Performance reporting: a comparative study of British and Irish charities

The British Accounting Review 36 (2004) 127–154 www.elsevier.com/locate/bar

Performance reporting: a comparative study of British and Irish charities Ciaran Connollya, Noel Hyndmanb,* a

b

School of Accounting, University of Ulster, Jordanstown, BT37 OQB, N. Ireland, UK School of Management and Economics, Queen’s University, Belfast, BT7 1NN, N. Ireland, UK Received 18 December 2001; revised 4 July 2003; accepted 3 October 2003

Abstract The concept of accountability seems inextricably linked with the view that accounting should provide information to satisfy the information needs of users. The user-needs model is now well established as a useful basis for a conceptual framework for charity reporting, and annual reports are recognised as key documents in the discharge of accountability to external users. It has been suggested that both financial information and also performance information should be disclosed to aid the discharge of accountability. However, previous empirical work conducted in Britain found that while audited financial information was most frequently disclosed by charities, users viewed wider performance information as being of greater importance. No comparable work has been conducted in Ireland. This paper focuses on information outside the financial statements and seeks to identify the type and extent of the reporting of performance information by charities in both Britain and Ireland. The main findings of the research are: performance reporting by British charities, although limited, is considerably better than that of their Irish counterparts; performance reporting by British charities has increased over time; and large charities (both in Britain and Ireland) provide more extensive performance information than small charities. These findings are discussed in both the context of accountability and in terms of conceivable economic incentives for disclosure. In addition, possible reasons for lower disclosure rates by Irish charities are explored. q 2003 Elsevier Ltd. All rights reserved. Keywords: Charities; Statement of recommended practice; Accounting

1. Introduction The charity sector in both Britain and Ireland comprises a vast and growing segment of economic activity. There were over 185,000 charities registered with the Charity * Corresponding author. Tel.: þ 44-2890273204; fax: þ 44-2890335156. E-mail address: [email protected] (N. Hyndman). 0890-8389/$ - see front matter q 2003 Elsevier Ltd. All rights reserved. doi:10.1016/j.bar.2003.10.004

128

C. Connolly, N. Hyndman / The British Accounting Review 36 (2004) 127–154

Commission of England and Wales at the end of 2000, with an estimated annual income of £24.6 billion (Charity Commission, 2001). It is estimated that there are approximately 5000 charities in Northern Ireland (NI) (NI Council for Voluntary Action, 1998), with a similar number in the Republic of Ireland (RoI) (Office of Information Commissioner, 1999). In each jurisdiction, charities have formed vital partnerships with government in the provision of a wide range of services. Growth in the size and influence of the charity sector has led to increased visibility and public scrutiny by diverse stakeholders including government oversight agencies, private donors and foundations, clients, the media and the public at large. A charity’s trustees are responsible for the preparation of the annual report and financial statements, which, together, are generally recognised as key documents in the discharge of accountability to external users. The Corporate Report (Accounting Standards Committee (ASC), 1975, p. 16) states that such information packages are ‘the primary means by which the management of an entity is able to fulfil its reporting responsibility’. The introduction to the 2000 Statement of Recommended Practice (SORP) (Charity Commission, 2000) states that the purpose of preparing a charity’s annual report and financial statements is to discharge the trustees’ duty of public accountability and stewardship. It is suggested that, among other things, the annual report and financial statements should enable the reader to understand the charity’s structure, activities and achievements. Recent research relating to charity external reporting has concentrated on the form and content of the financial statements (income statement, balance sheet, cash flow and notes) with the major thrust of the findings being that external financial reporting by charities is characterised by a diversity of accounting practices and a lack of standardisation which has resulted in difficulties for users in understanding financial statements (Ashford, 1989; Gambling et al., 1990; Hines and Jones, 1992; Williams and Palmer, 1998; Connolly and Hyndman, 2000, 2001). Differences between large and small charities have been highlighted, with larger charities providing more complete financial statements. Moreover, in the only comparative British/Irish study, it was seen that Irish financial statements were considerably less compliant with recommended accounting practice for charities than their British counterparts. While these weaknesses in financial statements are seen as problematical in terms of discharging accountability, research relating to British charities, published over a decade ago (Hyndman, 1990), highlighted that performance information included in charity annual reports is viewed as more useful than traditional financial accounting information, and more important in discharging accountability. Little subsequent research relating to this has been conducted in Britain, and no research at all regarding the state of performance reporting by charities in Ireland. This paper, while recognising the importance of reporting financial information, focuses on information included in annual reports of charities outside the financial statements. The main objectives of the paper are: to review the linkages between accounting and accountability, highlighting particularly the importance of performance information in discharging accountability by charities; and to conduct an empirical analysis of recent annual reports of charities, distinguishing between British and Irish, to ascertain the extent to which basic background information (for example, related to the legal and administrative arrangements of the charity) and more specific performance

C. Connolly, N. Hyndman / The British Accounting Review 36 (2004) 127–154

129

information is used in external reporting. Basic background information can be viewed as being important in providing an appropriate context for the understanding of the performance of a charity. Moreover, an attempt is made in the paper to explore the extent to which the reporting of performance information is related to size, with the possible expectation that larger charities provide more comprehensive reports of performance. A model for reporting to contributors to charities developed by Hyndman (1990) is used as a point of reference. In terms of the format of the paper, the following approach is taken. Section 2 discusses accounting, accountability and performance information, and relates these issues to charities. Previous research related to the area of reporting by charities is then reviewed. Following this, the method used in this study is outlined and the results are presented, discussed and analysed. In Section 6 some concluding comments are made and areas for further research identified.

2. Accounting, accountability and performance information The conventional view of accounting is that it is a purposive activity, directed towards a specified end, that is the meeting of users’ information needs (for example, American Accounting Association (AAA), 1966; Canadian Institute of Chartered Accountants (CICA), 1980; Macve, 1981; Accounting Standards Board, 1999). While the main focus of these publications has been on business accounting issues, they have helped to stimulate debate on user-needs with respect to the charity sector. The user-needs emphasis was evident in the Trueblood Report (American Institute of Certified Public Accountants, 1973) and the Statement of Financial Accounting Concepts No. 4 Objectives of Financial Reporting by Non-business Organisations (Financial Accounting Standards Board (FASB), 1980) in the United States, and in the work of Bird and Morgan-Jones (1981) in the United Kingdom (UK). This latter research ultimately led to the publication of SORP 2 Accounting by Charities (ASC, 1988), with subsequent revisions in 1995 (Charity Accounting Review Committee (CARC), 1995) and in 2000 (Charity Commission, 2000). These publications sought to reduce the diversity of accounting practice in charities’ published financial statements on the basis that a lack of consistency makes it difficult for users of accounts to understand (and therefore use) the information provided. While accepting the argument made by Jones and Pendlebury (1996), that accountability, in its widest sense, is more than accounting (however, widely accounting is defined), focusing on the information needs of users (a key aspect of a contemporary definition of accounting) seems clearly linked with ideas relating to accountability. A dictionary definition of accountability views it as being ‘responsible to someone or for some action’ (HarperCollins, 1993). Jackson (1982, p. 220) defines accountability in terms of ‘explaining or justifying what has been done, what is being done and what has been planned’. Rutherford (1983) views accountability as being related to the requirement to be answerable for one’s conduct and responsibilities. In an organisational context, these definitions might imply responsibility to an oversight agency (for example, the Charity Commission) with formal record keeping and reporting requirements as a means of demonstrating compliance, and explicit standards of performance or assessment by this higher authority.

130

C. Connolly, N. Hyndman / The British Accounting Review 36 (2004) 127–154

Stewart (1984) suggests a ladder of accountability, distinguishing between: accountability for probity and legality; process accountability; performance accountability; programme accountability; and policy accountability. While there are no standardised definitions for the types of accountability highlighted in Stewart’s ladder, and indeed there are possibly overlaps, the classifications used highlight that the information requirements vary with the differing bases of accountability. He argues that an accountability information system should report on all levels of accountability, providing financial information and also output and outcome information. While a detailed discussion of these various types, or rungs, of accountability is beyond the scope of this paper, what is suggested is that with some types of accountability (for example, accountability for probity and legality), one can have accountability by standards, whilst with higher levels of accountability (for example, with policy accountability), accountability by judgement is required. Similar ideas are presented by Kearns (1994). While the definitions of accountability presented above contain useful elements, discussions on accountability in the not-for-profit sector are fraught with problems. The Government Accounting Standards Board (1987) and Patton (1992) both suggest that accountability suffers from imprecise meaning, and Rutherford (1983) and Taylor (1989) argue that user-needs models do not explain clearly the information that should be provided. Furthermore, operational definitions of accountability often focus on compliance with specific legal mandates, such as the filing of accounts in a specified format, while conceptual definitions tend to become overly complicated, even ambiguous, in an attempt to meet the needs of diverse users (Stewart, 1984). Edwards and Hulme (1995) highlight the problem of multiple accountabilities, suggesting that difficulties in prioritising and reconciling varied user-needs may lead to weak accountability. This view is supported by the fact that discussions of accountability in the context of NFPOs have often focused on such disparate issues as: measuring the value-added performance of NFPOs in terms of actual outcomes and impacts (Kanter and Summers, 1987; Drucker, 1990); clear disclosure of fund-raising and executive compensation practices (Hills Bush, 1992; Kahn, 1992); and fulfilling obligations associated with tax exemptions (Ackerman, 1982; Simon, 1987). A number of writers (Gibson, 1978; Gray, 1983; Hedlund and Hamm, 1978; Hyndman, 1990) have argued that there may be a tendency to overemphasise the discharge of accountability through traditional financial accounts at the expense of wider performance information. The importance of performance information in the discharge of accountability by public sector organisations is argued forcefully by Mayston (1985), Miah (1991) and Boyne and Law (1991). In the latter case, it is suggested that in the absence of such information (p. 179) ‘the concept of accountability and indeed the whole democratic process is simply a sham’. Gray (1984, p. 84), in arguing for more disclosure of performance information by charities, expressed the view that ‘financial accounts can provide useful additional information. However, the financial accounts can only play a subsidiary role’. Despite problems relating to the utilisation of performance measurement and the development of performance-reporting systems within charities (such as the difficulty in setting explicit and quantifiable objectives, the weighting of objectives, problems in measuring quality, the ability to manipulate information, a concentration on quantification

C. Connolly, N. Hyndman / The British Accounting Review 36 (2004) 127–154

131

and a lack of regard for qualitative factors), performance measures are needed to allow the charitable sector to justify its existence (for a fuller discussion of the difficulties of measuring performance in the public sector, see Likierman, 1993). Unless performance measures are in place, it is difficult for the charitable sector as a whole, or for individual charities in particular, to counter criticisms of poor management and ineffectiveness. Performance measures give a visibility to the resources, activities and achievements of a charity, thus enabling informed discussions and decisions. In addition, the need to discharge accountability encourages management to concentrate on the issues that are of importance to those stakeholders who are outside the immediate management of the organisation and often provide the resources for the organisation to function. Edwards and Hulme (1995, p. 9) assert that the absence of accountability, and in particular performance accountability, ‘begins to make the likelihood of ineffective or illegitimate actions by an organisation much more probable’. In the context of charities, this suggests that where accountability is weak, management might have limited incentive to manage the charity’s funds efficiently and effectively.

3. Previous research It has been argued previously that the information needed to discharge accountability by charities may be extremely varied, and two important aspects of this are externally reported financial statements (an aspect of discharging financial accountability) and information relating to the performance of the charity (an aspect of discharging performance accountability). As mentioned earlier, previous research has focused on the financial information contained in charity financial statements, where it is argued that the quality of financial statements can be judged in terms of their compliance with the extant SORP (or generally accepted accounting principles). Broadly speaking, these studies indicated a degree of diversity of accounting practices by charities, particularly in the area of accounting measurement and disclosure. However, there was some evidence that the quality of financial accounts was improving over time (as judged by the increasing adherence to SORP requirements) and that the quality of the financial reporting of large charities was better than that of small charities (with size being measured in terms of total income). In addition, the only study to compare British and Irish charities (Connolly and Hyndman, 2001) suggested that, even after making allowances for size differences, British charities were more compliant with accounting recommendations than their Irish counterparts. With respect to the reporting of performance information by charities, there have been few studies. Based upon an analysis of charity annual reports and a survey of charity contributors, Hyndman (1990) sought to identify the information that is routinely made available to contributors through the annual report and the most important information sought by them. In the study, contributors were asked to rank 14 types of financial and non-financial information, including frequently disclosed information, in terms of importance to them. Hyndman found that while audited financial statements dominated reporting by charities, contributors viewed other information, particularly that relating to performance, as most important. The seven most important information types identified

132

C. Connolly, N. Hyndman / The British Accounting Review 36 (2004) 127–154

were (in order of importance): a statement of goals of the charity; information relating to the general problem or need area with which the charity was dealing; administration cost information (a financial indicator of efficiency); measures of the outputs of the charity; non-financial efficiency measures; a statement of the current objectives of the charity; and a statement of the future objectives of the charity. Three of the four least important information types were (in decreasing order of importance): audited operating statement; audited balance sheet; and audited funds flow statement. However, two of these were the most frequently disclosed (operating statement and balance sheet) of all the 14 information types, while the other (funds flow statement) was the fourth most frequently disclosed information type. Hyndman argued that many of the most important types of information sought by contributors relate to the assessment of performance, and suggested that the seven most important information types (indicated above) could all be related to past or future performance. Whereas Hyndman (1990) indicates a lack of external reporting of performance information by charities, Wise (1995) provides some evidence of the paucity of performance information available internally to charity managers. Wise asked the representatives of 54 charities (subdivided into small, medium and large charities) to provide details of the information made available to, and used by, managers in their charity. One of the main findings of the research was that while many of the charities had some non-financial performance information available at routine management meetings (with large charities having more than small charities), an examination of the detail of the information highlighted certain deficiencies. In particular, most of the performance indicators focused upon input or output volumes rather than concepts of efficiency or effectiveness. Furthermore, while the majority of respondents believed that their trustees were able to assess the performance of their charity, few could suggest how performance improvement might be measured. Wise asserted that these findings demonstrated significant weaknesses in the information systems of charities and undermined efforts to plan and control effectively.

4. Empirical work Annual reports are generally recognised as key documents in the discharge of accountability to external users. This paper, while recognising the importance of traditional financial statements, focuses on information contained in annual reports of British and Irish charities outside of the financial statements.1 This includes certain legal and administrative information which the 2000 SORP (Charity Commission, 2000) recommends should be in the trustees’ report, together with quantitative and qualitative 1 The term ‘financial statements’ is used in this context to include information contained in the financial documents (SOFA, Balance Sheet, Cash Flow Statement and the related notes) of the charity. This is the way the term is used in the audit reports of the charities (or indeed the way it would be used in the audit report of any company). It is possible that some of the performance information included in the analysis may have been included in a section of the annual report and financial statements that had a heading ‘Financial Statements’ (although this was rarely the case). Therefore, for example, if the ‘names of the principle officers’ was included anywhere in the annual report and financial statements, it was counted as having been disclosed.

C. Connolly, N. Hyndman / The British Accounting Review 36 (2004) 127–154

133

performance information. The results are presented on the basis of both income size2 and jurisdiction. 4.1. Method This research attempted to identify the extent to which basic background information and more specific performance information is provided in the annual reports of charities, distinguishing between British and Irish charities. Basic background information can be viewed as being important in providing an appropriate context for the understanding of the performance of a charity. In addition, the research attempts to explore whether the extent of the reporting of performance information is related to size. A matched set of British and Irish charities (in terms of income size) was collected. The analysis reported in this paper focuses on British and Irish charities in two size related groups, small (those with incomes between £100,000 and £650,000) and large (those with incomes greater than £650,000). It excludes those charities with incomes less than £100,000 because, depending on jurisdiction, charities with incomes of this magnitude may face less strict legislative controls and accounting recommendations (for a discussion of the historical background and legal framework in England and Wales, Scotland, NI and the RoI, see Connolly and Hyndman, 2001). In total, usable responses from 332 charities were analysed, compared and reported (80 large British charities, 15 large Irish charities, 169 small British charities and 68 small Irish charities).3 A checklist, focusing on information outside the financial statements, and distinguishing between basic background information and more specific performance information, was designed for use on these annual reports.4 Points of reference used for the development of this were the recommendations included in the various SORPs, guidance provided by The Charities Aid Foundation (2001) and the information types identified by Hyndman (1990) as being important to contributors. For analysis purposes, information was considered disclosed if it was highlighted in some way (highlighted copy) and not contained in the main narrative of the annual report (general narrative). The extent to which general narrative information was disclosed is not reported in the tables below since it may be difficult for the user to isolate and use this information as it is often embedded in a mass of subjective comment. In addition, it is sometimes given in a form that may make it difficult for a user to understand its significance. Definitions and rules were also developed for classifying the highlighted copy information in order to reduce the impact of subjectivity. The rules included: information presented in more than one way is only 2 Because of differences in currency in the Republic of Ireland (RoI) and Northern Ireland (NI), income figures in RoI are in punts (£IR) whereas income in NI (and in Britain) is in sterling (£Stg). For convenience, exchange differences are ignored and the symbol £ used to denote £IR in RoI and £Stg elsewhere. 3 The research reported in this paper largely uses a common database with research reported elsewhere which focused solely on financial accountability (Connolly and Hyndman, 2001), although more extensive reporting, and different categorisation, of the data set is used in this paper. A more detailed explanation of the rationale behind the collecting of the sample, the responses of each group of charities to requests for information, and the testing for non-response bias is contained in the 2001 paper and not repeated here. 4 A copy of the checklist is available from the authors on request.

134

C. Connolly, N. Hyndman / The British Accounting Review 36 (2004) 127–154

counted once (in its most detailed form); when there is an efficiency target and the actual achievement of efficiency is given, this is counted as a measure of efficiency only; and when there is a quality target and the actual achievement of quality is given, this is counted as a measure of effectiveness. This approach ensured that, as far as possible, terms were being used in the same manner for each charity.5 4.2. Results 4.2.1. Basic background information For external reporting purposes, British and Irish companies are bound by law to produce a set of accounts for publication. This must include a directors’ report, and may also include a chairman’s statement, which, according to Pendlebury and Groves (1994), should provide a narrative supplement to the financial statements. The directors’ report is required to provide information relating to the company’s activities during the period under review, its directors and employees, and likely future developments. Pendlebury and Groves claim that such information is an influential part of the annual report and is widely read by shareholders. In the context of charities, the trustees’ report fulfils a similar role to that of the directors’ report. In a similar fashion, and specifically focusing on charities, CAF (2001) suggests that certain basic background information should be provided at the front of each charity annual report. Like company legislation, this advocates the inclusion of: details as to how the charity is constituted; its aims and objectives; a full list of the trustees and officers of the charity; and an address at which the charity may be contacted. Furthermore, the 1988, 1995 and 2000 SORPs (ASC, 1988; CARC, 1995; Charity Commission, 2000) also recommend that certain legal and administrative information should be included by charities in their annual reports in order to enable the reader to understand better a charity’s structure and activities. The 2000 SORP suggests that a report describing what the charity is trying to do and how it is going about it should be provided. It is recommended that, as part of the report, there should be a statement outlining broad background information relating to the charity, including: an indication of the nature of the governing document (e.g. trust deed, memorandum and articles of association, Charity Commission Scheme or Royal Charter); how the charity is (or its trustees are) constituted (e.g. limited company, unincorporated association or trustees incorporated as a body); the names of all of the charity’s trustees (including the officers of the charity); and the address of the principal/registered office of the charity. Furthermore, the SORP recommends that the trustees’ report should assist the reader in interpreting the accounts and relating the numerical aspects to the organisational structure and activities. Table 1 highlights the basic background information provided by charities, distinguishing by jurisdiction and charity size. Seven items of information, drawn from the above discussion, were identified and reported. These were: an indication of the governing instrument; details of how a charity is constituted; review of the year; an explanation of the salient features of the accounts; names of trustees; names of principal officers; and address of principal or registered office. These information types, although 5

A copy of the rules is available from the authors on request.

Disclosure

Item (a) Indication of governing instrument (b) Details of how charity constituted (c) Review of the year (d) Salient features of accounts explained (e) Names of trustees (f) Names of principal officers (g) Address of principal/registered office

Significance

Large British (80)

Small British (169)

Large Irish (15)

Small Irish (68)

No. %

No. %

No. %

No. %

80 80 78 77 80 80 80

153 153 152 147 163 164 150

11 11 12 11 15 15 13

47 47 51 45 60 63 55

100 100 98 96 100 100 100

91 91 90 87 96 97 89

73 73 80 73 100 100 87

The values in parentheses represent total number of annual reports analysed.

Large Large Large Small Small Large British vs British vs British vs British vs British vs Irish vs large Irish small British small Irish large Irish small Irish small Irish

69 69 75 65 88 93 81

1% 1% 1% 1% No. No. 1%

1% 1% 5% 5% No. No. 1%

1% 1% 1% 1% 1% 5% 1%

5% 5% No. No. No. No. No.

1% 1% 1% 1% No. No. No.

No. No. No. No. No. No. No.

C. Connolly, N. Hyndman / The British Accounting Review 36 (2004) 127–154

Table 1 Basic background information

135

136

C. Connolly, N. Hyndman / The British Accounting Review 36 (2004) 127–154

not performance information, are viewed as important by CAF (2001) and the Charity Commission (2000), and provide a context to understanding the performance of a charity. For example, information relating to the governing instrument may outline the broad objects of the charity, how it relates to its external environment and how it is governed internally. Details of how a charity is constituted may indicate the extent to which an organisation is centrally recognised as having provided useful service over many years (as in the case of a charity incorporated by Royal Charter), and therefore, may lend credibility to its activities. A review of the year may attempt to explain and interpret the activities of the charity. An explanation of the salient features of the accounts may indicate the degree to which a charity has ‘lived within its means’ and the extent to which funds are presently available to expand its provision of services. The names of trustees and principal officers of the charity may help to promote confidence in the operations of the charity. Finally, the address of the principal or registered office provides a contact point for external parties and indicates a willingness to engage in communication. The numbers and percentages of charities providing each of these seven items of basic background information are reported in Table 1. In addition, chi-square tests were carried out to identify differences between the jurisdictions and the size categories. This involved six pair-wise comparisons (large British vs large Irish, large British vs small British, large British vs small Irish, small British vs large Irish, small British vs small Irish, and large Irish vs small Irish). Indications of significant differences (at the 5 and 1% levels) are presented. With five of the seven items (items a –d and g), the results shown in Table 1 display a very distinctive trend. Large British charities achieve the highest level of disclosure for each of the items, followed by small British charities, then large Irish charities and, lastly, small Irish charities. With each of these five items, the differences between large British and small British are significant at either the 1 or 5% level, suggesting size being a factor in disclosure by British charities with respect to these items of basic background information. For example, 98% of large British charities included item c in their annual reports, compared with 90% of small British (a difference that was significant at the 5% level). Regarding jurisdiction differences relating to these five items, small British charities had significantly greater disclosure (at the 1% level) than small Irish charities with four of the five items (items a –d). For example, 91% of small British charities provided item b, compared with 69% of small Irish charities. A similar pattern exists between large British and large Irish (with all five items showing significant differences at the 1% level) and, indeed, between small British and large Irish (with two of the items showing significant differences at the 5% level). Overall, this suggests that British charities provide more basic background information than their Irish counterparts. With respect to two of the seven items of basic background information (items e and f), large charities, both British and Irish, had greater disclosure than small charities, with small British charities showing greater disclosure than their Irish counterparts. For example, 100% of large British and large Irish disclosed item e in their annual reports compared with 96% (British) and 88% (Irish) of the small charities. When comparisons are made between large and small Irish charities, differences are not statistically significant with any of the seven items of basic background information, although with each of these the level of disclosure by large Irish charities was higher than

C. Connolly, N. Hyndman / The British Accounting Review 36 (2004) 127–154

137

that with small Irish charities. For example, item a was disclosed by 73% of large Irish charities compared with 69% of small Irish charities. 4.2.2. Performance information As was discussed earlier, the disclosure of performance information is important in discharging the accountability of charities. A number of studies of performance in the wider not-for-profit sector have modelled performance in terms of a production model consisting of three stages (inputs, outputs and results), with performance being judged in terms of efficiency and effectiveness (Brace et al., 1980; AAA, 1989; Hyndman and Anderson, 1995). While there are no standardised definitions for these terms, the following explanation may aid understanding. Inputs are the resources used to provide a product or service (e.g. staff, money, equipment). Outputs are the immediate products or services generated by the organisation; for example, number of children fed or number of individuals trained. Results represent the impact of the product or service on society; for example, a healthier population or safer roads. Effectiveness is concerned with the relationship between the outputs or results of an organisation and its objectives.6 Efficiency is the ratio of outputs to inputs, or the amount of input per unit of output. Each of these last two terms is almost always used in a comparative, rather than an absolute sense. For example, it is not normally said that an organisation is 70% efficient, but rather that it is more (or less) efficient than a comparable organisation, or that it is more (or less) efficient than it was last year, or that it is more (or less) efficient than budgeted for. The empirical investigation by Hyndman (1990) provided evidence that users of charity reports viewed performance information as paramount in meeting their information needs. As a conclusion to the study, it was suggested that a model of reporting be adopted by 6

While one may argue, ultimately, effectiveness should indicate the extent to which the ‘result’ or ‘outcome’ matches objectives; this is rarely the restricted manner in which the term is used in practice for planning and reporting purposes (although such views on the meaning of effectiveness may well be central when evaluating the achievements of an organisation or policy over the longer term). As is argued in a number of NFPO publications when considering performance, the ultimate result may occur much later than the intermediate output and is often affected by various factors unrelated to the action of the organisation (cf. HM Treasury, 1992, 2001; Cabinet Office, 2002; Charity Commission, 2002). For example, guidance related to performance targeting and measurement in the public sector distinguished between ‘direct outputs’ (a step towards producing something; for example, an operation conducted in a hospital) and ‘final outputs’ (the ultimate result or outcome; for example, the future quality of life of a patient following an operation) and argued that, because of the difficulty of targeting and measuring final outputs, ‘for the purposes of determining targets it is preferable to concentrate on the direct outputs’ (HM Treasury, 1992, p. 10). Similar themes are found in more recent guidance, where it is stated that (HM Treasury, 2001, p. 3) ‘ultimately, we are aiming to improve outcomes… but sometimes outcomes can be hard to measure.’ This discussion, because of length restrictions, is beyond the scope of this paper (for a fuller coverage of this issue see Connolly and Hyndman, 2003). It is interesting to note that the Charity Commission for England and Wales has recently set up an Annual Reporting Advisory Group to consider how performance reporting and governance reporting might be achieved (this group met for the first time in June 2003), and also provided some very broad guidance (Charity Commission, 2002). This guidance includes, among many other issues, some discussion of the difference between outputs and outcomes. In this research, the definition used to identify effectiveness is the relationship between the outputs or results of an organisation and its objectives (based on Anthony and Young, 2003, p. 12). As a consequence, it identifies output measures (or result measures) against objectives (or targets) as measures of effectiveness, as was done in Hyndman and Anderson (1995) and Hyndman and Eden (2001b). Had a more restricted definition been used (for example, outcome vs objective), there would have been very few measures of effectiveness recorded in this research.

138

C. Connolly, N. Hyndman / The British Accounting Review 36 (2004) 127–154

charities, with the model focusing heavily on the reporting of performance information. Essential performance-related aspects of this model, modified slightly to reflect the earlier discussion on performance information and some later reflections by the model’s originator (Hyndman and McKillop, 1999), are used to capture aspects of the reporting of performance by charities in this study. Table 2 focuses on the disclosure of selected performance information by charities, analysed by jurisdiction and by size. Eight items of performance information, based on the above discussion, were identified and reported. These were: goals and objectives; inputs; outputs; results; efficiency (including administration cost percentage information); effectiveness; future target information; and budget information. Examples of each of the eight items of performance information, shown by the charity making the disclosure, are included in Appendix A (goals and objectives, inputs, outputs and results) and Appendix B (efficiency, effectiveness, future target information and budget information). Goals are defined as intended outputs in broad terms, whereas objectives are specific, measurable targets to be achieved within a particular time frame (Anthony and Young, 2003). In a co-ordinated system of planning, objectives should be in harmony with the goals of the organisation. Objectives are essential in measuring the effectiveness of the organisation, where actual outputs (or results) are compared with objectives. As was highlighted earlier, most studies of performance in the not-for-profit sector have modelled performance in terms of a production model consisting of three stages (inputs, outputs and results) and it is suggested that the capturing of information at these stages is necessary to measure both efficiency and effectiveness. Using a similar model in relation to public sector organisations, Hyndman and Anderson (1998) argued that while the component parts of performance (i.e. inputs, outputs and results) may be useful in gaining some insight into performance, the higher measures of performance (i.e. efficiency and effectiveness) are more critical. While a measure of efficiency normally relates the inputs (financial or non-financial) to outputs (or results), an alternative, or supplementary, indicator of efficiency could be developed by relating administration costs to total expenditure, a performance information type that is viewed by contributors as being of considerable importance (Hyndman, 1990). Overall, with respect to efficiency and effectiveness measures in charities, Bird and Morgan-Jones (1981) expressed the view that such information is needed in assessing how resources have been used in meeting the needs of beneficiaries. The extent to which these performance information items, many of which were included in Hyndman (1990) model, are disclosed by charities is disclosed in Table 2. While the above information items focus on present or past performance, CICA (1980) and Bird and Morgan-Jones (1981) contend that performance information relating to what a charity intends to do in the future is also important. For example, this information would be useful for those deciding whether or not to fund, or continue to fund, a charity’s activities. Such information could include a statement of future objectives, which may be an expression of planned future output, possibly covering the next year. As discussed above, this would permit future performance to be compared with information on planned performance that had been published earlier. Furthermore, it may be useful to outline overall plans in terms of a financial budget. Such information would provide an indication of revenue requirements for the ensuing period. Indeed, Anthony and Young (2003)

Disclosure Large British (80)

Item (h) Goals and objectives (i) Inputs (j) Outputs (k) Results (l) Efficiency (m) Effectiveness (n) Future target information (o) Budget information

Significance Small British (169)

Large Irish (15)

Small Irish (68)

No.

%

No.

%

No.

%

No.

%

76 64 77 74 4 54 8 19

95 80 96 92 5 67 10 24

150 75 130 94 9 43 4 5

89 46 77 56 5 25 2 3

10 5 7 5 1 0 0 0

67 33 47 33 7 0 0 0

31 6 18 16 1 2 2 1

46 9 29 24 2 3 3 2

The values in parentheses represent total number of annual reports analysed.

Large British vs large Irish

Large British vs small British

Large British vs small Irish

Small British vs large Irish

Small British vs small Irish

Large Irish vs small Irish

1% 1% 1% 1% No. 1% 1% No.

No. 1% 1% 1% No. 1% 5% 1%

1% 1% 1% 1% No. 1% 5% 1%

5% No. 1% 1% No. No. No. No.

1% 1% 1% 1% No. 1% No. No.

No. No. No. No. No. No. No. No.

C. Connolly, N. Hyndman / The British Accounting Review 36 (2004) 127–154

Table 2 Selected performance information

139

140

C. Connolly, N. Hyndman / The British Accounting Review 36 (2004) 127–154

consider the preparation of a budget for an NFPO as being more important than for a business enterprise. This is because, in an NFPO, a larger proportion of expenditures is discretionary and the conditions affecting the operation of an NFPO are more predictable. Both of these future-oriented performance information items are included in Table 2. As in the case of basic background information, the numbers and percentages of charities providing each of the eight items of selected performance information are reported in Table 2. Chi-square tests were carried out to identify differences between the jurisdictions and the size categories, as before, with significant differences highlighted. As can be seen in Table 2, there was substantial disclosure (over 50% of charities in at least one of the groups) with five of the eight performance information items (items h –k and m). As was the case with many of the basic background information types, a very distinctive trend appears. Large British charities achieve the highest level of disclosure (all five items), followed by small British (all five items), then large Irish (four of the items), and, finally, small Irish (four of the items). For example, as can be seen in Table 2, item j was disclosed by 96% of large British charities, compared with 77% (small British), 47% (large Irish) and 29% (small Irish) of the other groups of charities. With respect to British charities, there were statistically significant differences between large and small charities with four out of five of these items (all at the 1% level), suggesting that size is a factor with disclosure. In relation to jurisdiction, differences between Britain and Ireland were stark. For example, small British charities had greater disclosure than small Irish charities with all five items, and each of the differences was significant at the 1% level. These differences suggest that British charities provide more performance information in their annual reports than Irish charities.7 With respect to the other three information types (items l, n and o), disclosure was not extensive (disclosure was less than 25% by all groups of charities). In fact, with both items l and n, disclosure was very low across all groups. For example, the disclosure of item n was 10% for large British charities, and 3% or less in the other three groups (large Irish, small British and small Irish). Although the number of charities disclosing these three information types was low, generally large British charities had greater disclosure than small British charities, and British charities in total had greater disclosure than Irish charities. 7 It is possible that differences in the extent of the reporting of performance reporting by charities may be affected by the type of charities reporting. For example, some types of charities may have greater incentive to produce performance information (for example, charities operating on the fringe of the public sector) while other charities are demand led and will have greater difficulty in providing such information (for example, disaster relief charities). A different ‘mix’ of charities in Britain and Ireland could possibly provide an explanation of differences in disclosure of performance information. To test for this, the data was reworked by charity type using a taxonomy based on three groups (charities operating on the fringe of the public sector—type 1; charities undertaking very specific activities such as cancer research and animal welfare—type 2; and charities responding to crises such as famine relief—type 3. Overall, differences were extremely minor between the charity types. Indeed, with 15 information items and three types of charity (giving 45 pair-wise comparisons) only two statistically significant results were identified (items i and o) when comparing charity types 2 and 3. In the interests of brevity, the detailed results of this reworking are not reported in the paper but are available from the authors on request. Overall, they suggest that the type of charity does not have a major influence on the reporting of the 15 information items.

C. Connolly, N. Hyndman / The British Accounting Review 36 (2004) 127–154

141

5. Analysis and discussion In this paper, it is argued that charities have an accountability to those outside their immediate management, and that the annual report is an important medium through which this accountability can be discharged. It is further argued that while the information disclosed in the financial statements may be of importance, other wider information, particularly relating to performance, is paramount in discharging accountability. Research by Hyndman (1990) provides backing for this viewpoint. Overall, the research seeks to ascertain the extent to which performance accountability is discharged through the medium of the annual report. Specifically, the empirical study reported in this paper has as its objective the analysis of recent annual reports of British and Irish charities to ascertain the degree to which basic background information and more specific performance information is used in external reporting. The analysis and discussion is arranged in the following sub-sections: changes over time; possible reasons for limited disclosure by charities; particular issues relating to Irish charities; and size factors. 5.1. Changes over time A comparison of this study with Hyndman (1990) study provides some limited insights into how performance reporting by charities has changed over time. Table 3 summarises the disclosure of the eight performance information items (as disclosed in Table 2) across all charities, and compares the results with those of Hyndman (1990) in cases where comparisons can be made, thus highlighting areas where disclosure has changed over time. In each of these studies, identical definitions of the performance-related terms were used. Although the two studies are not directly comparable (most obviously because the 1990 study focused solely on large British charities), five of the eight performance-related items analysed in this study were also analysed in the earlier Hyndman study. In all but one of the performance information items common to both studies, an increase in disclosure was evident over time. With two of the items, disclosure increased considerably. For example, Table 3 Summary of Results of Selected Performance Information This study

(h) Goals and objectives (i) Inputs (j) Outputs (k) Results (l) Efficiency (m) Effectiveness (n) Future target information (o) Budget information a b

Hyndman, 1990

No.a

%

No.b

%

267 150 232 189 15 99 14 25

80 45 70 57 5 30 4 8

62

38

47

29

35

21

1 8

1 5

Total No. of annual reports analysed: 332. Total No. of annual reports analysed: 163.

142

C. Connolly, N. Hyndman / The British Accounting Review 36 (2004) 127–154

with respect to item j, 70% of charities in this study disclosed at least one measure of output in their annual reports compared with only 29% in the earlier study. With two other items (items n and o) disclosure increased, but from a very low base. Such improvement overall in the reporting of performance information, while possibly being welcomed, is from a rather limited initial point of comparison. A possible explanation for this improvement may have been the increasing articulation of the importance of performance reporting by the CAF, evidenced by their promotion of the Charities Published Accounts Awards and the related guidance provided (CAF, 2001). The fact that performance reporting is greatest with those charities that are most likely to be involved in such high profile award competitions (large British charities, Table 2) may suggest this. Other possible reasons for improvement may be the influence of practices in the public sector, where performance-reporting systems have developed more strongly, and the increasing profile of such techniques as the Balanced Scorecard, especially in the private sector. It is interesting to note that with respect to the reporting of efficiency (see Table 3), disclosure decreased over time, from 21 to 5%.8 Efficiency measures, particularly those relating administration costs to total costs (which are captured as measures of efficiency in this study), are often perceived as potentially damaging headline figures. Beattie and Jones (1994, p. 221) have argued that charities may ‘have incentives to minimise or deemphasise their administration expenses vis-a`-vis their charitable expenditures’. This possibly encouraged charities to engage in ‘gaming’ by the rather judgemental classification of certain expenditures, often inconsistently, frequently resulting in an understatement of the administration cost percentage. Such behaviour made it difficult to gain an accurate impression of the efficiency of charities, something commented on by a number of writers over 20 years ago (Austin and Posnett, 1979; Bird and Morgan-Jones, 1981). This matter attracted attention in the original SORP 2 (ASC, 1988) and its subsequent revisions (CARC, 1995; Charity Commission, 2000), and led to much greater prescription regarding classification. Given this scenario, and given the potentially sensitive nature of such a disclosure, this may have resulted in charities being more hesitant to publish such headline information, although, as discussed previously, the more 8 It should be noted that, using the rules of classification, only charities that disclosed an efficiency measure (defined as ‘a ratio measuring the relationship of outputs to inputs. Sometimes expressed as an amount of output per unit of input, a cost per unit of output, a ratio relating overhead costs to total expenses or a ratio relating fundraising costs to total funds raised’) in the annual report were counted as having disclosed an efficiency measure. As indicated earlier in the paper, only information contained in the annual report outside the financial statements was analysed. A rule used for classification was that if an arithmetic calculation was made, for example dividing administration costs by total costs to give an administration cost percentage, this was counted as an efficiency measure. However, if total costs and total administration costs were disclosed separately, for example, as figures in the operating statement, but not specifically related to one another in an arithmetic calculation (a ratio), this was not included as a measure of efficiency. The analysis was conducted in this manner even though it might be argued that if the overhead costs and total costs were provided as figures within the operating statement (as they were for over 90% of British charities and approximately 50% of Irish charities in Connolly and Hyndman (2001) paper) an efficiency ratio could be calculated. However, given the fact that the majority of users of charity accounts, as is the case with the majority of users of commercial accounts, have limited understanding of audited financial statements (Bird and Morgan-Jones, 1981; CICA, 1980), disclosure within the operating statement as individual figures within a mass of other figures, was not deemed a communication of a measure of efficiency in this research.

C. Connolly, N. Hyndman / The British Accounting Review 36 (2004) 127–154

143

determined and experienced user may well be able to extract such information from the financial statements. Over 10 years have elapsed since the Hyndman (1990) study which identified that charities’ annual reports were deficient in meeting the information needs of users and that greater disclosure of performance information was needed to improve the situation. While there has been some limited increase in performance reporting through the medium of the annual report since that time, progress has been slow. Given that developments in the public sector with respect to performance reporting have been considerable, possibly facilitated by a more proactive central support function and greater prescription (Hyndman and Anderson, 1998), the change in charity reporting is perhaps disappointing. While some effort has been made to highlight the need for charities to disclose more performance information (CAF, 2001), the concentration on specific guidance has been on financial accounting matters (ASC, 1988; CARC, 1995; Charity Commission, 2000), information that was identified by Hyndman (1990) as being considerably less relevant to users of charity reports than performance information. Indeed, it could be argued that this focus on financial accounting matters may have detracted from a meaningful debate relating to the relevance of the information content of charity annual reports. 5.2. Possible reasons for limited disclosure by charities As with all cases of information provision, cost/benefit considerations should be considered. It could be argued that the provision of background information and performance information, which are voluntary disclosures, imposes costs on a charity, and the lack of disclosure is indicative of an efficient charity. However, this argument ignores important reasons for disclosing such information, some of which are related to economic incentives of disclosure. In terms of businesses, voluntary disclosure above and beyond regulatory requirements is often viewed as having a significant effect on an organisation’s financial success, particularly in terms of reducing the costs of finance and improving the access to financial markets (FASB, 2001). A number of studies have suggested that there are major market incentives to disclose information voluntarily, although managers’ attitudes to the perceived costs and benefits of such disclosure are important (Gray et al., 1990; Healy and Palepu, 1995). Ignoring accountability issues, it could be argued that in an environment where charities compete for contributions, the voluntary disclosure of performance information, possibly emphasising past performance and future plans, may enhance its attractiveness to existing and potential donors. With respect to efficiency information in particular, Callen and Falk (1993, p. 50) argued that large contributors would be interested because …there is probably little return from association with an inefficient, wasteful charity, especially if such wastefulness becomes known to the public at large. This concern motivates large donors to monitor the charity’s efficient use of resources. Similar sentiments were expressed by Bird and Morgan Jones (1981, p. 133) in the context of all contributors, where it was suggested that individuals ‘tend to give to charities

144

C. Connolly, N. Hyndman / The British Accounting Review 36 (2004) 127–154

which they are satisfied are doing the work they wish to assist and are spending the money that is given to them effectively…rather than wasting it’. Such scenarios suggest, for example, that the publication of efficiency information (in cases where the charity has operated efficiently) may well increase its attraction to potential contributors. Conversely, information indicative of poor efficiency, or no efficiency information at all, may well dissuade potential contributors. Similar arguments could be made in the case of other performance information. A lack of adequate performance accountability, particularly if it were highlighted in the media, could discourage potential donors. Related to the above in terms of economic impact, is that while it may be of some concern that charities do not include significant amounts of performance information in their annual reports, what may be of greater concern is that this possibly indicates that such information is unavailable in the internal/management accounting system. Such an argument was made by Gray (1983) with respect to the wider not-for-profit sector, and is alluded to by Wise (1995) in relation to the charity sector. It has been suggested that the need to discharge accountability through the publication of performance information acts as a disciplining pressure and encourages management to concentrate on the issues that are of importance to those stakeholders who are outside the immediate management of the organisation and often provide the resources for the organisation to function. Without pressure being brought to bear to discharge accountability through the disclosure of performance information, it has been argued that this can lead to managers of charities pursuing their own goals at the expense of the goals of the donors or the stated goals of the charity (Williams, 1984). In this case, it was suggested that (p. 11) Managers of charities whose private goals are not in accord with those that donors would prefer management to have, may take comfort from this displayed lack of interest in non-reporting…There may…be little incentive for management to develop meaningful accounting and reporting systems, or to devise appropriate methods of internal control. Similarly, Eden and Hyndman (1999), in reviewing the benefits and difficulties of reporting performance information in a public sector context, argue that while a lack of disclosure may suit some managers, the external pressure provided by the provision of performance information encourages a focus for the organisation and provides a catalyst for managers as they seek to improve performance. Making a comparable point, but in a more negative tone, Fox (1992) states that although accountability is viewed as desirable, empirical investigation commonly indicates that both leaders and subordinates in public and private organisations seek to avoid accountability. A more extreme version of the above argument suggests that in NFPOs with limited performance measurement and performance reporting, and weak forms of accountability, there may be no incentive to manage efficiently, and managerial shirking will result (Hansmann, 1996). With respect to charities, it could be argued that any cost savings in terms of less extensive disclosure (as previously intimated may perhaps be the case) might be more than eliminated because of less efficient management (and managerial shirking) caused by a lack of discipline and focus in the absence of performance measurement and

C. Connolly, N. Hyndman / The British Accounting Review 36 (2004) 127–154

145

performance-reporting systems. Indeed, recent research into the views of chief executives in public sector organisations provides evidence that the development and use of performance measurement and reporting systems (often at significant cost) can provide major benefits to many stakeholders, including the organisation itself, its clients and those who provide funding, and lead to a more effective and efficient organisation (Hyndman and Eden, 2001a). It is possible that the rather limited performance reporting by the charities in this research could result from a lack of awareness regarding the information needs of users or, alternatively, from charities being unwilling to disclose many of the items. However, regarding a lack of awareness, Hyndman (1990) study shows clearly that providers of charity reports are very aware of the dominant information needs of users of charity reports and are conscious of the relevance of performance information to such parties. With respect to unwillingness to disclose, there could be a number of reasons for this. For example, charities may be satisfied with existing reporting procedures, they may fear that such information would be misinterpreted or they may view that the cost of providing such information is disproportionately high. Whether all or any of these reasons justify nondisclosure is debatable. A more probable reason for non-disclosure may be a general complacency with existing reporting practices, since it is difficult to understand why, for example, a statement on goals and objectives would be omitted on the grounds of cost or that it might be misinterpreted. 5.3. Particular issues relating to Irish charities While the reporting of performance by all charities in this study was limited, the results indicate that the annual reports of British charities provide significantly greater levels of performance information than their Irish counterparts. This suggests greater levels of performance accountability. For example, this is clearly seen when small British charities are compared with small Irish charities. With respect to the five performance items that were most substantially disclosed by charities, small British charities had greater disclosure than small Irish charities, with differences being statistically significant in all cases. Similar trends are seen when large British charities are compared with large Irish charities. These results are consistent with those of a financial accounting study by Connolly and Hyndman (2001) on the impact of SORP 2 on British and Irish charities which found that compliance by Irish charities with provisions of both the original and revised SORPs was much less than their British counterparts. Possible reasons for the less extensive disclosure of performance information in Ireland may be due to: the weaker legislative framework in Ireland compared to Britain, possibly reflecting very little government focus on the sector; the relative lack of scrutiny of Irish charities; and the comparative lack of research into charity accounting practices. With respect to the comparative legislative frameworks, the regulatory environment with respect to charities in England and Wales (where the vast majority of British charities are located) is fairly strong. For example, both the 1992 and 1993 Charities Acts and the related Charities (Accounts and Reports) Regulations

146

C. Connolly, N. Hyndman / The British Accounting Review 36 (2004) 127–154

1995 (which applies to unincorporated charities), although not specifically requiring all charities to adopt the SORP, refer to the applicability of the SORP with respect to particular matters. NI charities are governed by the Charities Act (NI) 1964 and the Charities (NI) Order 1987, while the RoI Charities Acts 1961 and 1973 govern charities in the RoI. These governing instruments in both NI and the RoI obviously make no reference to the SORP because they predate it. Furthermore, in both NI and the RoI, there are no regulations governing either the form or the content of charities’ accounts. Interest in the scrutiny of charities in Britain began in the late 1970s (National Council for Voluntary Organisations (NCVO), 1976; Austin and Posnett, 1979) and gathered momentum with the publication of Bird and Morgan-Jones (1981). These publications often formed the basis, or at least provided a catalyst, for a plethora of government, quasigovernment and trade association reports and recommendations on the control and monitoring of the charity sector in Britain (for example, National Audit Office, 1987; Woodfield Report, 1987; Public Accounts Committee (PAC), 1988; NCVO, 1990; Home Office, 1999; Cabinet Office, 2002). However, the different systems of charity administration in England and Wales, Scotland, NI and the RoI have resulted in different levels of scrutiny across each jurisdiction. The Charity Commissioners for England and Wales regulate charities in England and Wales. Although there is no charity commission, the Inland Revenue in Scotland maintains an index of charities, and Scottish charities are legally obliged to respond (for a fee) within one month to a request for accounts made by a member of the public. There is, however, no register of charities in Ireland and Irish charities are not required to file accounts under charity legislation, nor do the public have a general right to see the accounts of Irish charities. Charity administration in NI is the responsibility of the Department for Social Development, which has no statutory role with regards to setting up new charities. Its main functions concern giving consent to the disposal of land or buildings by charity trustees, making schemes to change the objects of charities whose original functions can no longer be carried out effectively and giving informal advice to trustees and their solicitors. The Commissioners of Charitable Donations and Bequests perform a similar role in the RoI. While the effectiveness of the Charity Commissioners in England and Wales in encouraging improvements in charity accounting, and the related aspects of the development of performance measurement systems and the external reporting of performance, is unclear, the lack of such a body in Ireland is likely to have resulted in less pressure on charities to improve accounting and accountability. A focus on accounting and accountability in British charities is evidenced by the fact that almost all the previous empirical research (for example: Ashford, 1989; Gambling et al., 1990; Hyndman, 1990; Hines and Jones, 1992; Williams and Palmer, 1998; Connolly and Hyndman, 2000) deals with large British charities. Although each of these studies was primarily concerned with the quality of financial reporting, it is possible that this led to ‘good practice’ being adopted more quickly by many British charities with respect to the reporting of performance information. Early British research, particularly the Bird and Morgan-Jones research report, provided a foundation for the development, and subsequent revisions, of the SORPs (ASC, 1988; CARC, 1995; Charity Commission, 2000) and developments with respect to the control and

C. Connolly, N. Hyndman / The British Accounting Review 36 (2004) 127–154

147

monitoring of charities often complemented the recommendations regarding accounting practice in that direct reference was often made to the need to adhere to the SORP. Each of these reasons by themselves may not be of overriding importance, but taken together, they are likely to have contributed to greater ambivalence to reporting practices in Ireland compared with Britain, and hindered the development of performance measurement and performance-reporting systems in Irish charities. Given the weaker legislative and regulatory framework, a dearth of focus on control and monitoring, and limited investigation by researchers, it is perhaps unsurprising that the reporting of performance information by Irish charities is less extensive than their British counterparts. 5.4. Size factors This paper explored the extent to which performance reporting is related to the size of the charity. With respect to traditional financial accounting information, Williams and Palmer (1998) found that the size of a charity was a determining factor of the quality of financial accounts, quality being defined in terms of compliance with extant recommendations, with larger charities having higher quality reporting. No previous attempt had been made to identify whether this phenomenon also carried over to performance reporting. Regarding basic background information, which is viewed as important in providing a context for the understanding of the performance of a charity, large British charities had greater disclosure than small British charities with all seven information items, with the differences being statistically significant, at normal levels, in five of the seven cases. Similarly with the Irish charities, large charities had greater disclosure than small charities with all seven items, although these differences were not statistically significant. With respect to the selected specific performance information, similar patterns were found. For example, in the five items that had substantial disclosure (items h –k and m), large British charities had greater disclosure than small British charities with all five items, four of the differences being statistically significant at the 1% level. In relation to Irish charities, although large charities had greater disclosure than small charities with four of the five items, none of these differences was statistically significant at normal levels. Overall, this suggests that while accountability for reporting is weak with respect to all charities, it is particularly weak with respect to smaller charities. It is possible that smaller charities have more informal mechanisms than large charities for discharging accountability, with their donors more directly involved in the day-to-day affairs of the organisation. However, a more likely scenario may be that larger charities have responded to greater scrutiny by outside agencies and by potential donors and, as a consequence, have developed more extensive performance-reporting mechanisms. Although research on voluntary disclosure tends to focus on the information role of financial reports for capital markets, it can also help to explain some of the results of this research (in particular relating to the finding that larger charities disclose more performance information than smaller charities). Meek et al. (1995) found that, in

148

C. Connolly, N. Hyndman / The British Accounting Review 36 (2004) 127–154

general, large firms disclosed more than small ones, putting forward a number of reasons for this. Firstly, it was suggested that large firms have lower information production costs, and are more likely to employ highly skilled people to produce it. Furthermore, Meek et al. contended that because large firms often have a more complicated and wider ownership base than smaller firms, they need to disclose more information in order the satisfy the different needs. Also, as firms become multinational, it was argued that this places new demands for information and a greater variety of information. Gibbins et al. (1990) also see size as being a determining factor in the level of voluntary disclosure. They argue that external consultants tend to set the agenda in small companies whereas, in larger companies, more sophisticated consultants are used as an adjunct to the process, that is, to provide technical advice and opinion.

6. Concluding comments and areas for further research An important issue relating to the publication of performance information, and one which is described by Pendlebury et al. (1994, p. 45) in the context of public sector organisations as ‘the most significant technical problem’, is that of the reliability of externally reported performance information. If no verification of the performance numbers reported by a charity is required, then there may be a temptation to present performance in a manner which is perceived as more acceptable to the reader, for example, by exaggerating good performance, regardless of its accuracy. Although, here again, cost/benefit issues need to be considered, given that external parties may use externally reported performance information to make judgements and decisions regarding a charity, then there appears to be a case for some degree of independent verification. Mayston (1985), for example, has argued for bringing to bear similar disciplines on the external reporting of performance measures that are imposed on financial accounting information, i.e. disclosure requirements, external auditing and standard setting after consultation with interested parties. It should be noted that the performance information analysed in this study had not been subject to independent verification, as is the case with performance information reported by the vast majority of NFPOs. This issue has been discussed in the context of the public sector organisations with a suggestion being made that UK public sector organisations should aim to achieve ‘independent validation’ of performance information (PAC, 1997, p. vii), although developments to date have been limited. A fuller discussion of the issue in a public sector context is contained in Likierman (1997). Perhaps unsurprisingly, given the limited disclosure of performance information identified in this study, there has been no significant debate regarding this issue in the charity sector. If the annual reports of charities are to meet the information needs of users, it would seem appropriate that much greater focused guidance relating to performance information should be provided by those concerned with the administration and control of the charity sector, both in Britain and Ireland. This may include the Charity Commission in England and Wales, the CAF in Britain, the Scottish Charities Office, the Department of Social Development in NI and the Commissioners of Charitable Donations and Bequests in the RoI. To expect individual charities to develop meaningful

C. Connolly, N. Hyndman / The British Accounting Review 36 (2004) 127–154

149

and extensive performance-reporting systems is perhaps optimistic. In the context of performance reporting, it is particularly interesting that the Charity Commission in England and Wales has recently (June 2003) established a Reporting Advisory Group with a view to enhance support and guidance. This research focused specifically on fund-raising charities, largely ignoring grantmaking charities (except in so far as a small number of the fund-raising charities are also significant grant-making charities), and therefore, generalisations beyond this group should be made with caution. Comparative studies between grant-making and fund-raising charities would aid understanding, as would comparative studies across the wider not-forprofit sector. Indeed, such studies might increase the momentum for the adoption of best practice. Moreover, qualitative research, possibly utilising interviews with major stakeholders, might provide useful insights to help explain the reasons for the current level of disclosure. The research reported in this paper concentrates on the voluntary disclosure in annual reports of basic background information and more specific performance information by Irish and British charities. A model for reporting to contributors to charities developed by Hyndman (1990) is used as a point of reference. With respect to listed companies, Beattie et al. (2002), on the basis of an analysis of current narrative reporting practices of 11 foodprocessing companies, have argued that companies should be encouraged to make disclosures across a broad range of areas and to improve the quality of certain categories of disclosure. They use an integrated analytical framework that captures nine major topics, divided into 79 sub-topics, and three dimensions of disclosure (capturing time orientation, financial orientation and quantitative orientation) to identify the quality of reporting. In addition, they distinguish between standard disclosures (information that has some information value) and premium disclosures (those that have high information value; the ‘real nuggets’) as a means for identifying the focus for reporting. While the research reported in this paper largely focuses on information that has been empirically identified as having real value to users of charities (see Hyndman, 1990), as Beattie et al. might refer to as the ‘real nuggets’, a method similar to that used in the Beattie et al. study, but adjusted to reflect the NFPO dimension, might permit a more detailed analysis of the entire package of narrative reporting by charities (both as an entire sector, and within major sub-sectors of charitable activity). Such might be a challenging and fruitful way in which to extend the work reported here. While beyond the scope of this paper as a focus for investigation, it could be argued that those charities with ‘better’ governance regimes are more likely to develop more extensive internal systems to target, measure and report performance, particularly performance related to effectiveness and efficiency. It could be the case that charities with better governance regimes may have more complete foundations on which to base the external reporting of performance and are therefore more likely to provide such information in their annual reports. The identification of key governance variables, possibly by reference to recent research on governance and charities (Hind, 1995; Gambling and Jones, 1996; Cornforth and Edwards, 1998) and the relationship of these to performance reporting might provide another rich area for further research.

150

C. Connolly, N. Hyndman / The British Accounting Review 36 (2004) 127–154

Acknowledgements The authors are very grateful for the financial support of the Institute of Chartered Accountants of Scotland in this research, and the very useful comments of the anonymous referees on an earlier draft of the paper.

Appendix A. Examples of goals and objectives, inputs, outputs and results (charity annual reports)

Example Goals and objectives To offer practical and emotional support to men, women and children affected by HIV/AIDS related illnesses To preserve places of historic interest or natural beauty permanently for the benefit of the nation Inputs Analysis of total expenditure for the year split between direct awards made by the scientific committee and other expenditure Analysis of expenditure on the lifeboat service during the year, together with a description of the different types of lifeboat used by RNLI Outputs Emergency response (North Korea): more than 26,000 people received regular Red Cross parcels, containing basic foodstuffs, clothing and first aid kits Student Support Network: helped 1100 people to get a job Benefit Rights And Information Team: gave individual help to nearly 8000 people, mainly pensioners Results Numbers of lives saved, persons landed and persons brought in during the year (and comparisons with nine previous years) The prevalence of blindness has decreased by over 50% in areas where the programme has been running

Charity

AIDS Care Education and Training (Ireland) The National Trust

Cancer Research Campaign

Royal National Lifeboat Institution

British Red Cross

Royal National Institute for the Blind

Royal National Lifeboat Institution Sight Savers International

C. Connolly, N. Hyndman / The British Accounting Review 36 (2004) 127–154

151

Appendix B. Examples of efficiency, effectiveness, future target information and budget information (charity annual reports)

Example Efficiency Table indicating the cost per trainee over a five-year period Split of expenditure between administration and fund-raising, and direct support of donkeys Effectiveness Target: provide training and work experience for an average of 1350 unemployed persons Actual: an average of 1356 trainees were provided with training and work experience Target: to conduct six audits of the quality of residential services Actual: achieved Future target information To identify at least 25 coaches and to organise a development programme over the next three years To help half the total number of those seriously ill at home with cancer (rather than the current one third) Budget information Income, expenditure and cash flow budgets for four years Budget information for the coming year: provide grants of more than £4 million for 150 research projects

Charity

Enterprise Ulster Donkey Sanctuary

Enterprise Ulster

SENSE

Special Olympics Ireland Marie Curie Cancer Care

Special Olympics Ireland Association for International Cancer Research

References Accounting Standards Board (ASB), 1999. Statement of Principles for Financial Reporting. ASB, London. Accounting Standards Committee (ASC), 1975. The Corporate Report. ASC, London. ASC, 1988. Statement of Recommended Practice (SORP) 2, Accounting by Charities, ASC, London. Ackerman, S., 1982. Unfair competition and corporate income taxation. Stanford Law Review 34, 1017–1039. American Accounting Association (AAA), 1966. A Statement of Basic Accounting Theory. AAA, Illinois. AAA, 1989. Measuring the Performance of Non-profit Organisations: The State of the Art. AAA, Sarasota.

152

C. Connolly, N. Hyndman / The British Accounting Review 36 (2004) 127–154

American Institute of Certified Public Accountants (AICPA), 1973. Objectives of Financial Statements: Report of the Study Group on the Objectives of Financial Statements (Trueblood Report). AICPA, New York. Anthony, R.N., Young, D.W., 2003. Management Control in Non-Profit Organisations, seventh ed., McGrawHill/Irwin, New York. Ashford, J.K., 1989. Charity accounts. In: Skerratt, I.C.L., Tonkin, D.J. (Eds.), Financial Reporting 1989/90: A Survey of UK Reporting Practice, Institute of Chartered Accountants in England and Wales (ICAEW), London. Austin, M., Posnett, J., 1979. The charity sector in England and Wales—characteristics and public accountability. National Westminster Quarterly Review, 40–51. Beattie, V., Jones, M., 1994. An empirical study of graphical format choices in charity annual reports. Financial Accountability and Management 10, 215–236. Beattie, V., McInnes, W., Fearnley, S., 2002. Through the Eyes of Management: A Study of Narrative Disclosures (An Interim report). ICAEW, London. Bird, P., Morgan-Jones, P., 1981. Financial Reporting by Charities. ICAEW, London. Boyne, G., Law, J., 1991. Accountability and local authority annual reports: the case of Welsh district councils. Financial Accountability and Management 7, 179 –194. Brace, P., Elkin, R., Robinson, D.D., Steinberg, H.I., 1980. Reporting of Service Efforts and Accomplishments (Research Report). Financial Accounting Standards Board (FASB), Stamford. Callen, J.L., Falk, H., 1993. Agency and efficiency in nonprofit organisations: the case of special health focus charities. Accounting Review 68, 48 –65. Cabinet Office, 2002. Private action, public benefit: a review of charities and the wider not-for-profit sector, Strategy Unit Report, Cabinet Office, London. Canadian Institute of Chartered Accountants (CICA), 1980. Corporate Reporting: Its Future Evolution (A Research Study). CICA, Toronto. Charity Accounting Review Committee (CARC), 1995. SORP 2 (revised), Accounting by Charities, CARC, London. Charities Aid Foundation (CAF), 2001. Tell the Main Story: Guidance on Effective Annual Reporting, (accessed 10 October 2001). Charity Commission, 2000. Accounting and Reporting by Charities: Statement of Recommended Practice. Charity Commission, London. Charity Commission, 2001. Facts and Figures, (accessed 8 October 2001). Charity Commission, 2002. Reporting the Activities and Achievements of Charities in Trustees’ Annual Reports. Charity Commission, London. Connolly, C., Hyndman, N., 2000. Charity accounting: an analysis of the impact of recent changes. British Accounting Review 32, 77–100. Connolly, C., Hyndman, N., 2001. A comparative study on the impact of revised SORP 2 on British and Irish charities. Financial Accountability and Management 17, 73–97. Connolly, C., Hyndman, N., 2003. Performance Reporting by UK Charities: Approaches, Difficulties and Current Practice. The Institute of Chartered Accountants of Scotland, Edinburgh. Cornforth, C., Edwards, C., 1998. Good Governance: Developing Effective Board-management Relations in Public and Voluntary Organisations. CIMA Publlishing, London. Drucker, P., 1990. Judge nonprofits by their performance, not only by their good intentions. Chronicle of Philanthropy 2, 32. Eden, R., Hyndman, N., 1999. Performance measurement in agencies: poisoned chalice or holy grail? Optimum 29, 9– 15. Edwards, M., Hulme, D., 1995. Non-governmental Organisations—Performance and Accountability (Beyond the Magic Bullet). Earthscan Publications, London. FASB, 1980. Statement of Financial Accounting Concepts No. 4. Objectives of Financial Reporting by Nonbusiness Organisations, FASB, Connecticut. FASB, 2001. Business Reporting: Insights into Enhancing Voluntary Disclosures. FASB, Connecticut. Fox, J., 1992. Democratic rural development: leadership accountability in regional peasant organisations. Development and Change 23, 1–36. Gambling, T., Jones, R., 1996. The Financial Governance of Charities. CAF/CIMA/ICAEW, London.

C. Connolly, N. Hyndman / The British Accounting Review 36 (2004) 127–154

153

Gambling, T., Jones, R., Kunz, C., Pendlebury, M., 1990. Accounting by charities: the application of SORP 2, Certified Research Report No. 21, Chartered Association of Certified Accountants, London. Gibbins, M., Richardson, A., Waterhouse, J., 1990. The management of corporate financial disclosure: opportunism, ritualism, policies, and processes. Journal of Accounting Research 28, 121– 143. Gibson, J.L., 1978. Performance indicators in the court system. In: Greer, P., (Ed.), Accountability in Urban Society, Sage, London. Government Accounting Standards Board (GASB), 1987. Concepts Statement No 1 Objectives of Financial Reporting. GASB, Stamford. Gray, R., 1983. Accounting, financial reporting and not-for-profit organisations. AUTA Review 15, 3–23. Gray, R., 1984. Uncharitable view of accounting. Accountancy, 84. Gray, S.J., Radebaugh, L.H., Roberts, C.B., 1990. International perceptions of cost constraints on voluntary information disclosures: a comparative study of UK and US multinationals. Journal of International Business Studies, 597 –622.fourth quarter. Hansmann, H., 1996. The changing roles of public, private and nonprofit enterprise in education, healthcare and other human services. In: Fuchs, V., (Ed.), Individual and Social Responsibility: Child Care, Education, Medical Care and Long-term Care in America, Chicago University Press, Chicago. HarperCollins, 1993. HarperCollins Concise English Dictionary, third ed., HarperCollins Publishers, Glasgow. Healy, P.M., Palepu, K.G., 1995. The challenges of investor communication: the case of CUC International Inc. Journal of Financial Economics 38, 111 –140. Hedlund, R.D., Hamm, K.E., 1978. Accountability and political institutions. In: Greer, P., (Ed.), Accountability in Urban Society, Sage, London. Hills Bush, B., 1992. Regulators seek greater ‘control’ of fundraising practices. Nonprofit Times January, 5. Hind, A., 1995. The Governance and Management of Charities. Voluntary Sector Press, London. Hines, A., Jones, J., 1992. The impact of SORP 2 on the UK charitable sector: an empirical study. Financial Accountability and Management 8, 49–67. HM Treasury, 1992. Executive Agencies: A Guide to Setting Targets and Measuring Performance. Stationery Office, London. HM Treasury, 2001. Choosing the Right Fabric: A Framework for Performance Information. HM Treasury, London. Home Office, 1999. Charities: A Framework for the Future. Stationery Office, London, Cm. 2626. Hyndman, N., 1990. Charity accounting: an empirical study of the information needs of contributors to UK fundraising charities. Financial Accountability and Management 6, 295–307. Hyndman, N., Anderson, R., 1995. The use of performance information in external reporting: an empirical study of UK executive agencies. Financial Accountability and Management 11, 1 –17. Hyndman, N., Anderson, R., 1998. Performance information, accountability and executive agencies. Public Money and Management 18, 23 –31. Hyndman, N., Eden, R., 2001a. Rational management, performance targets and executive agencies: views from agency chief executives in Northern Ireland. Public Administration 79, 579–598. Hyndman, N., Eden, R., 2001b. The Co-ordination of Mission, Objectives and Targets in UK Executive Agencies. CIMA Publishing, London. Hyndman, N., McKillop, D., 1999. Conversion Ratios in Charities in England and Wales: An Investigation of Economies of Scale. Financial Accountability and Management 15, 135–153. Jackson, P.M., 1982. The Political Economy of Bureaucracy. Philip Allan, London. Jones, R., Pendlebury, M., 1996. Public Sector Accounting, fourth ed., Pitman, London. Kahn, R., 1992. How much is too much in nonprofit compensation? Nonprofit Times, 18. Kanter, R.M., Summers, D., 1987. Doing well while doing good: dilemmas of performance management in nonprofit organisations and the need for a multiple-constituency approach. In: Powell, W., (Ed.), Handbook of Nonprofit Organisations, Yale University Press, New Haven. Kearns, K.P., 1994. The strategic management of accountability in non-profit organisations: an analytical framework. Public Administration Review 54, 185– 203. Likierman, A., 1993. Performance indicators: 20 early lessons from managerial use. Public Money and Management 13, 15–22.

154

C. Connolly, N. Hyndman / The British Accounting Review 36 (2004) 127–154

Likierman, A., 1997. Applying accruals-based accounting and budgeting to UK central government. The Irish Accounting Review 4, 55 –68. Macve, R., 1981. A Conceptual Framework for Financial Accounting and Reporting: The Possibilities for an Agreed Structure. ASC, London. Mayston, D., 1985. Non-profit performance indicators in the public sector. Financial Accountability and Management 1, 51–74. Meek, G., Roberts, C., Gray, S., 1995. Factors influencing voluntary annual disclosures by US, UK and continental European multinational corporations. Journal of International Business Studies 26, 555–571. Miah, N.Z., 1991. Attempts at developing a conceptual framework for public sector accounting in New Zealand. Financial Accountability and Management 7, 83– 97. National Audit Office, 1987. Monitoring and Control of Charities in England and Wales. Stationery Office, London. National Council for Voluntary Organisations (NCVO), 1976. Charity Law and Voluntary Organisations (Report of the Goodman Committee). NCVO, London. NCVO, 1990. Effectiveness and the Voluntary Sector (Report of the Working Party—The Nathan Report). NCVO, London. Northern Ireland Council for Voluntary Action (NICVA), 1998. The State of the Sector 11—Northern Ireland Voluntary Sector Almanac 1998. NICVA, Belfast. Office of Information Commissioner, 1999. Case 98042—Mr John Burns, The Sunday Times Newspaper and the Office of the Revenue Commissioners, (accessed 11 April 2000). Patton, J.M., 1992. Accountability and governmental financial reporting. Financial Accountability and Management 8, 165–180. Pendlebury, M., Groves, R., 1994. Company Accounts: Analysis, Interpretation and Understanding, third ed., Routledge, London. Pendlebury, M., Jones, R., Karbhari, Y., 1994. Developments in the accountability and financial reporting practices of executive agencies. Financial Accountability and Management 10, 33–46. Public Accounts Committee (PAC), 1988. Monitoring and Control of Charities in England and Wales. Stationery Office, London. PAC, 1997. Resource Accounting and Proposals for a Resource-based System of Supply, Committee of Public Accounts, Ninth Report, Session 1996–97. Stationery Office, London, HC 167. Rutherford, B.A., 1983. Financial Reporting in the Public Sector. Butterworths, London. Simon, J., 1987. The tax treatment of nonprofit organizations: a review of federal and state policies. In: Powell, W., (Ed.), Handbook of Nonprofit Organisations, Yale University Press, New Haven. Stewart, J., 1984. The role of information in public accountability. In: Hopwood, A.G., Tomkins, C.R. (Eds.), Issues in Public Sector Accounting, Philip Allan, Oxford. Taylor, N.V., 1989. Local authority accounting—the development of a conceptual framework. Financial Accountability and Management 5, 19–38. Williams, D.J., 1984. Regulating the accountability of charitable institutions. Accounting and Finance 6, 1–16. Williams, S., Palmer, P., 1998. The state of charity accounting—developments, improvements and continuing problems. Financial Accountability and Management 14 (4), 265– 279. Wise, D., 1995. Performance Measurement for Charities. Institute of Chartered Secretaries and Administrators, Hemel Hempstead. Woodfield Report, 1987. Efficiency Scrutiny of the Supervision of Charities. Stationery Office, London.