of authority in democratic countries. They perform a legislative function because, in addition to introducing legislation on their own, they have the power to amend, approve or reject government bills. And they perform an oversight function, ensuring that governments implement policies and programs in accordance with the wishes and intent of the legislature. They undertake this oversight function in two ways: they oversee the preparation of a given policy (ex ante oversight) or can oversee the execution and the implementation of a given policy (ex post oversight). Though most legislatures have the power to keep the government accountable for its actions and its policies, there is considerable variation in the legislative tools that legislatures can employ to perform their oversight function. This variation reflects to a large extent differences in the form of government and other constitutional arrangements. These means include parliamentary committees, questions in the legislature, interrogations, urgent debates, the estimates process, scrutiny of delegated legislation, private members' motions and adjournment debates that allow legislators to raise issues relating to the use or proposed use 2 of governmental power, to call upon the government to explain actions it has taken and to require it to defend and justify its policies or administrative decisions11. One of the tools that a legislature can use to further enhance oversight of the financial operations of government is a specialized committee. In the "Westminster model" of democracy12, the committee is known as the Public Accounts Committee, or PAC; it is the "audit committee" of Parliament and as such is the core institution of public financial accountability.13 As Frantzich pointed out more than two decades ago, legislatures need useful information to perform their representative, legislative and oversight function effectively14. So do PACs. This information is generally provided by the legislative auditor, or Auditor General. The auditor reports to the legislature and the public at large on whether public sector resources are appropriately managed and accounted for by the executive government. ii) Concept of Public Financial Accountability Following implementation of a government's budget, a legislative auditor audits government accounts, financial statements, and operations. In most countries, this audit is followed by the consideration of the audit findings ­ which may include value for money and performance auditing as well as financial or compliance auditing ­ by the legislature. If the legislatures' role in the budget process is effective, legislative recommendations to the executive based on the deliberation on audit findings put forward by the auditor are reflected in future budgets, thus allowing for continuous improvements in public financial accountability. The exact nature of the relationship and interaction between the legislature and the auditor partly depends on the model of the legislative auditor and the reporting relationship to the legislature. In most Commonwealth countries, the legislative auditor is the Auditor General, whose office is a core element of parliamentary oversight; he/ she reports directly to Parliament and the Public Accounts Committee. In some instances, the Auditor General is an officer of parliament, which guarantees his/ her independence from the Executive (as in the case of Australia and the United Kingdom), while in some other instances, he/ she is independent of both the executive and the legislature (as in the case of India). 11See Pelizzo, Riccardo and Rick Stapenhurst Tools of Legislative Oversight Policy Research Working Paper # 3388, The World Bank, September 2004 12 The expression, "Westminster model of democracy' was developed by political scientist Arend Lijphart. According to Lijphart "in this book I use the term Westminster model interchangeably with majoritarian mode to refer to a general model of democracy." See Arend Lijphart, Patterns of Democracy, New Haven, Yale University Press, 1999, p.9. This model of democracy is defined by: concentration of executive power in one- party and bare majority cabinets, cabinets dominance, two-party system, majoritarian and disproportional systems of elections, interest groups pluralism, unitary and centralized government, concentration of legislative power in a unicameral legislature, constitutional flexibility, absence of judicial review, a central bank controlled by the executive. 13In some countries such as India and Sri Lanka the Public Undertakings Committee co exists with the PAC to cover the oversight of autonomous public enterprises 14 Stephen E. Frantzich, "Computerized Information Technology in the US House of Representatives", Legislative Studies Quarterly, vol. 4, n. 2, 1979, pp. 255-280. See also Robert Miller, Riccardo Pelizzo and Frederick Stapenhurst, "Parliamentary Libraries, Institutes and Offices. The Sources of Parliamentary Information", WBI Working Paper, 2004. 3 The structure and function of the PAC date back to the reforms initiated by William Gladstone, when he was British Chancellor of the Exchequer in the mid-19th century. The first PAC was established in 1861 by a resolution of the British House of Commons. Replicated in virtually all Commonwealth and many non-Commonwealth countries, PACs are seen as the apex for financial scrutiny and have been promoted as a crucial mechanism to facilitate transparency in government financial operations (see Diagram 1). Diagram 1 Across the Commonwealth, however, there is considerable variation in PACs' terms of FIDUCIARY OBLIGATION reference and modus operandi. In some Parliament instances, for example, the terms of reference are narrowly defined; in these cases PACs concentrate exclusively on Report financial probity. In other instances, the Mandate terms of reference are more widely defined; Public ConferredResponsibility Accountability Reporting here, the committee does not simply focus on financial probity but also on the efficiency and effectiveness of programs in Watchdogs Examine Executive achieving the objectives for which they had been established. (Indeed, our findings indicate that this "scope of work" is one of the principal factors affecting PAC performance.) There is considerable variation, too, regarding the relationship between the Auditor General and the PAC, the status of the PAC within Parliament, how the PAC conducts its business, on PAC reporting to the legislature and on requirements for government follow-up on PAC recommendations. These and related issues are examined in detail in McGee (2002)15 An important feature in virtually all jurisdictions is the fact that PACs do not question the desirability of a particular policy ­ that is the mandate of parliamentary departmental committees; rather, PACs examine the efficiency and effectiveness in the implementation of policy. But if McGee outlined how PACs work, in an effort to identify good practices across the Commonwealth, he did not consider, in any detail, the reasons behind the good practice -- what might be called the critical success factors: why are some PACs considered more successful, while others are not? Nor did he, in the less successful cases, identify constraints hindering performance. This paper explores such issues in some depth. Section 2: Survey Results ­ Overview In the survey undertaken by WBI, PAC Chairs were asked to self-assess the impact of the Committee's work; the responses are summarized in Table 1. It would appear that PACs are most successful acting as a catalyst for improvement in government's implementation of policy decisions and for improvement in the availability of government information to 15op cit 4 parliament. In 78.8 percent of cases, PAC chairs state that Committee recommendations are accepted by government, while in 63.6 percent of cases, governments implement such recommendations. Typical, perhaps, is India, where, over the period 1980 through 1999, the government accepted 60.68 percent of recommendations made by the PAC16. Similarly, in 60.8 percent of cases, PAC recommendations have improved the availability of information to parliament. However, the findings show that PACs are less frequently the catalyst for the government to change legislation or major policy objectives or to prosecute officials who break the norms of probity or present misleading financial information to the public. One notable exception is Uganda. Since the mid-1990s, the Ugandan PAC and its sister committee, the Committee on Commissions, Statutory Authorities and State Enterprises, have worked in close collaboration with police CID officers ­ so that cases of financial wrong-doing that emerge from committee hearings lead directly to police investigations and many times to court cases17. Table 1. Results achieved by the PAC: how frequently has the PAC achieved the following result? Percentages Result achieved Frequently Rarely of N Recommendations accepted 78.8 15.2 33 Recommendations implemented 63.6 27.3 33 Better information 60.8 18.2 33 Disciplinary action 27.3 15.2 33 Modification of legislation 15.2 54.5 33 PAC Chairs were also asked what powers and practices they considered important in achieving results. When given a list of 37 potential success factors, most respondents considered most of the suggested factors to be "very important" (see Annex 3 for the full listing of the 37 factors). However, when asked what the top three factors critical for success were, a consensus began to emerge, especially when Chairs were subsequently asked to identify important powers or practices that their committee did not have or follow but which would be useful to have. The responses are summarized in Tables 2 and 3. Table 2 shows that there was substantial agreement among respondents concerning the first four powers ­ namely the power to formulate recommendations and publish the conclusions, to investigate all past and present expenses, to choose topics without government interference and to focus on financial probity rather than policy issues. For all these powers, more than 90 percent of Chairs rated them as "very important". There was also considerable, although lesser, agreement among respondents on the next four powers ­ to call witnesses, to examine the budget of the legislative auditor, to require cabinet ministers to appear before the committee and to view any proposed legislation affecting the legislative auditor. Here, at 16G.C. Malhotra "Ensuring Executive Accountability : India's Public Accounts Committee", The Parliamentarian April 2000, pp 179-186 17A Member's Handbook Report Based on a Parliamentary Workshop in Entebbe with the World Bank Institue and the Parliamentary Centre of Canada, , 1999 5 least 70 percent of respondents thought these powers to be either "very important" or "important" ­ although a significant minority of respondents considered these powers to be "not important". Table 2. Success and Powers of PACs: How Important is this factor? Percentages Very Important Not N important important Formulate recommendations and 97 3 0 33 publish the conclusions Investigate all past and present public 93.5 6.5 0 31 expenditures Choose topics for investigation 90.9 9.1 0 33 without government interference Focus on financial accountability 90.9 9.1 0 33 rather than on policy Require witnesses to answer questions 87.1 6.45 6.45 33 Examine the budget of the Legislative 58.8 35.3 23.5 17 Auditor Require cabinet ministers to appear 55 15 30 33 before the committee View proposed legislation or the 47.8 30.4 21.8 23 amendments to the Legislative Auditor's Act Regarding composition of the committee, only two of the 37 factors mentioned in the list given to the respondents belong to this category. The first factor is the `balanced representation of all major political parties in the commission", while the second factor is "exclusion of government members from the commission". There was again a high degree of consensus among respondents, with more than 85 percent reporting these factors as "very important" (and 100 percent believing that the exclusion of MPs with cabinet posts a "very important" or "important" factor). Table 3. Success and Composition of the PAC: How important is this factor? Percentages Very Important Not N Important important Proportional representation of the 86.2 10.3 3.4 29 various parliamentary parties Exclusion of MPs with cabinet posts 85.2 14.8 0 27 And finally, with regard to committee practices and procedures, there was again a fairly strong consensus among respondents regarding success factors. As Table 4 indicates, these include the keeping of transcripts of meetings, adequate preparation before committee 6 meetings, follow-up procedures on government action regarding committee recommendations and close working relations between committee members of different political parties. Table 4. Are Practices and Procedures important for the Success of a PAC? Percentages Very Important Not N important important Keeping the transcripts of the 87.9 9.1 3.0 33 meetings Preparation before Committee 78.8 18.2 3.0 33 Meetings Procedures to determine whether 75.0 18.7 6.3 32 the government has taken any step to implement the recommendations of the Committee Close working relationship 75.0 15.6 9.4 32 between the members of the various political parties Section 3: Success Factors for PACs Discussed below are the top success factors and best practices, including some comments about why they are important and how they help achieve results for the committees that suggested them. Also discussed are some reasons why these success factors might not be present, or why they don't work well in certain contexts. We have incorporated related success factors together, where this is appropriate. 1. Broad Scope: The power to investigate or review all past, current and committed expenditures of government18 The power to oversee all public expenditure provides the PAC with a broad mandate and target audience. It gives the PAC a vantage point with respect to a wide range of public services, and an entry point into the administration of many departments and agencies even if other parliamentary committees are involved in their policy aspects. The wider the PAC mandate, the greater it's potential to deter waste and wrongdoing and encourage better management of public resources. Most committees have this power, although some cannot examine committed expenditures; most can also examine revenue administration. But sometimes the full range of public 18Reasons given for selecting this success factor included: "Be perceived as the principal government watch- dog;" "To get details and the rationale behind government decisions;" "Keeps a check on administration being slack, arbitrary;" "Monitor fiscal discipline;", and "Effective check on finances." 7 functions is divided with other committees, such as a committee for public enterprises or undertakings. Other possible scope limitations could restrict the committee's investigations of important institutions such as the central bank, or important fields of public administration like statutory expenditures that are not voted each year, or tax expenditures. In India, for example, the central government PAC is unable to probe into the workings of financial institutions or the central bank or the accounting policies and standards of government financial reporting, or the adequacy of existing auditing standards in the public sector. Even with a broad mandate, there are situations that could nullify or reduce PAC effectiveness. Some supporting factors that would help to make the best use of a broad mandate are: Having a specific permanent reference to examine the public accounts and legislative auditor's reports; Legislative audit reports that get at the causes of problems rather than just recounting the problems themselves; Timely publication of the public accounts and legislative auditor's reports; Co-operation of all committee members especially those from the governing party - a willingness to both criticize the government's performance, and to help it perform better; Clear demarcation of responsibilities with other legislative committees; Good support from the legislative auditor (but not too much dependence); Effective planning and priority-setting process in the committee; Adequate briefing, support, resources for identifying important issues. All these further conditions are important. For example, it is common, and important, to have a permanent reference to examine the public accounts and legislative audit reports. Through their examination of legislative audit reports, many PACs have the opportunity to examine questions of economy, efficiency and effectiveness, depending, of course, on the type of information provided by the legislative auditor and the attitude of the committee. Some committees are more comfortable dealing with instances of fraud, waste and abuse with brief audit notes about problems, rather than with broad value for money issues such as discussing systems weaknesses, policy specifications, legislative loopholes or cultural factors that caused the problems in the first place. The preferences of the Committee Chair may be a contributing factor. Another problem noted was that sometimes the tabling of the accounts and audit reports are delayed, leaving the PAC to deal with old issues or without work until the documents are available; similarly, there are times when the heavy workload of old reports delays getting to the most current material. While some committees face a serious backlog of work,19 others have found ways to deal with this problem. Here are two examples: 19We know of several committees that did not respond to our survey and are not up to date with audit reports; for example one has not yet started to review the audit reports of 1998-99 which were submitted to the legislature in April 2000. Some are even further behind. Some are waiting for work because the audit reports are late. One could speculate that embarrassment about this situation may have been a factor in not responding to the survey. 8 "[In the past] the practice of the committee was to take [Comptroller and Auditor General] CAG reports on a first-in first-out basis ...the reports of the CAG kept accumulating, resulting in a lackadaisical attitude on the part of government officials. Therefore, to improve the state of affairs ...the Committee now takes up recent reports, and accords paragraphs ...different priorities." "The Committee was able to finish a year's work in less than a year through regular meetings." Clearly a move towards last in first out (LIFO) may be a preferred option to the conventional first in first out (FIFO) method of arranging deliberation of audit reports. Some terms of reference state or imply a detailed examination of the accounts, or a concentration on compliance rather than value for money20. Case 1 UNITED KINGDOM The United Kingdom PAC has 10 government and 6 opposition members with an opposition member as chair. It was established in 1861. It issues an average 10-20 reports a year based on the reports of the National Audit Office (NAO). The PAC sometimes looks beyond the audit reports to root causes of problems. For example, in 1994 it issued a groundbreaking report called "the Proper Conduct of Public Business" that provided a broad checklist of improvements to help departments avoid the failings in financial administration that had brought them before the auditors and the PAC in the past. In 1999, the PAC asked the NAO to examine the Committee's impact. The NAO agreed, and provided a detailed, frank assessment for the Committee's private information. The review identified and assessed three main aspects of the PAC's impact: The government's implementation of the Committee's recommendations; The views of senior public servants who were most often witnesses before the Committee and whose departments were the subject of its reports; and The quality of press coverage of PAC hearings and reports, as measured in various ways including coverage by leading journalists. The assessment reviewed some known examples of PAC reports that had been implemented, including a recent follow-up report that convinced the government to abolish an agency. This case illustrates a very close relationship between a PAC and a legislative auditor. Both have achieved useful results, some of which can be quantified. Although the review of impact was not an independent study of the Committee's effectiveness, it breaks new ground in the measurement of PAC performance. 20 For example, Singapore and India, although in the case of India the PAC's mandate extends beyond compliance with rules and regulations to control over expenditure, its probity, wisdom, faithfulness and economy. 9 Some committees interpret a permanent reference of documents as meaning they must systematically review all aspects of all reports, rather than selecting the most important ones. Others take the reference of these documents as a limitation that they cannot examine anything else. But most set priorities and select important matters within their scope to examine, sometimes taking a creative look at root problems (see Case 1)21. 2. Free Choice: The Power to Choose Subjects for Examination without Government Direction and Advice22 This power provides the PAC with the freedom of action to focus on relevant, important matters. Having a broad scope mandate is of little use if the committee is unable to identify major issues or if the government directs the committee away from these. PAC Chairs reported that they are most effective if they can initiate enquiries on their own as necessary, without having to wait for a specific reference from the government.23 The elements of speed and even of surprise are often important here: if the committee can act quickly to select incisive issues, and if the choice and timing cannot be predicted by the government and the public service, the committee's enquiries can have a deterrence effect. Some committees focus primarily on specific audit notes or individual transactions uncovered by the legislative auditor. Relevant, important subjects would certainly include looking beyond individual grants, contracts and other transactions for the larger system weaknesses that make the individual problems inevitable. Committees should factor in the likelihood and practicality of the government taking corrective action on the subjects addressed. Consideration should also be given to whether the PAC is the most effective vehicle for dealing with an issue or whether other committees or institutions could not have dealt it with adequately. The selection of important, relevant issues depends in turn on a number of supporting factors, some of which were identified and supported in the survey: Co-operation of all parties on the committee Clear focus on accountability, rather than on policy Effective research staff with knowledge of public administration and accountability Members with relevant experience Clear vision of improvements needed in public administration Effective chairing 21British Columbia (Canada) is another good example of a committee that works with their legislative auditor to identify important matters for follow up. 22 Reasons given for selecting this success factor are: "Helped the ad-hoc committee work independently without government interference;" "Issue rigorous reports and decide on own agendas, inquiries;" "Select matters of concern and hold government accountable;" "Maintain independence;" "Undertake enquiries in areas of concern;" and, "Allows PAC to scrutinize all areas of government accountability." 23New South Wales is an example - much more activity and greater effectiveness once its mandate was changed 10 If these conditions are missing, for example if there is a highly partisan atmosphere, the committee's ability to choose subjects freely will be reduced. With a government majority on most committees, some restrictions in selection are inevitable. Having a clear focus on accountability is one way to instill a unity of purpose among different parties. Almost all PACs surveyed thought this was very important. For example, if all committee members can agree with the objective of holding the government accountable for spending, and for the stewardship of public assets, then the committee can focus on implementation of government programs rather than policy matters, and focus on the accountability of civil servants for administrative and financial operations rather than on the political accountability of ministers. This will help at all stages of a committee's work - selecting issues, carrying out an enquiry, and drafting a report. If all parties in the committee agree on a clear focus at the beginning of an enquiry, the chair will have a much easier time keeping the committee on track towards a useful conclusion. 3. Effective Analysis and Reporting: The power to make recommendations and publish conclusions24 Virtually all committees value the power to report publicly. Most committees produce specific reports, with recommendations at the end of each enquiry, although not all enquiries can be brought to a successful conclusion in a report. Some committees also produce an annual report and, further, may request a debate in the legislature. Some also require a response from the government within a set time. In many jurisdictions, the reports and recommendations follow closely those of the legislative auditor, and some25 depend on the auditor to follow up their recommendations when the work of the audit office itself is being followed up. Where both the PAC and legislative auditor make recommendations, close collaboration and support of each other's work is essential. A division of labor that seems to be effective is when the legislative auditor reports on the status of government administration and performance of civil servants and the PAC adds a broader perspective of the political economy to this information by identifying the root causes and practical solutions. The completion of a successful enquiry and report depends in turn on a number of conditions (some of which were identified in the survey as success factors): A focused enquiry with clear objectives (for example, in Australia the purpose of questioning witnesses in connection with a legislative audit report is not to analyze all the findings and recommendations in detail, but to seek action commitments that can be reported and then followed up) Effective drafting procedures for reports All party agreement on the recommendations, to increase the likelihood of acceptance 24Reasons given for selecting this success factor are: "Publish reports;" "Seek government response Ensures implementation and review; Allows committee to express itself; Forms basis for debate and follow-up; To ensure accountability to taxpayer 25United Kingdom, Canada, Ontario, British Columbia, and probably many others 11 Good research and analysis support from staff Effective follow-up procedures, to increase the likelihood of implementation. One of these additional factors in particular received strong support from PAC Chairs - effective follow-up procedures. Follow-up is closely related to the power to make recommendations and publish conclusions, because suggested improvements are of little use if they are not implemented. Surprisingly, two Chairs thought this was not important 26but most rated it "very important" to encourage the government to act, and to track the committee's record and therefore effectiveness (see Case 2). All these comments highlight the importance of a coherent process of follow-up and accountability to the committee for dealing with the matters raised in the relevant reports. 4. Solid Technical Support: From the legislative auditor and research staff The diligence and skill of committee investigations depends on many factors such as bipartisan co-operation (see below), good chairmanship, and the ability of individual members to probe reluctant witnesses. Underlying many of these is solid research support. Good advice and information is helpful in selecting important issues and pursuing them to a conclusion. The legislative auditor is usually the chief support for the Committee - the "friend, philosopher and guide"27 as one committee explained. In some cases, the legislative auditor sits with the committee28, or provides staff directly to the PAC or performs work otherwise done by committee staff. The committees in our survey had an average of less than two professional staff assigned full-time, but the range was wide, from 0 to 22. These numbers could include professional staff providing procedural advice and report drafting skills. We are not aware of any cases of research being contracted out, but this probably happens. Surprisingly, two of the survey respondents thought that relations with the legislative auditor was not important, while three thought the use of outside expertise and research support was not important. But most others gave these factors a high rating. Especially when a committee decides to tackle difficult issues, it needs skills and practices to obtain and analyze information and testimony. In some countries outside the Commonwealth, more of an adversarial relationship with witnesses is brought to bear either by the legislative auditor, who acts as a "court of accounts", or by committees with legal help and more judicial powers. In most Commonwealth PACs, it has been suggested that "fair play and decorum" usually prevail, in contrast, for example, to the life and death partisan battles in US congressional committees29. The need for legal training and help in cross-examining 26South Australia, Prince Edward Island 27See parliament of India web site description of the workings of the PAC. 28In India and Ontario, for example, the legislative auditor is allowed to ask questions of witnesses 29Teething the Watchdogs is a 1992 University of Strathclyde report of a panel discussion comparing US and United Kingdom legislative committees. See the quote from a United Kingdom official on p. 4: "I've seen senior officials from my old department the Home Office putting up very effective and courteous stone-walling performances in front of the Home Affairs Select Committee and, if it is courteously done, I think the select committee genuinely accepts that the officials are playing within the rules and it doesn't press them too hard for 12 witnesses seems to be less important in Commonwealth countries, although there are examples where it has been important30. Case 2 INDIA The Committee on Public Accounts (PAC) of the Government of India is composed of 14 government and 8 opposition members, with an opposition chair. Committee members are elected from both houses of Parliament. The PAC has been in existence since 1921 and is referred to as "the eyes and ears of the legislature". It generally produces more than 10 reports a year, and works very closely with the Comptroller and Auditor General, whose reports the PAC can examine even before they are tabled in the House (although it cannot report on them until they are tabled). The Committee keeps track of recommendations implemented by the Government - over 60% in the past 20 years, and more once the Committee follows up. An early example of an important accomplishment was the implementation of the PAC's 1947-48 recommendation to establish internal audit. On occasion the PAC has initiated its own independent enquiries, for example into tax evasion by a foreign bank. The PAC usually examines a selection of the instances of unauthorized or extravagant expenditure and other cases raised in the auditors' reports, and sometimes these lead to larger issues such as in the following example. In 1993-94, a PAC report commented on the failure of India's Export Processing Zones to live up to expectations. The Committee was following up on previous recommendations aimed at helping these special zones meet their target of producing 8-10% of total national exports. In the 1980s, export zones had contributed only 2-3% of total exports. The Committee was concerned to learn that in the 1991-93 period, the rate was still in the 2-3% range, even though instructions had been issued to implement the PAC`s previous recommendations. In response to the PAC follow-up report, the government commissioned a special study to evolve a strategy for attracting foreign capital and technology and improving export performance. The Committee may follow up again, although it has not yet done so. From past experience, the Committee claims that on occasion the announcement of its intention to hold further hearings has had a deterrent effect. This case illustrates an attempt to deal with a major issue: the fact that the results of an important government program were not being monitored or reported adequately. The Committee has taken a high-level approach to the problem and focused on the final result, leaving it to the government to decide on and monitor the means to improve export performance. information." Compare this with p.5: "If there's something partisan going on [in a US committee], if it's Republican against Democrat, hard politics, then politics in Washington is very much a blood sport and they would not hesitate to go in for the kill and I think that's much different than the way it's practiced in England". One panelist concluded: "Governance is not a matter for amateurs any more" (p. 10) 30 See Case 3. A study could be done of whether PAC members with legal training are more effective questioners 13 Some supporting conditions that help create solid research support for committees are: Communication skills of all concerned, especially the legislative auditor who has to translate financial and technical material into lay terms Adequate resources for expert staff and advisors, for independent expert witnesses, and perhaps for travel and public hearings; Trust between staff, experts and committee members, in particular the assurance that staff are non-partisan; Good relationship between researchers and the public service and/or strong access to information rules. Case 3 CANADA In the last parliament, the federal PAC in Canada had 9 government and 8 opposition members with an opposition chair. It produced 10-20 reports a year based mainly on the reports of the Auditor General of Canada. The PAC also conducted enquiries on its own, most recently about international financial reporting standards for the public sector, and internal audit and evaluation. The government pays close attention to PAC recommendations and has already implemented those made a few years ago as a result of another independent enquiry supporting more frequent reporting for the Auditor General. This case concerns an independent PAC investigation about 25 years ago beginning with a short reference in the Auditor General's report to unsubstantiated payments made by Atomic Energy of Canada Limited (AECL), a crown corporation. The Committee conducted a detailed enquiry into more than C$20 million that AECL had paid to agents abroad in hopes of selling nuclear reactors it manufactured. The Committee concluded the Corporation followed totally unacceptable business practices, and reported that they suspected some of the payments were used for illegal or corrupt purposes. The PAC went far beyond the audit report, wrote many letters, called many witnesses including the retired company President, held 17 hearings over the course of a year, and wrote a far-reaching report that recommended major changes in the accountability regime for crown corporations that were subsequently implemented. During the course of the enquiry, all committee members actively cross- examined hostile witnesses. The media gave the enquiry front-page attention and were able to uncover additional evidence the Committee could not have obtained on its own. This case illustrates a number of features of an effective PAC: selection of important issues, a focused enquiry tenaciously pursued, and a big-picture report with recommendations that are implemented. Perhaps surprisingly, lack of resources was not mentioned too often as a problem - the survey results are not conclusive on this point31. The survey did not enable us to obtain sufficient 31We would have thought resources were the key variable here, but there was no correlation between the number of committee professional staff and the results committees reported achieving. United Kingdom, one of the most effective committees, has only one professional staff member. We did not collect information on how many staff or other resources legislative auditors provide to committees. We have no clear idea what the large 14 information on the value added by outside expertise, or how often independent witnesses testify, although the findings are clear that all committees value/ need strong research support. 5. Bi-Partisan Climate: Strong disagreements between and/ or within political parties can be very damaging to the PAC. Parliaments are, by definition, arenas for the clash of ideas, and it is not surprising that differences of principle and even partisan issues in the legislature sometimes carry over into the PAC. Sometimes the partisan differences reach a point where the government is unwilling to accept any criticism or to act on valid complaints, and the opposition for its part overplays minor misdemeanors.32 Some of the survey respondents alluded to partisan problems. Having a balanced representation among parties within the committee - most Chairs thought it was "very important" and five respondents identified it among their top three success factors. Developing a close working relationship among members from different political parties also applies - most Chairs thought this very important, although none identified it among their top three. Another factor contributing to a bi-partisan climate is having a clear focus on implementation of policy, and not on whether policies are good or bad.33 Is a close balance of party representatives on the committee important, rather than a large government majority? There are examples of effective PACs with large government majorities, but this undoubtedly places more pressure on the need for a skilled and experienced opposition representation and chairmanship. A legislature with only a few opposition members will likely have a difficult time staffing the PAC as well as other committees. The survey provided some evidence that the United Kingdom and Australia have been quite successful in defusing partisan disagreements. There are checks and balances built into these models -- opposition chair but government majority; focus on administration but not policy; focus on public servants and not ministers (in United Kingdom only) -- that help to work around partisan problems and make it adaptable to varied situations. In addition there are special factors in the United Kingdom Parliament that support committee work as a career path, including the longer tenure of MPs on the committee and the large number of members who cannot hope to become government ministers. The United Kingdom PAC also declines to mediate between the legislative auditor and the government in disputes about the facts of a case, thus eliminating one cause of partisan friction in the committee and avoids time- number of professional committee staff in India does - one committee with a large staff still identified lack of staff as a problem. Perhaps the type or skills of staff may be a problem rather than the number. This would require more work 32 Malaysia did not respond to the survey, but we were told that 30 years ago the PACs in that country were chaired by opposition members, but since then have been chaired by a government member. If true, it would be useful to investigate this further 33In one case a PAC has expressed the view that there is need for flexibility regarding dealing with important matters of public concern regardless of the line between policy and implementation. This view is based on the ground that policy is good only when it is implementable 15 consuming discussions about whether something happened, rather concentrating on why it happened and what has been done to prevent it happening again. In Australia (federal and state parliaments), the Committee chairs - most of whom are government members - place great emphasis on operating in a non-partisan manner and on reaching unanimous agreement about agendas and reports. But unlike in the United Kingdom, the federal PAC in Australia also deals with disputed issues between the executive branch and the legislative auditor. This Committee is able to select a wide range of issues of public accountability on a non-partisan basis regardless of whether the issue is raised in the Auditor General's Report.34 Interviews provided some further suggestions for diffusing partisan disagreements: Reducing ministerial involvement both as members and as witnesses (but having former ministers may provide valuable experience) Sometimes a minority government or close balance of parties in the legislature heightens partisan tensions in the PAC; while at other times it creates the need for co-operation. One would expect a minority government to tolerate or compromise on administrative issues that don't threaten to become election rallying points. Sometimes an experienced and skilled chair can develop strategies to avoid or minimize partisan deadlock; sometimes not. In-camera meetings to set agendas and review or draft reports help.35 Sometimes election by the legislature of the committee chair, provision of a special salary and/ or other perquisites can increase the prestige of the position and make it attractive to politicians who might otherwise seek more partisan rewards. In short, while there is no panacea for tempering the partisan climate, it can sometimes be reduced by the changes discussed above and by other means, such as clearly focusing on accountability for implementing approved policy objectives without questioning the objectives themselves. 6. Public Involvement and Media Coverage: There are regional differences in the extent of public involvement in the PAC and parliamentary affairs generally and opinion is divided on whether the media should have access to PAC meetings. However there are examples where the media has provided valuable 34The committee has held hearings on the working of the wider sector accounting profession. 35The recent European Union study of Supreme Audit Institutions (SAI) and Parliamentary Committees (PC) identified 6 factors that help prevent politicisation of audit-related meetings: a) the organisation of the parliamentary committee's work according to the Parliament's Standing Orders, b) the participation in PC meetings of officers responsible for the matters being discussed, and other experts, c) the participation in PCs of members from all political parties and the right of all members to be heard, d) the oath each MP takes which provides for adherence and respect for the Constitution and other legislation, e) in some of the countries the MPs endeavour to avoid politicising PC sittings when the SAI-related matters are discussed, f) in case of committees that are, exclusively or mainly, in charge of state audit-related issues, the position of a chairman is sometimes held by a representative of an opposition party (the Czech Republic, Malta, Poland) or a representative of an opposition party holds the position of a vice-chairman. 16 information for a PAC enquiry (see Case 3) and media presence may also keep public servants and auditors on their toes and perhaps encourage more realistic government commitments for follow-up. Case 4 NEW SOUTH WALES (Australia) The Public Accounts Committee of New South Wales, Australia, has 3 government and 3 opposition members, with the chair from the governing party. It was established in 1902, but its powers were significantly increased in the early 1980s, with bipartisan support. At that time it began to initiate its own enquiries and hold public hearings, supported by its own secretariat and working closely with the Auditor General. This case concerns several PAC inquiries into the justice system. Increasing caseloads and litigation have burdened the many agencies involved in the justice system, including the police and courts themselves. The end result, for the public, was a feeling of powerlessness and concern about the quality of justice and about serious delays in the courts. In 1995, the Auditor General conducted a preliminary audit of the performance of the court system. The PAC decided to focus on one aspect of the subject on which parliamentarians had a special vantage point - the impact of the system on the public. It invited briefs from interested parties, held public hearings, and published a bipartisan report with many recommendations to improve the court experience for the public. Since then, many changes, several government enquiries, and a follow-up audit have taken place. There have been big improvements in waiting times for criminal cases in the Supreme and District courts, and case management procedures and time standards have been introduced. But the rapid growth in civil cases has continued to cause problems. So, in 2001, the PAC began another public enquiry. By focusing this time on the results achieved by the courts in the context of the justice system as a whole, they attempted to ensure that bottlenecks are removed, not just moved. Their report has now been published, and it recommends broad changes including greater use of alternative dispute resolution, compulsory mediation, and penalization of legal practitioners who unnecessarily delay civil cases. The Committee anticipates that this report will receive a positive response from the government, as did their first one. This case takes a system-wide look at a problem from the public's perspective. In doing so it illustrates a useful and inexpensive way to involve the public directly in a PAC enquiry that builds on the work of the legislative auditor but does not rehash it. It also illustrates a tenacious follow-up program that builds credibility for the PAC's future work. Committees in the "developed Commonwealth" (Australia, Britain, Canada and New Zealand) tend to meet in public while elsewhere the practice is mixed. Committees in South Asia generally do not meet in public, but the committees in Jamaica, Singapore and South Africa do, as did the ad-hoc committees in Pakistan between 1999 and 2002. The argument that is made in South Asia is that secrecy facilitates frank exchanges with witnesses; however the success of PAC performances in these countries is no better than in other jurisdictions. 17 Keeping meetings closed from the public is a little surprising in the case of Rajasthan, India, which has made great strides in opening up the budget process to public input but where there does not appear to be any corresponding move to open up the PAC to public discussion of how the budget is implemented. The same situation exists in the Indian states of Orissa and Karnataka. The United Kingdom is also somewhat of a surprise as, until 1977, the PAC met in camera and still does not hear outside witnesses. But there are some changes under way to encourage public involvement in the budgeting process in the United Kingdom, and there are already other avenues of direct public feedback on government services through the Citizen's Charter. In fact, the emphasis on published performance measures in Australia, Canada and the United Kingdom involves the public directly in assessing the quality of service delivery. Closer direct contact between the executive and the public is not inconsistent with public involvement in the PAC, and can provide new and relevant information to help the PAC hold the government to account. Important powers include the ability to compel ministers to attend, to subpoena witnesses and to hold witnesses in contempt of Parliament if they fail to answer questions, can be useful. Where such power is lacking, the glare of publicity can compensate for the lack of this power. Getting the media involved in an enquiry can sometimes bring to light evidence that would not have appeared otherwise (see Case 4). Case 5 IRELAND The PAC has operated since the foundation of the State in 1922 and both it and the Office of Comptroller and Auditor General (C&AG) are modelled on the United Kingdom equivalents. In early 1998 media reports suggested there might be significant evasion of Deposit Interest Retention Tax through the use of bogus non-resident bank accounts. (The total amount of money held in non-resident bank accounts was in the billions on pounds, much of it legitimate.). The PAC sought to investigate the matter but concluded it had insufficient powers to do so. This led to the enactment of the Comptroller and Auditor General and Committees of the Houses of the Oireachtas (Special Provisions) Act, 1998 which granted extensive quasi-judicial powers to the C&AG and allowed him to investigate the operation of the tax by the revenue authorities and the financial institutions, and to report his findings to parliament. This investigation broke new ground and the C&AG's report formed the basis for a subsequent public inquiry by a sub-committee of the PAC into the whole affair which generated enormous public interest. It led ultimately to the payment of significant sums by the financial institutions and others to the revenue authorities. The affair raised public awareness of the PAC and the Office of the C&AG significantly. Some committees, for example in Australia, Ireland36 and Canada tend to accept testimony from outside experts as well as the general public, rather than relying totally on the legislative auditor and public servants (see Cases 4 and 5). The PAC in New South Wales, Australia is a particularly good example of this trend because it considers public education as 36Not a Commonwealth country, but has a Westminster parliamentary system 18 one of its primary responsibilities, and often picks subjects for examination that benefit from direct public input, such as the level of service that the government provides. In general, those committees that do not meet in public, or accept input from the general public and particular interest groups, are missing a key source of information, which likely hinders their effectiveness. In summary, the success factors that PACs themselves consider most important include: Having a broad scope Power to select issues without government direction Power to report conclusions, suggest improvements and follow up on these Strong support from the legislative auditor, members and research staff that creates a unity of purpose about PAC work Having a bi-partisan relationship among Committee Members Involving the public and encouraging media coverage There are other factors that were rated highly, many of which support these. The survey results support the view, for example, that a PAC will be more effective if it meets regularly to keep up to date with the progress of public business, if Members are well prepared for Committee meetings and if detailed records are kept of Committee meetings. It should be noted, however, that the possession of these powers does not guarantee success, nor does their absence necessarily hinder PAC effectiveness. Said one Chair [The committee] "does not have a mandate to report to the House any of its deliberations, opinions or recommendations pertaining to government operations. Furthermore, although committees of the legislative assembly in general theoretically enjoy the power to send for persons, papers and records, the Committee has never done so. Correspondingly, the Committee has never recommended to the House that it censure the government for contravening ... [the] law, nor has it ever recommended to the House that it exercise its penal powers of arrest and detention over government officials who fail to comply with orders of the House. The deputy chairman hails from the governing party, and members of the governing party dominate the Committee in number, thus effectively silencing any attempts to substantively investigate matters pertaining to the Public Accounts, policy objectives and program delivery. Since the Committee has rarely, if ever, presented a substantive report to the House, the government has never had to issue a comprehensive response. The legislative assembly ... has never referred a substantive matter to the Committee. Lastly, the Committee is hamstrung by a lack of a budget to employ the services of a research and investigative staff ..." This particular committee lacks many of the success factors discussed in this paper and considers its current mandate a serious constraint. This situation is atypical; however, as many other committees elsewhere seem to cope with similar restrictions, and yet appear to function reasonably well. Some of the common problems, such as not holding meetings or 19 writing reports; not addressing important issues; not keeping up to date with the reports of the legislative auditor, or giving too much attention to superficial issues, have already been highlighted. From the responses received, however, it would appear that an analysis of constraints has to go beyond the absence of particular powers and practices and ask why powers have not been granted or, if granted, are not used. By looking at the broader environment in which legislative oversight takes place, it is possible perhaps to learn more about why some PACs work better than others. As noted above, the PAC is part of a broader external public financial accountability system that includes many other actors, including the government, the Auditor General and the public. These relationships bear further investigation. How the individual actors interact determines the effectiveness of the whole. But this is complex territory - it involves conjecture into political and other constraints that may well be indigenous to each of the jurisdictions examined - and there are no clear answers. One determining factor is the willingness of government to allow parliament in general, and the PAC in particular, sufficient political space to carry out their constitutional mandates. There are several interrelated problems: the inherent weaknesses of parliamentary committees and of parliamentarians, and the lack of commitment on the part of some governments in correcting these weaknesses. And then there is the perception that given the history of an adversarial relationship between the auditor and the auditee the `findings' of the legislative auditor are too negative and not always sufficiently balanced or presented to invoke better operational performance. Oversight requires knowledge, skills and experience that many parliamentarians do not have, and who often report difficulty in "learning on the job". Representing, as they do, all walks of life and all geographic regions, they often face a huge information deficit when questioning an important senior official backed by the technical resources of the government. It is not always in the government's interest to enhance the knowledge and skills to make this confrontation a success. Even if committee members do not feel intimidated, they can easily be put off by procedural tactics, or baffled by detail, or discouraged from asking in-depth questions that firmly establish the causes or responsibility for problems. As a result, MPs may feel powerless, possess less knowledge, and prefer other activities to committee work. Their lack of knowledge, depth or diligence may create a self-fulfilling prophecy - the government may lose respect for the role of legislative oversight, and have no confidence that useful results can come out of it37. In the United States, there is a more even balance of power between the legislature and the executive. There is no equivalent to the PAC in the congressional system, where all committees have forward-looking policy responsibilities and review or post-audit functions with a strong audit office supporting the work of the committees. In the Westminster system the executive branch is very powerful and committees have fewer powers and resources than in the United States. Why, then, in a parliamentary system, would the government be willing to suffer public criticism of its activities when it doesn't have to? It could, instead, depend 37In Australia, the PAC did not meet for almost 20 years from 1932-1951 because the Government decided it was unnecessary. 20 more on internal audit and review, or force the PAC and legislative auditor to operate in camera, like internal review mechanisms generally do. The government could, if it wishes, control the PAC by the way it uses its powers, for example, to appoint the majority of members and the chair, select issues, shut down enquiries and decline to implement recommendations. Yet there is a trend in certain Commonwealth countries (namely the United Kingdom, Canada and Australia and more recently, emerging in countries such as Ghana, Kenya, Sri Lanka, and Uganda) to strengthen the role of parliamentarians and parliamentary committees. In the United Kingdom, the chair is usually a former Treasury minister who has first-hand experience in the subjects the PAC deals with. And these countries have opened up their PACs and other committees to public input and exposure. Section 4: Benchmarks for Measuring PAC Performance? A question we sought to answer as we examined factors for success for PACs is: "Are there potential indicators or benchmarks that can be used when examining PACs?" At present there is no international standard setting process for legislative functions such as oversight and control of the public purse. Is it possible to develop an index for the effectiveness of PACs? It is not surprising that public sector accounting and auditing is perceived to have fallen behind comparable improvements in private sector and other aspects of good corporate governance despite some advances on the part of legislative auditors. This particular gap between the public and private sector financial accountability standards and practices is likely to grow given there is no existing force to address this critical issue. However, there are some encouraging signs. For example, international auditing associations have set some international audit standards; Auditors General have already made considerable progress38. But the same cannot be said for PACs, where development of standards has been uneven and slow. Nonetheless, it would appear that a comprehensive framework for assessing PAC performance would encompass several different dimensions or types of information: Activity level - the measurement of events and input resources used such as: keeping up to date with legislative auditor's reports, costs and time of staff, members, witnesses and others involved. Output level - the immediate visible results of work the committee does like: recommendations made, followed up and implemented Outcomes level - durable improvements in public administration such as: increased economy, efficiency or effectiveness of government programs better compliance with laws or regulations, improvements in financial and control structures, such as prosecution of wrongdoers, stronger powers for legislative auditor more accurate, timely government information enhanced public awareness of government programs 38International Organization of Supreme Audit Institutions (INTOSAI), Proposed Strategic Plan 2005-2010 21 enhanced legislative knowledge about the state of the management of programs All three dimensions of performance are important. Beyond these indicators, there is a question whether the PAC is actually dealing with important issues such as the root causes rather than the symptoms of problems (like corruption), and the deterrence effect of the PAC on the behavior of public servants. Some call this dimension of performance "relevance"39. Case 6 QUEBEC (Canada) The Committee on Public Administration (CPA) in Quebec has 6 government and 4 opposition members, with an opposition chair. The CPA took over from the former PAC in 1997 with new responsibilities that stressed administrative, non-partisan action as well as review of the Auditor General's reports. The CPA reports twice a year. In 2002, the Chair of the Committee started an informal review of the first five years of the committee's work by asking six key questions: Could the CPA make the distinction between political and administrative aspects in the management of departments and agencies? Could senior public servants testify without getting their political masters into hot water? Could a bipartisan approach work in the CPA given the highly partisan nature of the legislature? Does the Legislature permit the CAP to hold timely public accountability sessions with departments? Will the CPA be able to review the many new performance reports that departments are required to produce each year? Has the CPA ensured that its recommendations and the government's action plans to deal with them are followed up effectively? The Chair of the Committee presented this framework for a PAC scorecard at the 2002 meeting of all Canadian public accounts committees. He has made his own preliminary personal assessment of his committee's score for each question/criterion, and he is planning to involve the committee in a more formal assessment in the next few months. The results could be presented to the legislature. This case identifies most of the key problems that PACs face in achieving effective results. Moreover it illustrates a courageous attempt by a PAC to assess its own effectiveness. We found very few committees that published reports dealing with their own performance, or assessed their success in informing the legislature about the accountability of the government. This is somewhat surprising in view of the fact that performance reporting is an important feature of effective legislative oversight, and some PACs already review and critique departmental performance reports, or performance reports of the Auditor General. It makes sense for the committee to set a good example and do what it insists others do. 39 Our thanks to Bob Miller, Executive Director of the Parliamentary Centre for pointing us in the direction of an IDRC publication on Organizational Assessment, a Framework for Improving Performance that suggests relevance as an indicator. 22 Unfortunately, at the present time, few of them are. Why? Perhaps some committees are assessing their performance in camera, or perhaps there is a problem in identifying success factors that are non-partisan. Case 7 VICTORIA (Australia) The Australian state of Victoria has had a Public Accounts and Estimates Committee since 1895. Its members are senior, high profile parliamentarians 5 of whom are from the government side (including the chair), and 5 are opposition members one of whom is an independent. It has three main objectives: to encourage economic, efficient and effective utilization of public resources; to enhance accountability to Parliament and the public; and to enhance the presentation and disclosure of information to Parliament and the public. In part because of its estimates responsibilities, the Committee is more active than most PACs, averaging over 50 meetings a year. The Committee has very wide responsibilities which include initiating and carrying out its own enquiries, and to help with these the Committee has a panel of specialist advisers that deal with complex and technical issues. In addition, they work closely with the Office of the Auditor General. They frequently suggest changes to the terms of reference for performance audits, and selectively follow up on unresolved issues in its reports. The Committee also recommends the appointment of the Auditor General, and commissions a performance audit of the legislative audit office every 3 years. An interesting and possibly unique activity is the program of seminars that the Committee holds for members of Parliament on financial and accountability subjects. The Victoria PAC has carefully thought out what it would like to achieve and how to get it done. The Committee's annual report has a chart depicting what the committee does, and the impact of its work. This performance and results framework starts with the inputs - the knowledge and skills of members, and the information received from stakeholders about the operations of Government - and the outputs, which are advice and recommendations. Then it moves into a detailed analysis of the intermediate outcomes - such as better performance of government - and final outcomes such as improved public confidence and better parliamentary control. In its annual report, the Committee publishes performance targets and plans for the year, as well as statistics on past performance. Thus, for example, the Committee sets a target for the Government's acceptance of its recommendations, and it keeps count and publishes the results. It also keeps track of the timeliness of its review and coverage of the Auditor General's reports, and is not afraid to report slippage if they do not get done all the work they expected to do. The Victorian PAC is a leader in performance reporting for PACs. In part, the reluctance to assess performance is a result of the difficulty of the questions that must be addressed, such as who should agree to the measures, who should see the results, and how they will be used. Measurement can sometimes have perverse effects, such as possibly encouraging a Committee to select easily resolvable issues or make soft recommendations. There is a danger that a committee might shy away from difficult but valuable work with long term impact that cannot be measured, given the current state of the art. 23 We sought to identify examples of "good practice" in PAC reporting and performance measurement. The outcome of our interviews with selected Chairs and document reviews are presented in Case Studies 6 and 7 (as well as Case 1, earlier), which show some examples of performance results that some PACs have achieved. Section 5: Conclusions In his study of Public Accounts Committees and Auditors General, McGee (2002) identifies three main priorities for action. First of all, given the importance of institutional capacity of the overseers for the effective financial oversight of the government expenditures, McGee suggests that the ability of the parliament, of the PAC and of the Auditors General to carry out their oversight functions should be improved. In his view, this result (capacity building) could be achieved by providing staffing and resources, training and access to information. Second, McGee underlines that a key determinant of the success in the activity of Auditors General is represented by their independence from partisan and political influence. Auditors General need to have the freedom to carry out their duties independently and impartially. Finally, McGee stresses that a key determinant for success in the activities of the PACs is represented by the availability of information and information exchange. McGee argues in fact that PACs need to exchange information and ideas in order to stay up-to-date with developments, changing standards and best practices as they emerge. The evidence presented in this paper is to a considerable extent consistent with the conclusions formulated by McGee. The PACs Chairs surveyed in the study conducted by the WBI reported that information availability and bipartisanship/non-partisanship are critical conditions for the success of the PACs as two of McGee's three conclusions previously suggested. But the analysis conducted in this paper allows one to gain a much better understanding of how the capacity of the PACs can be built. Two sets of factors seem of great importance in this regard. One concerns the institutional design of the PAC, while the other concerns their behavior and functioning. The success of the PACs depends to a large extent on how they are institutionalized, on what institutional features and characteristics they have, that is their power and mandate. In this respect, we suggest that PACs should focus on governments' financial activity and accountability rather than evaluating or assessing the content of the governments' policies. Second, we believe that PACs should have the power to investigate all past and present government expenses regardless of when they were made. Third, PACs should be given the power to check whether the government actually undertakes some steps to implement the recommendations of the PAC itself. And finally, PACs must have a close working relationship with the Auditors General. The success of the PACs does not depend exclusively on institutional design, but it also and more importantly depends on the behavior of its members and the functioning of the committee itself. Here we have been able to identify some obvious best practices: PACs' members must act in a non partisan fashion and should try to have a good working 24 relationship with other committee members in spite of possible partisan differences. In its functioning the PAC should always strive for consensus. Second, our study has revealed that effectiveness in the PACs' activity increases whenever the PACs members study the documentation and prepare themselves before the PAC meetings. Third, the PACs should keep the transcripts of their meetings, they should publish their conclusions and recommendations, and they should involve the public and the media. Public opinion can in fact provide a strong incentive for governments to improve their financial accountability and avoid possible allegations of ineffective management of public resources. In the light of these conclusions, we are now able to present, however cautiously, some aspects of an ideal PAC (see Case 8 below). Case 8 An "Ideal Committee"? · The Committee is small; committees seem to work well with 5-11 members, none of whom should be government Ministers; · Senior opposition figures are associated with the PAC's work, and probably chair the Committee; · The Chair is a senior parliamentarian, fair minded and respected by parliament; · The Committee is appointed for the full term of the parliament; · The Committee is adequately resourced, with an experienced clerk and a competent researcher(s) · There is clarity on the Committee's role and responsibilities; · The Committee meets frequently and regularly; · Hearings are open to the public; a full verbatim transcript and summary minutes are quickly available for public distribution; · A steering committee plans the Committee's work in advance and prepares an agenda for each meeting to the full Committee; · The typical witness is a senior public servant (the "accounting officer") accompanied by the officials that have a detailed understanding of the issues under examination; · The Auditor's Report is automatically referred to the Committee and the Auditor meets with the Committee to go over the highlights of the report; · In addition to issues raised by the Auditor, the Committee occasionally decides to investigate other matters; · Committee strives for some consensus in their reports; · The Committee issues formal substantive reports to parliament at least annually; · The Committee has established a procedure with the government for following up its recommendations and is informed about what, if any, action has been taken; · In all its deliberations, the Committee uses the Auditor as an expert advisor; · Parliaments hold an annual debate on the work of the Committee. 25 Annex 1: Main Conclusions from McGee book, The Overseers The Study Group identified three main priorities for action: 1. Capacity building. A constant theme is the need to improve institutional capability, that is, the ability of Parliaments, PACs and Auditors General's offices to carry out their functions by being provided with sufficient resources and having adequate training and access to expertise that they require. 2. Independence. Particularly for Auditors General it is essential that they be free from political or legal constraints that could inhibit them carrying out their duties diligently and impartially. 3. Information exchange. PACs in particular need to have the means to exchange information and ideas so as to keep them up-to-date with important developments, changing standards and best-practices as they emerge. The following are the main individual conclusions and recommendations of the Study Group: The International Dimension: · There should be greater direct contact between Parliaments, especially PACs, and international financial institutions. · The CPA include good government as a subject of the theme or sub-theme of its conferences Auditors General: · The Auditor General should be an Officer of Parliament independent of the Executive · The appointment process for an Auditor General should involve consultation with a wide range of stakeholders · An Auditor General should only be removed from office on limited grounds that are specified in advance by law · Auditors General should actively participate in international Auditors General associations · Auditors General should actively introduce themselves and their services to all parliamentary committees, not just PACs · Auditors General have a role in approving internal audit standards · Central banks should be subject to the Auditor General's audit mandate in the same way as other public sector agencies · The Auditor General should take account of the views of PACs in framing their work programs · Parliaments should be involved at the pre-Budget stage in determining the resources to be allocated to the Auditor General · Auditors General and their staff must have appropriate legal protections conferred on them to enable them to carry out their duties · PACs should keep under review any proposals to change the Auditor General's audit mandate · Any company receiving public funding to deliver public services should be subject to the Auditor General's audit mandate in respect of those services 26 · Auditors General should present their reports in an attractive form and devise active communications strategies · Auditors General should take steps to measure their own performance · The main of PACs work should be guided by the work of the Auditor General Public Accounts Committees: · Parliaments should regard the PAC as their pre-eminent committee · Senior opposition figures must be associated with the PAC's work · There should always be sufficient experience and seniority among the membership of the PAC · Specially structured training be provided to PAC members · It is crucial that the Chairperson of the PAC has the qualities to ensure that the PAC works effectively · PACs must be adequately resourced to carry out their functions · PACs, while not being bound to act unanimously, should strive for some consensus in their reports · PACs should promote greater public awareness of their role · PACs should consider using subcommittees for specific inquiries · The Internet should be used to disseminate information on PACs · Procedures for follow-up action in recommendations in PAC reports are critical · Parliament should hold an annual debate on the work of the PAC · PACs in smaller and developing parliaments need improved access to information technology · A rational local method of allocating funding to PACs needs to be put in place to ensure that they have adequate resources · Smaller Parliaments need to take innovative steps to expand the pool of personnel available to serve on the PAC · Special attendance allowances, rather than a special salary, should be considered for PAC attendance · Links between PAC websites should be developed · The CPA should explore the potential for the use of a news group to encourage information exchange on PAC matters · The CPA should examine what options exist for conferences of associations of PACs · A compendium of Commonwealth PAC practice be established to be managed by a CPA branch or Parliament · Research should be undertaken into establishing a basis for making international comparisons of PAC performance 27 Annex 2: Survey Instrument The purpose of this questionnaire is to identify factors that can improve the success of the Public Accounts Committee (PAC) of the legislature in monitoring the financial accountability of government. While not all countries have a committee with the precise name PAC, the questionnaire applies to any committee that performs the functions of a PAC, overseeing public administration to achieve value for money spent. 1. What is the exact name of the committee in your jurisdiction that performs the functions of the Public Accounts Committee? .................................................................................................................................................................................... 2. Is the Chair of the Committee from (Please check one): the Governing Party? (____) or an Opposition Party? (____) 3. How many Committee members are there? Government _____ Opposition _____ Total_____ 4. How many Committee staff are there? Professional _____ Support staff _____ Total_____ 5. In what year was the Committee first established? __ __ __ __ 6. Please provide an indication of the amount of Committee activity in the past 3 years, i.e. since 1 January 1999? Number of meetings (check one): 0-9_____ 10-24_____ 25-49_____ more than 50 _____ Number of subjects dealt with (check one): 0-9_____ 10-24_____ 25-49_____ more than 50 _____ Number of reports issued (check one): 0-9_____ 10-24_____ 25-49_____ more than 50 _____ 7. Did the Committee achieve any of the following types of results or successes in the past 5 years i.e. since 1 January 1997? If so, please indicate how often: Government responds favorably to Committee recommendations: frequently_____ seldom_____ never_____ Government implements Committee recommendations: frequently_____ seldom_____ never_____ Changes in legislation were adopted as a result of Committee work: frequently_____ seldom_____ never_____ Improvements in the integrity of government information or data bases: frequently_____ seldom_____ never_____ Legal action was taken against officials who contravene laws: frequently_____ seldom_____ never_____ Disciplinary action was taken against officials who contravene administrative guidelines: frequently_____ seldom_____ never_____ Please briefly describe any other types of results or successes achieved: .......................................................................................................................................................................................................................................... .................................................................................................................................................................................................................................................... .................................................................................................................................................................................................................................................... 28 8. Some suggest the following factors contribute to the success of a Public Accounts Committee. With reference to the powers and practices of your Committee, how important do you think each of the following factors has been in the past three years? Please circle the appropriate number rating each factor as either: 1 - very important; 2 - somewhat important; 3 - not important; or 4 - not applicable. You are welcome to add any other factors that are missing in the space at the end of this question. Some possible success factors 1=Very important 2=Somewhat important 3=Not Important 4=Not Applicable Composition of the Committee 8.1 Balanced representation of all major political parties on the committee 1 2 3 4 8.2 Membership should exclude ministers of the Government 1 2 3 4 Committee Powers 8.3 Clear focus on holding the government accountable for its spending of 1 2 3 4 taxpayers` money and its stewardship over public assets: 8.4 Clear focus on administration of policy, and not on whether policies are good or bad. 1 2 3 4 8.5 Having a permanent reference to examine the Public Accounts: 1 2 3 4 8.6 Having a permanent reference to examine all reports of the Legislative Auditor: 1 2 3 4 8.7 Power to call independent witnesses: 1 2 3 4 8.8 Power to investigate or review all past, current and committed expenditures of government, 1 2 3 4 organizations receiving funds from the government and all state or crown corporations: 8.9 Power to request (but not compel) the legislative auditor to perform specific reviews or tasks: 1 2 3 4 8.10 Power to compel officials to attend and be held accountable for administrative, 1 2 3 4 performance even after they have left office: 8.11 Power to compel witnesses to answer questions: 1 2 3 4 8.12 Power to make recommendations and publish conclusions: 1 2 3 4 8.13 Power to hold press conferences and issue press releases: 1 2 3 4 29 8.14 Power to hold in camera meetings, if dealing with sensitive or national security issues: 1 2 3 4 8.15 Power to hold meetings and conduct enquiries even when the legislature is not in session: 1 2 3 4 8.16 Power to review proposed legislation or amendments to the Legislative Auditor's Act: 1 2 3 4 8.17 Power to review the Legislative Auditor's budget: 1 2 3 4 8.18 Power to choose subjects for examination without government direction and advice: 1 2 3 4 8.19 Power to compel Ministers to appear before the committee: 1 2 3 4 Committee Practices 8.20 Close working relationship between members from different political parties: 1 2 3 4 8.21 Advance preparation of members before hearings: 1 2 3 4 8.22 Close working relationship with and research support from the legislative auditor: 1 2 3 4 8.23 Independent technical expertise and research support for hearings: 1 2 3 4 8.24 Separate subcommittees for groups of related departments: 1 2 3 4 8.25 Strategic prioritization of items for committee review, with time limits for stages of 1 2 3 4 committee work like consideration of departmental replies, reporting, implementation reports: 8.26 Transcripts kept of all hearings and meetings: 1 2 3 4 8.27 Committee appointed for the life of the Legislature or Parliament 1 2 3 4 and stays active between sessions: 8.28 Televised public hearings: 1 2 3 4 8.29 Report to the legislature annually, and ask for the report to be debated: 1 2 3 4 8.30 Comprehensive response to recommendations from the government: 1 2 3 4 8.31 Having committee members with at least 2 years of prior Committee experience: 1 2 3 4 30 8.32 Having committee members with prior administrative or business experience: 1 2 3 4 8.33 Extra pay or other incentives for members to participate in hearings outside 1 2 3 4 the normal legislative session: 8.34 Meeting place suitable for media and public access to hearings: 1 2 3 4 8.35 Incentives to encourage administrative action on committee recommendations: 1 2 3 4 8.36 Effective follow-up procedures to determine if action has been taken 1 2 3 4 to implement the committee's recommendations: 8.37 Good relations with other parliamentary oversight mechanisms 1 2 3 4 such as the budget committee Other important success factors for your Committee: 8.38.............................................................................................. 1 2 3 4 8.39.......................................................................................... 1 2 3 4 8.40.............................................................................................. 1 2 3 4 9. Of those factors you rated "1 - very important" in question 8, which three have been most crucial in achieving successful results for your committee in the past three years? Why? #1 ...................................... what has this helped you to do?................................................................................... #2 ...................................... what has this helped you to do?................................................................................... #3 ...................................... what has this helped you to do?................................................................................... 10. Of those factors in Question 8 that do not currently apply to your Committee (i.e. those you rated "4"), do you think that any of these powers and practices might make your committee more effective? Please list them, and explain why. ........................................................................................................................................................................................................................................................ ................................................................................................................................................................................................................................................................... ................................................................................................................................................................................................................................................................... ......................................................................................................................................................................................................... 31 11. Are there any documents that describe the history, powers and practices of the Public Accounts Committee? If so, could you please send us copies, or list them and let us know where they can be obtained? ....................................................................................................................................................................................................................................................... ................................................................................................................................................................................................................................................................... ........................................... 12. Please feel free to add or attach any additional comments that you would like to make. ....................................................................................................................................................................................................................................................... ....................................................................................................................................................................................................................................................... ................................................................................................................................................................................................................................................................... ................................................................................................................................................................................................................................................................... ................................................................................................................................................................................................................................................................... ................................................................................................................................................................................................................................................................... ................................................................................................................................................................................................................................................................... ................................................................................................................................................................................................................................................................... ................................................................................................................................................................................................................................................................... ................................................................................................................................................................................. Thank you very much for your time and assistance. The results of this survey will be summarized and a copy sent to all those who respond, after the study is completed. 32 Annex 3: Full Listing of Success Factors How many of the potential success factors do you consider very important? Somewhat important? Not important? Not applicable? (Sample = 34) Potential Success Factors Very Some Not n/a Composition of the Committee 8.1 Balanced representation of all major political parties on the committee 25* 3 1 4 8.2 Membership should exclude ministers of the Government 23 4 0 6 Committee Powers 8.3 Clear focus on holding the government accountable for its spending of 30 3 0 0 taxpayers` money and its stewardship over public assets: 8.4 Clear focus on administration of policy, and not on whether policies are good 19 11 2 1 or bad. 8.5 Having a permanent reference to examine the Public Accounts: 28 4 1 0 8.6 Having a permanent reference to examine all reports of the Legislative 21 7 0 5 Auditor: 8.7 Power to call independent witnesses: 24 6 1 2 8.8 Power to investigate or review all past, current and committed expenditures of 29 2 0 2 government, organizations receiving funds from the government and all state or crown corporations: 8.9 Power to request (but not compel) the legislative auditor to perform specific 17 7 3 6 reviews or tasks: 8.10 Power to compel officials to attend and be held accountable for 19 9 3 2 administrative performance, even after they have left office: 8.11 Power to compel witnesses to answer questions: 27 2 2 2 8.12 Power to make recommendations and publish conclusions: 32 1 0 0 8.13 Power to hold press conferences and issue press releases: 13 12 5 3 8.14 Power to hold in camera meetings, if dealing with sensitive or national 20 7 3 3 security issues: 8.15 Power to hold meetings and conduct enquiries even when the legislature is 24 5 2 2 not in session: 8.16 Power to review proposed legislation or amendments to the Legislative 11 7 5 10 Auditor's Act: 8.17 Power to review the Legislative Auditor's budget: 6 4 7 16 8.18 Power to choose subjects for examination without government direction and 30 3 0 0 advice: 8.19 Power to compel Ministers to appear before the committee: 11 3 6 13 Committee Practices 8.20 Close working relationship between members from different political parties: 24 5 3 1 8.21 Advance preparation of members before hearings: 26 6 1 0 8.22 Close working relationship with and research support from the legislative 22 6 2 3 auditor: 8.23 Independent technical expertise and research support for hearings: 17 8 3 5 8.24 Separate subcommittees for groups of related departments: 5 6 6 16 8.25 Strategic prioritization of items for committee review, with time limits for 12 18 2 1 stages of committee work like consideration of departmental replies, reporting, implementation reports: 33 8.26 Transcripts kept of all hearings and meetings: 29 3 1 0 8.27 Committee appointed for the life of the Legislature or Parliament and stays 19 5 1 8 active between sessions: 8.28 Televised public hearings: 3 6 10 14 8.29 Report to the legislature annually, and ask for the report to be debated: 17 5 6 5 8.30 Comprehensive response to recommendations from the government: 25 6 1 1 8.31 Having committee members with at least 2 years of prior Committee 3 11 10 9 experience: 8.32 Having committee members with prior administrative or business 4 14 10 5 experience: 8.33 Extra pay or other incentives for members to participate in hearings outside 8 7 7 11 the normal legislative session: 8.34 Meeting place suitable for media and public access to hearings: 17 6 2 8 8.35 Incentives to encourage administrative action on committee 10 7 5 11 recommendations: 8.36 Effective follow-up procedures to determine if action has been taken to 24 6 2 1 implement the committee's recommendations: 8.37 Good relations with other parliamentary oversight mechanisms such as the 10 10 4 9 budget committee: *Should be read as: 25/34 committees responded that factor 8.1 was very important in contributing to their success in the past 3 years 34 Annex 4: List of Respondents to the Survey 1. Canada 2. Alberta, Canada 3. British Columbia, Canada 4. New Brunswick, Canada 5. Newfoundland, Canada 6. Ontario, Canada 7. Prince Edward Island, Canada 8. Quebec, Canada 9. United Kingdom, Central Government 10. Northern Ireland 11. Scotland 12. Southern Ireland 13. Australian Commonwealth Government 14. New South Wales, Australia 15. Northern Territory, Australia 16. Queensland, Australia 17. Victoria, Australia 18. Tasmania, Australia 19. South Australia 20. Western Australia, Australia 21. Australian Capital Territory, Australia 22. New Zealand 23. India 24. Karnataka, India 25. Orissa, India 26. Punjab, India 27. Rajasthan, India 28. Tamil Nadu, India 29. Pakistan 30. Sindh, Pakistan 31. Sri Lanka (PAC) 32. Sri Lanka (Committee on Public Enterprises) 33. Singapore 34. Nepal 35. Uttar Pradesh, India 35 REFERENCES Stephen E. Frantzich, "Computerized Information Technology in the US House of Representatives", Legislative Studies Quarterly, vol. 4, n. 2, 1979, pp. 255-280. International Organization of Supreme Audit Institutions (INTOSAI), Proposed Strategic Plan 2005-2010, 2005. Arend Lijphart, Patterns of Democracy, New Haven, Yale University Press, 1999 Charles Lusthaus, Marie-Hélène Adrien, Gary Anderson, Fred Carden, and George Plinio Montalván, Organizational Assessment: A Framework for Improving Performance, IDRC/IDB 2002. David McGee, The Overseers ­ Public Accounts Committees and Public Spending, Pluto Press, London 2002. Robert Miller, Riccardo Pelizzo and Frederick Stapenhurst, "Parliamentary Libraries, Institutes and Offices - The Sources of Parliamentary Information", WBI Working Paper, 2004. Riccardo Pelizzo and Rick Stapenhurst, Tools of Legislative Oversight, Policy Research Working Paper # 3388, The World Bank, September 2004. SIGMA, Relations Between Supreme Audit Institutions and Parliamentary Committees. SIGMA paper No.33, OECD and European Union. Survey of parliamentary and external auditing institutions in European Union candidate countries, 2002 University of Strathclyde, Teething the Watchdogs, Report of a panel discussion comparing US and United Kingdom legislative committees, 1992. 36