Journal of International Financial Management and Accounting 9:3 1998
Why Firms Comply Voluntarily with IAS: an Empirical Analysis with Swiss Data Pascal Dumontier* Pierre Mendès-France University, Grenoble, France
Bernard Raffournier University of Geneva, Switzerland
Since its foundation in 1973, the International Accounting Standards Committee (IASC) has become established as the most authoritative agency for accounting harmonisation. Its influence has spread widely to all parts of the world. Some countries, in particular developing, newly industrialised or newly capitalistic ones, have adopted IAS with few or no amendments as their national standards.1 In countries with older accounting traditions, local standards have not been replaced or even modified. However, IAS generally serve as references for the development of new standards. Some countries have gone further. In France and Switzerland, for example, companies can use IAS (or any other set of accounting standards compatible with national regulation) for the preparation of consolidated financial statements. This opportunity has been widely used in both countries. According to a study by Price Waterhouse,2 about 20% of French listed companies included in the SBF 120 market index3 declare their consolidated accounts in conformity with IAS. As shown later, this percentage amounts to almost 40% in Switzerland. This study is aimed at determining why companies voluntarily comply with IAS. Knowledge of the characteristics of companies which voluntarily adopt a particular set of accounting standards may be of particular interest for standard setting. It may give an indication of the type of companies which will naturally be in favour of accounting regulation and, adversely, of firms which standard setters will have to convince. No prior research has, to our knowledge, been conducted on this issue, probably because in most industrial states, the national regulation does not * We thank Peter Walton, Rahim Bah and two anonymous reviewers for their helpful comments. Pascal Dumontier, Pierre Mendès-France University, Grenoble, France, E-mail:
[email protected]; Bernard Raffournier, University of Geneva, HEC-Uni Mail, 102 bd Carl Vogt, 1211 Geneve 4, Switzerland, E-mail:
[email protected]. © Blackwell Publishers Ltd. 1998, 108 Cowley Road, Oxford OX4 1JF, UK and 350 Main Street, Malden, MA 02148, USA.
Why Swiss Firms Comply Voluntarily with IAS 217 allow firms to choose the set of accounting standards they want to follow. However, in some countries whose national accounting regulation is low, as in Switzerland, companies have a large choice of alternative accounting policies which enables them to comply with international accounting standards or foreign GAAP. Similarly, some countries whose national accounting regulation is not market-oriented (e.g., France) accept deviations from national rules in the preparation of consolidated financial statements in order to avoid their firms being penalised on international financial markets. Although the use of IAS for consolidated financial statements is not restricted to Switzerland, this country was chosen for two reasons. First, the high percentage of Swiss companies which voluntarily adopt international accounting standards makes it possible to build an homogeneous sample whose size is sufficient to derive statistical inferences without incurring biases inherent to the inclusion of data from several countries with different degrees of accounting regulation. Secondly, and as discussed further, Swiss GAAP are much less stringent than international standards for valuation rules as well as disclosure requirements. Consequently, compliance with IAS implies additional disclosure and renunciation of a considerable discretion in accounting choices which make the decision to adopt IAS probably more costly for Swiss firms than for companies from countries with a higher degree of accounting regulation. This paper is organised as follows. The next section provides an overview of Swiss accounting regulation. The second section discusses the possible determinants of the decision to comply with IAS. Section 3 describes the sample and the explanatory variables. Empirical results are reported and discussed in the last section.
I. Accounting in Switzerland4 Accounting in Switzerland is characterised by (1) low disclosure requirements, (2) few and permissive accounting standards and (3) a high degree of tolerance for income smoothing. Legal disclosure requirements are contained in the company law (Code of Obligations, hereafter CO), which stipulates on the one hand some general accounting principles which apply to all enterprises (art. 957 to 964) and, on the other hand, more detailed rules which are applicable only to companies limited by shares (art. 662 to 673). The law defines a minimum structure for the balance sheet and the income statement (CO, art. 633 and © Blackwell Publishers Ltd. 1998.
218
Pascal Dumontier and Bernard Raffournier
633a) and provides a list of 12 information items which must be disclosed in the notes to the accounts (CO, art. 633b). The other source of accounting regulation is the Foundation for Accounting and Reporting Recommendations (FER), the Swiss standard setting body created in 1984 on the initiative of the Swiss Institute of Certified Accountants and Tax Consultants. Its objective is to harmonise accounting practices, improve comparisons, and increase the quality of financial statements. To date, the FER has issued 12 Accounting and Reporting Recommendations (ARR) and 5 others are in process. Appendix 1 shows that ARR deal with many less issues than IAS do. Important subjects, such as inventory valuation, depreciation, accounting for leases, investments, and retirement benefit costs are not yet covered by the FER recommendations. As shown in appendix 1, FER disclosure requirements are very low, compared to the wide range of information required by IAS. In addition, and as already mentioned, the application of the FER recommendations is still optional. All firms, even listed companies,5 only have to provide information required by law. In this context, it is clear that compliance with IAS implies considerable additional disclosure for Swiss companies. The FER recommendations are also much less detailed and more permissive than the corresponding International Accounting Standards. Their maximum length, including explanations, is 5 pages, while several IAS are more than 20 pages long. The purpose of ARR is less to prescribe detailed methods than to provide guidelines sufficient to prevent erroneous treatments. Appendix 2 compares the IAS and FER provisions for some major issues. This comparison makes it obvious that FER recommendations are much more permissive than IAS. This characteristic of the FER recommendations is probably a consequence of the absence of any standard setting body prior to 1984. Because of this lack of regulation, Swiss firms have for a long time benefited from a large range of accounting alternatives. They would probably oppose the development of stringent accounting standards which would restrict their choice of accounting principles. Since compliance with their recommendations cannot be enforced by law, Swiss standard setters need to gain support from companies. Accordingly, they develop recommendations with many options and low disclosure requirements. In Switzerland, as in several other countries influenced by the German conservative view of accounting,6 income smoothing is formally permitted by law. Hidden reserves can be created (and cancelled) practically without limitation, since CO article 669 states that “additional hidden © Blackwell Publishers Ltd. 1998.
Why Swiss Firms Comply Voluntarily with IAS 219 reserves are permissible to the extent that they are justified to secure a continuing growth of the business or to distribute consistent dividends, provided that such justifications are in the interest of the shareholders.” The only requirement is that a write-down of hidden reserves should be disclosed in the notes if material, and that every addition to or release from hidden reserves should be reported to the auditors. As a result, most Swiss companies make large use of hidden reserves by recognising excessive depreciation of assets or creating unjustified provisions. From 1989 to 1992, Bilan, a Swiss economic journal, published a ranking of the most profitable Swiss companies, based on an estimation of their “true” (undistorted) income. In some cases, reported income represented less than 25% of estimated real earnings.7 Hidden reserves are, of course, created when profits are high and released (i.e., recognised as income) in periods of losses or low earnings. Consistency of valuation methods theoretically precludes such income smoothing behaviour. But this convention, although expressly recognised by law, is not strictly respected in Switzerland. The Code of Obligations allows deviations from this principle and the FER states that these deviations “are admissible as long as they meet the legal maximum valuation and disclosure regulations” (ARR 3). From these observations, it can be presumed that Swiss firms which choose to comply with IAS renounce a considerable discretion in their accounting choices. They also give up the creation of hidden reserves, with the result that earnings management is more difficult. In addition, they commit themselves to disclose much more information than legally required. They probably do so because they expect advantages from these limitations. The next section will discuss some possible determinants of this behaviour. II. Reasons for Voluntary Compliance with IAS In the absence of prior research on voluntary compliance with IAS or with any other set of accounting standards, hypotheses are drawn from studies which investigated other behaviour whose cost and advantages are expected to be similar. Since the adoption of IAS implies an increase in the extent of information disclosed, studies on the determinants of voluntary disclosure provide a first theoretical background. In addition, since IAS are more stringent than Swiss GAAP, compliance with international accounting standards may also be considered as a bonding activity, as purchase of a © Blackwell Publishers Ltd. 1998.
220
Pascal Dumontier and Bernard Raffournier
discretionary audit. Relevant hypotheses can thus be derived from studies on the decision to submit the firm to a voluntary audit. Listing status Financial statements of internationally diversified companies probably have a larger proportion of foreign users than those of purely domestic enterprises. Most of these users are not familiar with Swiss accounting rules and do not have easy access to other sources of information. Compliance with IAS should give them an assurance that financial statements are not misleading. Enhanced disclosure should also provide these users with information they could not obtain otherwise. Accordingly, firms complying with IAS should be considered as providing superior information which may help them obtain better financing conditions.8 Compliance with IAS may also facilitate admission on foreign stock exchanges (other than U.S.) and exempt firms from preparing financial statements consistent with local accounting rules. The resulting hypothesis is: H1: Firms listed on foreign stock exchanges are more likely to comply with IAS than companies listed on the Swiss equity market only. The evidence on voluntary disclosure gives support to this hypothesis. Cooke (1989, 1992) reports that Swedish and Japanese companies whose shares are listed on foreign markets exhibit a higher level of disclosure than those quoted on their national market only. Similar results were obtained in Malaysia (Hossain, Tan and Adams, 1994; and in New Zealand, Hossain, Perera and Rahman, 1995). Gray, Meek and Roberts (1995) also found that U.S. and U.K. multinationals disclose more information in their annual reports than domestically listed companies. Internationality Listing on foreign stock exchanges is not a necessary condition for an inducement to apply IAS. Foreign financing can also be obtained from banks and other financial institutions. Since foreign operations are often financed with local external capital (Cooke, 1993), internationally diversified companies may be incited to comply with IAS, even if they are not listed on foreign stock exchanges. Benefits other than financial ones may also be expected from compliance with IAS. Because they are more visible on foreign markets, firms which operate internationally may have an interest in preparing financial statements which can easily be read by © Blackwell Publishers Ltd. 1998.
Why Swiss Firms Comply Voluntarily with IAS 221 local customers, suppliers and governments. This leads to the following hypothesis: H2: The propensity to comply with IAS increases with the level of international diversification. This hypothesis is supported by Raffournier (1995b) who found that the extent of disclosure in annual reports of Swiss firms is positively related to the level of international activity, as measured by the exports-to-sales ratio. Size The influence of size on the decision to comply with IAS may be based on several arguments. First, since IAS render earnings management more difficult because of the prohibition of hidden reserves,9 compliance with these standards may be used by firms sensitive to political costs in order to give their financial statements superior confidence. If, as generally admitted, political costs are positively related to size (Watts and Zimmerman, 1986), large firms should be more likely to apply IAS than small companies. Secondly, since compliance with IAS implies an increase of disclosure, there should be a relationship between the decision to apply IAS and the cost of additional disclosure. It is generally argued that disclosing detailed information is less costly for large firms because these are assumed to produce this information already for internal purposes (Singhvi and Desai, 1971). There may also be a fixed component to disclosure costs, so that the cost per unit of size is decreasing (Lang and Lundholm, 1993). In addition, because their annual report is the main source of information for their competitors, smaller firms may be reluctant to make additional disclosures which might place them at a competitive disadvantage (Firth, 1979). All these reasons support the following hypothesis: H3: The propensity to comply with IAS increases with size. A similar hypothesis was tested by Blacconiere et al. (1991) in the savings and loans industry. Their results show that, as predicted, firms voluntarily adopting regulatory accounting principles are larger than those which do not. The influence of size on voluntary disclosure is well documented. All empirical studies, although conducted in different environments, show a positive relationship between the size of the firm and its extent of disclosure in annual reports (Singhvi and Desai, 1971; Buzby, © Blackwell Publishers Ltd. 1998.
222
Pascal Dumontier and Bernard Raffournier
1975; Firth, 1979; Chow and Wong-Boren, 1987; Cooke, 1989, 1992; Hossein, Tan and Adams, 1994; Hossein, Perera and Rahman, 1995; Meek, Roberts and Gray, 1995; Raffournier, 1995b; Wallace and Naser, 1995). Several other studies have investigated the influence of size on the decision to purchase voluntary auditing. Ettredge et al. (1994) report that U.S. companies which contract for quarterly reviews by external auditors exhibit larger size than firms having their accounts audited only annually. Similarly, Prémont and Dionne (1993) note that, in Quebec, firms which have their financial statements audited prior to an initial public offering are larger than companies which submit unaudited accounts. On the other hand, however, Chow (1982) did not find a significant relationship between size and the decision to hire external auditors at a time (1926) when there were no externally imposed audit requirements in the U.S.A. Ownership structure Separation of ownership and control generates agency costs resulting from conflicts of interests between managers and shareholders. Jensen and Meckling (1976) have shown that, to the extent that managers do not hold the entire capital of their firm, they are motivated to divert a part of the firm’s value into perquisites. Accounting contributes to limit these wealth transfers by making them more apparent. A necessary condition for this monitoring role of accounting to be efficient is that managers are not able to distort accounting numbers through the use of inadequate methods or frequent accounting policy changes. The larger the discretion in the choice of accounting methods, the less efficient the monitoring role of accounting. Since IAS are more numerous and more stringent than Swiss accounting principles, compliance with international accounting standards considerably restricts the opportunity set of accounting choices. In addition, by prohibiting the creation of hidden reserves, IAS give financial statements greater credibility. Compliance with IAS may thus be either a monitoring activity imposed by shareholders or a bonding activity decided by managers in order to reduce agency costs. Several empirical studies have shown that financial statements are the main source of information for small shareholders (Epstein and Pava, 1993; Epstein and Anderson, 1994). Because of their small part in the firm’s wealth, these investors cannot incur large expenditure in the search for additional information and in controlling the wealth transfers of managers. They should thus favour the use of stringent accounting standards © Blackwell Publishers Ltd. 1998.
Why Swiss Firms Comply Voluntarily with IAS 223 which restrict the managers’ discretion in accounting policy choices and require extensive disclosure. Accordingly, companies with numerous small shareholders should be more motivated to comply with these standards than those controlled by large shareholders. Thus we hypothesise that: H4: The propensity to comply with IAS increases with ownership diffusion. Previous studies on the extent of disclosure do not provide support to this hypothesis. Neither Raffournier (1995b) in Switzerland, nor Wallace and Naser (1995) in Hong Kong, found a significant relationship between ownership diffusion and the content of annual reports. The results of studies on voluntary auditing are more encouraging. Ettredge et al. (1994) report that the percentage of stock owned by officers and directors is smaller in firms purchasing quarterly reviews. Prémont and Dionne (1993) found a similar influence on the decision to make financial statements audited prior to an initial public offering (IPO). Leverage External financing creates the opportunity for stockholders to transfer wealth to the prejudice of creditors. Since borrowing costs are based on a firm’s current risk, shareholders have an incentive to invest borrowed funds in projects riskier than current assets in order to increase their expected return without supporting any additional cost. In this case, creditors are penalised since they are not compensated for the resulting increase of default risk (Jensen and Meckling, 1976). To the extent that earnings volatility can proxy for a firm’s risk, financial statements can be used to monitor agency relationships between shareholders and creditors. Since income smoothing artificially reduces earnings volatility, compliance with IAS helps facilitate the monitoring role of financial statements by making this smoothing behaviour less easy. The more levered the firm, the more necessary to ensure an efficient monitoring of agency relationships between shareholders and creditors. We thus hypothesise that: H5: The propensity to comply with IAS increases with leverage. Empirical studies do not generally support the influence of leverage on the extent of disclosure in annual reports (Chow and Wong-Boren, 1987; Hossain, Tan and Adams, 1994; Raffournier, 1995b; Wallace and Naser, 1995). The only exception is Hossain, Perera and Rahman (1995) who © Blackwell Publishers Ltd. 1998.
224
Pascal Dumontier and Bernard Raffournier
report a positive and weakly significant relationship in New Zealand. In contrast, the influence of leverage on voluntary auditing is well documented (Chow, 1982; Prémont and Dionne, 1993; Ettredge et al., 1994). Capital intensity From a theoretical analysis of conflicting interests between stockholders and creditors, Myers (1977) predicts that wealth transfers from creditors to shareholders can be more easily operated through the acquisition of new assets than by replacing those already in place because increasing the firm’s risk by means of new assets is easier than shifting the risk of existing ones. He also argues that there is a positive relationship between assets in place and capital intensity. In as much as the proportion of fixed assets is a proxy for capital intensity, the need for monitoring should be less for firms whose wealth is invested mainly in fixed assets than for those with a high proportion of current assets. Accordingly, compliance with IAS should be less necessary for firms in the first category. This suggests the following hypothesis: H6: The propensity to comply with IAS is inversely related to the proportion of fixed assets. Contrary to Anderson et al. (1993) who found that monitoring expenditures decrease with the proportion of assets in place, studies which investigated the influence of this variable on voluntary disclosure in annual reports do not report any significant relationship (Chow and WongBoren, 1987; Hossain, Tan and Adams, 1994; Hossain, Perera and Rahman, 1995; Raffournier, 1995b). Profitability Since earnings volatility is related to total risk, companies may use earnings management with the hope of reducing their risk, as perceived by the market. Since it makes earnings management more difficult, compliance with IAS may be a signal used by firms with superior performance. The following hypothesis may thus be tested: H7: The propensity to comply with IAS increases with profitability. The evidence concerning the influence of profitability on voluntary disclosure is unclear. Raffournier (1995b) and Meek, Roberts and Gray (1995) found no significant relationship, while Singhvi and Desai (1971) © Blackwell Publishers Ltd. 1998.
Why Swiss Firms Comply Voluntarily with IAS 225 found a positive relation, and Wallace and Naser (1995) found an inverse influence. With regard to voluntary auditing, Prémont and Dionne (1993) report that firms presenting audited financial statements at an initial public offering are more profitable than those which present unaudited accounts. Auditors’ reputation Several authors have suggested that auditors play a role in defining the disclosure policy of their clients. Specifically, it has been argued that large and well-known audit firms may incite companies to disclose more information (Singhvi and Desai, 1971; Firth, 1979). Similarly, it can be hypothesised that it is in their own interest to have their clients comply with complicated accounting standards. The rationale for that is twofold. First, large auditor firms have high reputation capital at stake. According to Watts and Zimmerman (1986), independence is a necessary condition for an auditor’s reputation. By forcing their clients to apply stringent accounting standards, large audit firms prove they are independent and thus strengthen their reputation. Secondly, large and international audit firms probably have a competitive advantage in controlling the application of international accounting standards because of the superior international training of their employees and because of the existence of economies of scale in the development of competence in international accounting standards. We posit that firms comply voluntarily with IAS to give their financial statements greater credibility. Credibility can be enhanced by making these statements audited by large and well-known audit firms. In other words, the choice of auditors may be a consequence, as well as a cause, of compliance with IAS. In any event, and since it is generally considered that membership in a Big 6 network proxies for auditor reputation, we hypothesise that: H8: Companies audited by a Big 6 firm are more likely to comply with IAS than firms audited by a smaller auditor. The empirical evidence on the extent of disclosure is conflicting. Firth (1979), Hossain, Tan and Adams (1994), and Hossain, Perera and Rahman (1995) conclude that the type of auditors is not an important factor in explaining voluntary disclosure. Adversely, Singhvi and Desai (1971) and Raffournier (1995b) found a significant relation between auditors’ reputation and the extent of disclosure in annual reports. © Blackwell Publishers Ltd. 1998.
226
Pascal Dumontier and Bernard Raffournier
III. Data Collection and Sample Design The sample At the end of 1994, 215 Swiss companies had their shares listed on the Swiss Stock Exchange.10 This number includes 36 banks and 13 insurance companies excluded from the study because of their specific disclosure requirements which prevent them from freely selecting the accounting standards they apply. 31 companies out of the 166 remaining firms were also excluded because of annual reports availability. Since the study is aimed at determining the characteristics of firms which comply voluntarily with IAS, we also excluded 2 companies which declared their financial statements in accordance with U.S. GAAP. Consequently, the final sample includes 133 firms and covers 80% of the target population. All data refer to consolidated financial statements of the year 1994. Accounting standards used by Swiss firms are very diverse: 59 companies declare their group accounts in conformity with the FER recommendations, 38 with the EU directives, and 52 with the IAS, while 15 do not mention any reference. These groups are not independent. Because of options in valuation methods and different levels of disclosure requirements, financial statements can comply with several sets of standards simultaneously. In particular, IAS are compatible with both EU directives and FER recommendations. That means that financial statements in compliance with IAS are also de facto in conformity with EU directives
Table 1. Accounting Standards Used by Firms Non-IAS group (82 firms) EU-non-IAS group (30 firms)
IAS group (51 firms)
Neither-IASTotal nor-EU group sample (52 firms) (133 firms)
EU FER No IAS only IAS + EU IAS + FER only EU + FER only standard 34
8
9
17
13
In tables 4 and 5: – the IAS group is denoted by subscript 1 – the non-IAS group is denoted by subscript 0 – the EU-non-IAS group is denoted by subscript 01 – the neither-IAS-nor-EU group is denoted by subscript 00 © Blackwell Publishers Ltd. 1998.
37
15
133
Why Swiss Firms Comply Voluntarily with IAS 227 and FER recommendations, even if not mentioned in the notes to the accounts. Compliance with a particular set of accounting standards was determined on the basis of the firm’s assertion which, in most cases, was corroborated by the auditors in their report on consolidated financial statements. The IAS group is composed of 51 firms (i.e., 38%) which declare their financial statements in conformity with IAS. The remaining 82 firms constitute the non-IAS group. Companies which referred to IAS but with some disclosure exceptions were nevertheless classified in the IAS group because it was apparent that most Swiss firms which declare compliance with IAS do not, in fact, satisfy the entire set of disclosure requirements of the IASC. Although this study focuses on the decision to comply with IAS, we also classified firms into three groups depending on whether they comply with IAS, with EU directives only, or with none of these standards. The purpose of this additional classification is to analyse differences between firms complying only with EU directives and those satisfied with applying Swiss accounting principles. To the extent that advantages expected from voluntary compliance with stringent accounting standards require that this behaviour should be disclosed, firms which do not refer to any set of standards are supposed to apply neither IAS nor EU directives. Accordingly, they were grouped with companies which only referred to FER recommendations.11 It has been previously shown that FER recommendations are much less stringent than IAS. Similarly, there is no doubt that EU directives are more permissive than IAS. EU directives are only minimum requirements whose aim is to harmonise regulation inside the European Union. As such, they admit several options allowing member states to promulgate a more stringent regulation, providing it is consistent with these guidelines.12 Since Switzerland is not a member of the European Union, Swiss firms which want to mention compliance with EU directives are not obliged to refer to the accounting regulation of a particular country. They can make their choice among numerous options proposed to member states and enterprises and comply with minimum disclosure requirements which, as demonstrated by Raffournier (1995b), are much lower than the corresponding information required by IAS. Consequently, for Swiss firms, compliance with EU directives does not imply a substantial loss of discretion in accounting choices, nor a significant increase of disclosure. Accordingly, Swiss firms which conform to EU directives should not be very different from those which apply Swiss accounting principles only. In order to © Blackwell Publishers Ltd. 1998.
228
Pascal Dumontier and Bernard Raffournier
verify this assertion, the non-IAS group (82 firms) was categorised into two sub-samples: a first group (denoted by subscript 01 in the following tables) composed of 30 firms complying with EU directives and a second group (subscript 00 in the tables) including the 52 remaining companies. Explanatory variables Independent variables are shown in Table 2. According to our hypotheses, these variables measure internationality, firm size, ownership diffusion, leverage, capital intensity, profitability, auditor type, and listing status. The choice of a particular set of accounting standards probably depends on the geographical area in which the firm operates. For Swiss companies whose activities are mainly concentrated in Europe, compliance with EU directives may be sufficient, but for firms operating worldwide, adoption of IAS is probably necessary. Consequently, the percentage of sales realised outside Switzerland, in consequence of exports or overseas manufacturing, may be inadequate to correctly assess the incitement to comply with IAS. Accordingly, internationality was measured by two variables: percentage of sales realised abroad (INT1) and percentage of sales outside Europe (INT2).13 Size can be measured in numerous different ways. Four variables frequently used to proxy for size are considered: sales (SALES), total assets (ASSETS), decimal logarithm of sales (LSALES) and decimal logarithm of total assets (LASSETS). Measures of ownership diffusion are constrained by available information. Swiss companies generally do not provide an extensive description of their ownership structure, but some of them reveal the identity and proportion of stock held by their main shareholders. When missing in the annual report, this information can generally be found in the Guide des actions suisses.14 We thus measured ownership diffusion (DIFF) by the percentage of stocks not held by known shareholders, as identified by these two sources. Selected leverage ratios are total-debt-on-total-assets (TDTA) and longterm-debt-on-total-assets (LTDTA). Capital intensity is proxied by the assets-in-place ratio (AIP) obtained by dividing the book value of fixed assets (net of depreciation) by total assets. Three measures of accounting profitability were defined: a return-onassets ratio (ROA) and two return-on-equity ratios (ROE1 and ROE2). To the extent that companies which do not comply with IAS are more likely to smooth income, these measures may be biased. Consequently, a market © Blackwell Publishers Ltd. 1998.
Why Swiss Firms Comply Voluntarily with IAS 229 Table 2. Explanatory Variables Internationality sales outside Switzerland
INT1 =
total sales sales outside Europe
INT2 =
total sales
Size SALES = 1994 total sales LSALES = decimal logarithm of 1994 total sales ASSETS = total assets at the end of 1994 LASSETS = decimal logarithm of 1994 total assets Ownership diffusion DIFF = 100 – percentage of stock held by known shareholders Leverage TDTA = LTDTA =
total debt total assets long term debt total assets
Capital intensity AIP
=
fixed assets total assets
Profitability ROE1 = ROE2 = ROA
=
MRET =
net income equity earnings before interests and taxes equity earnings before interests and taxes total assets 1 1994 APt – APt –1 + D t ∑ 3 t =1992 APt –1
where APt = Average stock price in year t Dt = Dividend paid in year t Auditors’ reputation BIG6 = 1 if the auditor is a Big 6 firm BIG6 = 0 otherwise Listing status LIST LIST
= 1 if the firm’s shares are listed on a foreign stock exchange = 0 otherwise © Blackwell Publishers Ltd. 1998.
230
Pascal Dumontier and Bernard Raffournier
measure of profitability was also used: the average stock return from 1992 to 1994 (MRET). Finally, the influence of auditor type is assessed by a dummy variable taking the value 1 when the company is audited by a “Big 6” firm and 0 otherwise. The listing status is taken into account by a dummy variable (LIST). This variable equals 1 if the firm is listed on a foreign stock exchange and 0 otherwise. Very few Swiss companies are listed abroad mainly because the economy of Switzerland is composed of a small number of multinational companies along with numerous small- or mediumsize firms. Only 13 out of 133 firms in the sample are listed on one or several foreign stock exchanges: 7 are listed only in Europe (mainly on the London Stock Exchange, SEAQ, or the Frankfurt Stock Exchange), and 6 are listed both in and outside Europe (on the NYSE, and NASDAQ, and the Tokyo Stock Exchange). Table 3 provides descriptive statistics on all these variables for the entire sample.
Table 3. Descriptive Statistics and Normality Test Continuous variables INT1 (%) INT2 (%) SALES (106 CHF) ASSETS (106 CHF) DIFF (%) TDTA (%) LTDTA (%) AIP (%) ROE1 (%) ROE2 (%) ROA (%) MRET (%)
Dichotomous variables BIG6 LIST
Number of observations
Mean
Standard deviation
K-S test for normality
92 125 133 133 131 133 133 133 133 133 133 127
55.03 21.76 2,163 2,529 46.82 59.68 30.95 49.33 8.99 17.16 5.21 11.12
33.84 21.18 6,472 6,838 26.42 17.44 17.57 21.00 16.67 19.55 4.58 21.09
Z = 1.08 Z = 1.69 *** Z = 0.67 Z = 0.65 Z = 1.23 * Z = 0.91 Z = 1.28 * Z = 0.61 Z = 2.32 *** Z = 1.42 ** Z = 0.65 Z = 0.70
Number of observations
Value = 1
Value = 0
133 133
118 13
15 120
*** Hypothesis of normality rejected at the 0.01 level. ** Hypothesis of normality rejected at the 0.05 level. * Hypothesis of normality rejected at the 0.10 level. © Blackwell Publishers Ltd. 1998.
Why Swiss Firms Comply Voluntarily with IAS 231 IV. Empirical Results Univariate analyses A Kolmogorov-Smirnov test was performed to check the normality assumption of independent variables (Table 3). It indicates that 3 variables (INT2, ROE1 and ROE2) deviate from normality at the 5% level. They were treated with the non-parametric Mann-Whitney U-test, while the traditional Student t-test was used for the other variables. Both tests were performed for DIFF and LTDTA, whose normality assumption was rejected at the 10% level only. Chi-square tests were used for dummy explanatory variables (LIST and BIG6). Four comparisons were made: (1) IAS versus non-IAS firms (t1/0, Z1/0, χ21/0), (2) IAS versus EU-non-IAS firms (t1/01, Z1/01, χ21/01), (3) IAS versus neither-IAS-nor-EU companies (t1/00, Z1/00, χ21/00), and, finally, EU-non-IAS versus neither-IAS-nor-EU firms (t01/00, Z01/00, χ201/00). Table 4 summarises the results obtained when comparing firms which comply to IAS to those which do not. As expected, IAS firms are larger and more internationally diversified than others. Tests show that SALES, ASSETS, INT1 and INT2 are statistically significant at the 1% level. These results support hypotheses H2 and H3. Moreover, in accordance with hypothesis H4, IAS firms exhibit a higher ownership diffusion (DIFF) than non-IAS ones. Chi-square tests also display a significant influence of auditor type (BIG6) and listing status (LIST). These results validate hypotheses H1 and H8. However, IAS firms are neither more profitable, nor more levered than the others. Differences on ROE1, ROE2, ROA or MRET, and on TDTA or LTDTA, are not statistically significant at the usual levels. Moreover, both groups also exhibit the same percentages of fixed assets since AIP values are not statistically different. Accordingly, hypotheses H5, H6 and H7 are not validated by empirical evidence. Other results are shown on Table 5. The comparison between firms which comply with IAS and those which refer neither to IAS nor EU directives (subscript 1/00) shows that the former are significantly larger and more internationally diversified. A similar result is obtained when comparing EU-non-IAS to neither-IAS-nor-EU firms (subscript 01/00). The opposition between IAS firms and companies referring to EU directives only (subscript 1/01) also provides interesting results. IAS firms are slightly bigger and realise a larger part of their activity outside Europe. The fact that INT2 is highly significant but not INT1 confirms the assumption that the primary motive for compliance with IAS is extraEuropean activity. As long as their business is concentrated in Europe, © Blackwell Publishers Ltd. 1998.
232
Pascal Dumontier and Bernard Raffournier
Table 4. IAS Firms Versus Non-IAS Firms—Results of Univariate Analyses Variables
IAS firms
Non-IAS firms
LIST
µ1 = 66.38 σ1 = 28.15 µ1 = 32.26 σ1 = 21.65 µ1 = 4,161 σ1 = 10,763 µ1 = 4,486 σ1 = 10,563 µ1 = 52.63 σ1 = 27.63 µ1 = 56.80 σ1 = 12.77 µ1 = 30.24 σ1 = 16.01 µ1 = 47.05 σ1 = 15.89 µ1 = 9.37 σ1 = 9.28 µ1 = 16.58 σ1 = 14.13 µ1 = 5.72 σ1 = 4.37 µ1 = 13.09 σ1 = 21.29 N1 = 11 (51)
BIG6
N1 = 35 (51) N0 = 69 (82)
INT1 INT2 SALES ASSETS DIFF TDTA LTDTA AIP ROE1 ROE2 ROA MRET
µ0 = 46.91 σ0 = 35.75 µ0 = 15.19 σ0 = 18.28 µ0 = 790 σ0 = 1,064 µ0 = 956 σ0 = 1,452 µ0 = 42.34 σ0 = 25.17 µ0 = 62.43 σ0 = 26.51 µ0 = 33.76 σ0 = 27.61 µ0 = 50.53 σ0 = 23.61 µ0 = 8.54 σ0 = 22.52 µ0 = 17.95 σ0 = 22.21 µ0 = 5.10 σ0 = 4.78 µ0 = 9.88 σ0 = 21.03 N0 = 2 (82)
Student Mann-Whitney t-test U-test Chi-square-test t1/0 = 3.68 (p = 0.00)
NA
NA
NA
Z1./0 = 4.68 (p = 0.00)
NA
NA
NA
NA
NA
Z1/0 = 2.22 (p = 0.03)
NA
NA
NA
Z1/.0 = –0.09 (p = 0.92)
NA
NA
NA
t1/0 = 3.21 (p = 0.00) t1/0 = 2.83 (p = 0.00) t1/0 = 2.36 (p = 0.02) t1/0 = –1.07 (p = 0.29) t1/0 = –0.74 (p = 0.46) t1/0 = –0.73 (p = 0.47) NA NA t1/0 = 0.33 (p = 0.74) t1/0 = 0.38 (p = 0.71)
Z1/.0 = 0.33 (p = 0.75) Z1/.0 = –0.35 (p = 0.72)
NA NA
NA
NA
NA
NA
NA
NA
NA
NA
χ21/0 = 23.47 (p = 0.00) χ21/0 = 23.05 (p = 0.00)
NA: Not appropriate.
Swiss firms seem to be satisfied with applying EU directives, but when they begin to operate on a worldwide basis, a need for compliance with IAS appears. Irrespective of the groups considered, differences in terms of listing status (LIST) and auditor type (BIG6) are highly significant, suggesting that the incentive to use first EU directives, then IAS, is stronger for firms listed on foreign stock exchanges and for companies audited by a Big 6 firm. © Blackwell Publishers Ltd. 1998.
Tabe 5. IAS Versus EU-Non-IAS and Neither-IAS-Nor-EU Firms—Results of Univariate Analyses Variables INT1 INT2 SALES ASSETS DIFF TDTA LTDTA AIP
Groups
Mean
Standard deviation
IAS EU-non-IAS Neither-IAS-nor-EU IAS EU-non-IAS Neither-IAS-nor-EU IAS EU-non-IAS Neither-IAS-nor-EU IAS EU-non-IAS Neither-IAS-nor-EU IAS EU-non-IAS Neither-IAS-nor-EU IAS EU-non-IAS Neither-IAS-nor-EU IAS EU-non-IAS Neither-IAS-nor-EU IAS EU-non-IAS Neither-IAS-nor-EU
µ1 = 66.38 µ01 = 58.91 µ00 = 37.33 µ1 = 32.36 µ01 = 20.28 µ00 = 11.95 µ1 = 4,161 µ01 = 1,212 µ00 = 559 µ1 = 4,486 µ01 = 1,695 µ00 = 552 µ1 = 52.63 µ01 = 42.76 µ00 = 42.18 µ1 = 56.80 µ01 = 66.82 µ00 = 60.40 µ1 = 30.24 µ01 = 34.34 µ00 = 33.60 µ1 = 47.05 µ01 = 50.31 µ00 = 50.74
σ1 = 28.15 σ01 = 33.43 σ00 = 34.39 σ1 = 21.65 σ01 = 17.81 σ00 = 18.19 σ1 = 10,763 σ01 = 1,300 σ00 = 836 σ1 = 10,563 σ01 = 2,088 σ00 = 684 σ1 = 27.63 σ01 = 25.55 σ00 = 24.61 σ1 = 12.77 σ01 = 19.10 σ00 = 29.47 σ1 = 16.01 σ01 = 22.60 σ00 = 29.91 σ1 = 15.89 σ01 = 22.70 σ00 = 24.57
Student t-test t1/01 = 0.87 t1/00 = 3.65 t01/00 = 2.32
(p = 0.39) (p = 0.00) (p = 0.02)
NA Z1/01 = 2.37 Z1/00 = 5.01 Z01/00 = 2.80
N/A t1/01 = 1.74 t1/00 = 2.43 t01/00 = 2.77 t1/01 = 1.65 t1/00 = 2.71 t01/00 = 3.66 t1/01 = 1.58 t1/00 = 1.95 t01/00 = 0.11 t1/01 = –2.61 t1/00 = –0.72 t01/00 = 1.19 t1/01 = –0.89 t1/00 = –0.65 t01/00 = 0.13 t1/01 = –0.71 t1/00 = –0.83 t01/00 = –0.08
Mann-Whitney U-test
(p = 0.09) (p = 0.01) (p = 0.01) (p = 0.10) (p = 0.00) (p = 0.00) (p = 0.13) (p = 0.06) (p = 0.92) (p = 0.01) (p = 0.47) (p = 0.24) (p = 0.38) (p = 0.52) (p = 0.90) (p = 0.48) (p = 0.41) (p = 0.94)
(p = 0.01) (p = 0.00) (p = 0.00) NA NA
Z1/01 = 1.39 Z1/00 = 1.75 Z01/00 = 0.15
(p = 0.17) (p = 0.08) (p = 0.87) NA
Z1/01 = –0.92 Z1/00 = –0.02 Z01/00 = 0.66 NA
(p = 0.36) (p = 0.98) (p = 0.51)
Table 5. Continued Variables ROE1 ROE2 ROA MRET
Groups IAS EU-non-IAS Neither-IAS-nor-EU IAS EU-non-IAS Neither-IAS-nor-EU IAS EU-non-IAS Neither-IAS-nor-EU IAS EU-non-IAS Neither-IAS-nor-EU
Variables
Mean µ1 = 9.37 µ01 = 6.37 µ00 = 9.68 µ1 = 16.58 µ01 = 20.64 µ00 = 16.52 µ1 = 5.72 µ01 = 5.08 µ00 = 5.13 µ1 = 13.09 µ01 = 14.65 µ00 = 8.88
Standard deviation σ1 = 9.28 σ01 = 32.22 σ00 = 15.39 σ1 = 14.13 σ01 = 28.72 σ00 = 18.01 σ1 = 4.37 σ01 = 5.57 σ00 = 4.30 σ1 = 21.29 σ01 = 20.21 σ00 = 24.21
Student t-test
Z1/01 = 0.51 Z1/00 = –0.48 Z01/00 = –0.85 Z1/01 = –0.73 Z1/00 = 0.02 Z01/00 = 0.74
NA NA t1/01 = 0.52 t1/00 = 0.65 t01/00 = –0.04 t1/01 = –0.30 t1/00 = 0.88 t01/00 = 1.11
Mann-Whitney U-test
(p = 0.59) (p = 0.52) (p = 0.96) (p = 0.76) (p = 0.38) (p = 0.27)
(p = 0.60) (p = 0..62) (p = 0.39\) (p = 0.46) (p = 0.98) (p = 0.46) NA NA
Groups
Number of cases
Student t-test
LIST
IAS EU-non-IAS Neither-IAS-nor-EU
N1 = 11 (51) N01 = 1 (30) N00 = 1 (52)
NA
χ21/01 = 30.56 χ21/00 = 28.54 χ201/00 = 9.17
(p = 0.00) (p = 0.00) (p = 0.00)
BIG6
IAS EU-non-IAS Neither-IAS-nor-EU
N1 = 49 (51) N01 = 27 (30) N00 = 42 (52)
NA
χ21/01 = 17.63 χ21/00 = 5.03 χ201/00 = 41.69
(p = 0.00) (p = 0.03) (p = 0.00)
NA : Not appropriate
Chi-square-test
Why Swiss Firms Comply Voluntarily with IAS 235 Multivariate analyses To check univariate results, logistic regressions were used, which jointly test hypotheses formulated above. Since several redundant variables were proposed to proxy for size, leverage, profitability, and internationality, regression estimates could be substantially biased because of collinearity among independent variables. Two alternative procedures were used to circumvent this problem. The first one relies on a selection of the most relevant measures of size, leverage, profitability or internationality, based on their explanatory power in univariate analyses. The second procedure uses factor analysis to define a composite variable for each hypothesis by transforming original variables into a smaller set of orthogonal factors. Concerning the first procedure, because of its high correlation with both LASSETS and INT2, LIST was not included in the regression along with these variables. The resulting logit models are thus: PROB = a0 + a1ROA + a2LASSETS + a3 TDTA + a4INT2 + a5DIFF + a6AIP + a7BIG6 + u and: PROB = a′0 + a′1ROA + a′3TDTA + a′5DIFF + a′6AIP + a′7BIG6 + a′8LIST + u′. Three analyses were successively conducted. In model 1, PROB equals 1 if the firm complies with IAS and 0 otherwise. In model 2, PROB equals 1 if the firm complies with IAS and 0 if it only applies EU directives. Finally, model 3 opposes firm complying with EU directives only (PROB = 1) to firms which comply neither with IAS nor EU directives (PROB = 0). Results of logistic regressions are reported on Table 6. With a few exceptions, the univariate findings are confirmed. As expected, IAS firms are significantly larger and more internationally diversified than non-IAS firms. When size and internationality are replaced by listing status, this variable also has a positive and significant coefficient. These results are in accordance with hypotheses H1, H2 and H3. Contrary to the univariate tests, multivariate analyses do not support hypotheses H4 and H8 since the coefficients of DIFF and BIG6 are not significant at usual levels. Finally, in accordance with univariate tests, but contrary to predictions, profitability, leverage and capital intensity do not seem to have an influence on the decision to comply with IAS. © Blackwell Publishers Ltd. 1998.
236
Pascal Dumontier and Bernard Raffournier
Table 6. Logit Regression Estimates Based on Independent Variables
Variables (expected sign)
IAS versus non-IAS firms (model 1)
IAS versus EU-non-IAS firms (model 2)
EU-non-IAS versus neitherIAS-nor-EU firms (model 3)
Coefficient (Wald statistics)
Coefficient (Wald statistics)
Coefficient (Wald statistics)
ROA (–)
–0.017 (0.09)
LASSETS (+)
0.090 (5.72***)
TDTA (+)
1.592 (1.37)
INT2 (+)
0.037 (8.44***)
DIFF (+)
0.011 (1.63)
0.013 (2.52)
0.015 (1.97)
0.009 (0.72)
0.009 (0.68)
0.001 (0.00)
AIP (–)
–0.745 (0.33)
–0.301 (0.07)
–1.575 (0.68)
–0.249 (0.02)
–1.059 (0.48)
–0.496 (0.16)
BIG 6 (+)
0.158 (0.03)
1.102 (1.85)
6.034 (0.06)
6.018 (0.05)
8.205 (0.56)
9.348 (0.12)
LIST (+)
–0.024 (0.29)
–0.041 (0.39)
–0.036 (0.56)
0.226 (3.02*) 1.661 (1.80)
3.449 (1.06)
–0.072 (0.47)
0.201 (5.17**) 4.483 (2.27)
0.034 (3.75*)
1.805 (4.43**)
–0.053 (0.61)
1.708 (2.20)
1.364 (1.61)
0.022 (1.53)
1.886 (3.62*)
Constant (?)
–4.034 (5.23**)
–1.271 (0.89)
–5.617 (0.05)
–8.561 (0.11)
% correctly predicted Model chi–square
70.27 25.65***
68.10 14.91**
70.59 64.18 14.81** 11.67*
0.429 (0.08) –15.372 –10.56 (0.19) (0.15) 68.42 13.05*
66.05 9.57
*** Significant at the 0.01 level. ** Significant at the 0.05 level. * Significant at the 0.10 level.
Similar conclusions can be drawn from the comparison between IAS and EU firms, the only slightly significant variables being size, internationality and listing status. When comparing EU firms with companies which refer neither to IAS nor to EU directives, the influence of internationality and listing status disappears, size being the only significant variable. An alternative two-step procedure was used to better control for collinearity between independent variables, especially between size, internationality and listing status. The first step consisted in transforming © Blackwell Publishers Ltd. 1998.
Why Swiss Firms Comply Voluntarily with IAS 237 original variables into a smaller set of orthogonal factors, by means of a principal component factor analysis. These factors were then included in the logistic regression in place of the independent variables. This procedure circumvents any collinearity problem since, by design, factors are orthogonal. In as much as only 5 factors with eigenvalues greater than 1 were extracted by the principal component factor analysis, all of them were included in the logistic regression. They explain 75% of the variance of the original data (Table 7). The factors identification is based on factor loadings, which measure correlation between factors and original variables. To facilitate identification, a varimax orthogonal rotation was performed. It provides a simple structure, reducing problems associated with variables exhibiting significant loadings on more than one factor or with too many variables related to the same factor. The classification pattern of original variables makes the interpretation of factors relatively obvious. Since ROE1, ROE2, ROA and MRET exhibit a strong positive correlation with factor 1, it is clear that this factor represents profitability. Because of high positive loadings of size and listing status variables, factor 2 can be considered as representing firm size. The fact that size and listing status variables are positively correlated with Table 7. Classification Pattern of Original Variables Variables
Factor 1
Factor 2
Factor 3
Factor 4
Factor 5
ROE1 ROE2 ROA MRET
0.896 0.906 0.823 0.511
0.112 0.006 –0.060 0.304
–0.123 0.269 –0.232 –0.477
–0.107 –0.015 0.209 0.010
0.026 0.124 –0.078 –0.191
TDTA LTDTA
–0.019 –0.092
0.077 –0.106
0.900 0.878
–0.001 –0.047
0.089 –0.235
0.016 0.075 0.053
0.941 0.870 0.726
0.068 –0.084 –0.720
0.067 0.151 0.301
0.114 0.273 –0.110
AIP INT1 INT2 DIFF
0.095 0.154 0.037 –0.048
0.303 0.086 0.214 0.186
0.338 0.083 –0.248 0.608
–0.493 0.872 0.805 0.472
–0.358 0.090 –0.004 –0.029
BIG6
0.026
0.205
–0.038
0.044
0.888
25.2%
42.1%
56.9%
67.5%
75.0%
LASSETS LSALES LIST
Cumulative variance explained
© Blackwell Publishers Ltd. 1998.
238
Pascal Dumontier and Bernard Raffournier
factor 2 means that firms listed on foreign stock exchanges are among the largest in the sample, which is not surprising. Given that the two leverage ratios have significant positive loadings on factor 3, this factor obviously proxies for leverage. Interpreting factor 4 is more critical, since this factor is negatively correlated with AIP and positively correlated with internationality variables and, to a lesser degree, with ownership diffusion. This suggests that the more internationally diversified firms are those with the highest ownership diffusion and that these firms are not capital intensive. Given that this factor opposes internationality and capital intensity, we labelled it ‘intercap’. Big6 is the only variable with a significant loading on factor 5, indicating that this factor represents auditors’ reputation. The individual impact of each factor was assessed by including all of them in the following logistic regression: PROB = b0 + b1 Profitability + b2 Size + b3 Leverage + b4 Intercap + b5 Big6 + v where Profitability, Size, Leverage, Intercap and Big6 are the five previously defined factors. Results reported in Table 8 show that, with only one exception, all factors have the expected sign. With regard to the opposition IAS versus non-IAS firms (model 1), results are consistent with those of univariate tests. Factor 2 (size and listing status) and factor 4 (internationality, ownership diffusion and capital intensity) exhibit coefficients statistically significant at the 1% level, indicating that large and internationally diversified companies with a low percentage of fixed assets and a strong ownership diffusion are more likely to comply with IAS than small, domestic and capital intensive firms controlled by a small number of stockholders. Factors representing profitability, leverage and auditors’ reputation are not statistically significant. Nevertheless, with a percentage of correct prediction superior to 71% and a chi-square value statistically significant at the 1% level, the model is highly significant. Similar results are obtained when comparing IAS firms to those which only refer to EU directives (model 2), but factors 2 and 4 are now significant at the 10% level only. Again, results of univariate tests are confirmed: firms which comply with IAS are larger and more internationally diversified than those which are satisfied with applying EU regulation. In the last model (EU-non-IAS versus neither-IAS-nor-EU firms), no factor is significant at the usual levels, indicating that the hypotheses tested fail to explain why some firms declare their accounts in conformity with EU © Blackwell Publishers Ltd. 1998.
Why Swiss Firms Comply Voluntarily with IAS 239 Table 8. Logit Regression Estimates on Factors
Factors (expected sign)
IAS versus non-IAS firms (model 1)
IAS versus EU-non-IAS firms (model 2)
EU-non-IAS versus neitherIAS-nor-EU firms (model 3)
Coefficient (Wald statistics)
Coefficient (Wald statistics)
Coefficient (Wald statistics)
Profitability (factor 1) (–)
–0.304 (1.35)
–0.132 (0.18)
–0.187 (0.42)
Size (factor 2) (+)
1.113 (7.64***)
1.234 (3.01*)
0.687 (2.21)
Leverage (factor 3) (+)
0.023 (0.008)
–0.402 (1.07)
0.310 (1.15)
Intercap (factor 4) (+)
0.973 (9.24***)
0.902 (3.66*)
0.278 (0.62)
Big6 (factor 5) (+)
0.131 (0.26)
1.164 (2.49)
0.685 (2.35)
Constant (?)
–0.269 (1.07)
0.598 (2.17)
–0.150 (0.16)
% correctly predicted Model chi–square
71.43 23.83***
66.67 10.95*
64.58 7.68
*** Significant at the 0.01 level. ** Significant at the 0.05 level. * Significant at the 0.10 level.
directives while others do not. This result was expected since, as previously discussed, compliance with EU directives does not imply, for Swiss firms, a considerable loss of discretion in accounting choices, nor a significant increase of disclosure. Conclusion The aim of the paper was to identify the motivations of Swiss listed companies which voluntarily comply with IAS. Switzerland is of particular interest for such studies because of its low level of accounting regulation and its accounting permissiveness. In this context, compliance with IAS is particularly costly since it implies additional disclosure and renunciation of considerable discretion in accounting practices. © Blackwell Publishers Ltd. 1998.
240
Pascal Dumontier and Bernard Raffournier
On the basis of prior literature on voluntary disclosure and auditing, we formulated a set of hypotheses which were tested on a sample of 133 Swiss listed companies. Univariate analyses showed a positive influence of size, internationality, listing status, auditor type and ownership diffusion on voluntary compliance with IAS. Inversely, no significant relationship was found for leverage, profitability and capital intensity. Multivariate analyses designed to control for collinearity among independent variables confirmed these results. They revealed that firms which comply with IAS are larger, more internationally diversified, less capital intensive and have a more diffuse ownership. Similar results, although less significant, were obtained in opposing firms complying with IAS to those which refer only to EU directives. In contrast, this latter group did not appear as statistically significant from firms which do not mention any set of accounting standards. Taken as a whole, these findings suggest that political costs and pressures from outside markets play a major role in the decision to apply IAS. Inversely, they do not validate the hypothesis that voluntary compliance with stringent accounting standards is used to solve monitoring problems between managers, shareholders and creditors. Notes 1. Examples of such countries are Nigeria, Malaysia and Singapore (see Nobes, 1994). 2. Cited in Price Waterhouse—Befec, IAS—Normes comptables internationales, Editions Francis Lefebvre, Paris, 1995, pp. 41–43. 3. The SBF 120 index is a French market index. It includes the 120 shares most actively traded on the Paris Stock Exchange. 4. This section provides an overview of the main features of the Swiss accounting regulation. For a more complete description, see Raffournier (1995a) and, for an historical perspective, Achleitner (1995). 5. The FER should gain additional support in the near future since a new listing regulation which should have been implemented by the end of 1996, but whose application date has been deferred, states that the consolidated financial statements of listed companies will, in the future, have to comply with the FER recommendations. 6. For a survey of international classifications of financial reporting systems, see Nobes, 1984. 7. See Bilan, No. 7–8, juillet-août 1992, p. 77. 8. This assertion is indirectly supported by Lang and Lundholm (1996) who recently reported that firms with more information disclosure policies exhibit a larger analyst following and more accurate analyst earnings forecast. 9. Creation of hidden reserves is expressly forbidden by IAS 1. 10. Source: Geneva Stock Exchange, 1994 report, p. 27. 11. Two other arguments justify this grouping. First, because of the permissiveness of FER recommendations, there should be no significant differences between firms which comply with these standards and those which do not. Secondly, since FER recommendations © Blackwell Publishers Ltd. 1998.
Why Swiss Firms Comply Voluntarily with IAS 241 are generally considered as Swiss GAAP, many large firms probably apply these standards without finding it necessary to mention them. 12. The fact that EU directives are more permissive than IAS does not mean, of course, that IAS are systematically more stringent than the accounting regulation of any particular country within the European Union. 13. Many large firms do not indicate the amount of sales realised in Switzerland and consider Europe as a unique market in their breakdown of sales by geographical area. That explains why the sales-outside-Switzerland-on-total-sales ratio is measured for 92 firms only. 14. Guides des actions suisses 94/95, published by Verlag Finanz und Wirtschaft AG, Zürich.
References Achleitner, A.-K., “The history of financial reporting in Switzerland,” in European Financial Reporting: A History, P. Walton, ed. (London: Academic Press, 1995), pp. 241–258. Anderson, D., J. Francis and D. Stokes, “Auditing, directorships and the demand for monitoring,” Journal of Accounting and Public Policy (1993) vol. 12, pp. 353–375. Blacconiere, W., R. Bowen, S. Sefcik and C. Stinson, “Determinants of the use of regulatory accounting principles by savings and loans,” Journal of Accounting and Economics (1991) vol. 14, pp. 167–201. Buzby, S., “Company size, listed versus unlisted stocks, and the extent of financial disclosure,” Journal of Accounting Research (1975) vol. 13, pp. 16–37. Chow, C., “The demand for external auditing: Size, debt and ownership influences,” The Accounting Review (1982) vol. 57, pp. 272–291. Chow, C. and A. Wong-Boren, “Voluntary financial disclosure by Mexican corporations,” The Accounting Review (1987) vol. 62, pp. 533–541. Cooke, T., “Disclosure in the corporate annual reports of Swedish companies” Accounting and Business Research (1989) vol. 19, pp. 113–124. Cooke, T., “The impact of size, stock market listing and industry type on disclosure in the annual reports of Japanese listed corporations,” Accounting and Business Research (1992) vol. 22, pp. 229–237. Ettredge, M., D. Simon, D. Smith and M. Stone, “Why do companies purchase timely quarterly reviews,” Journal of Accounting and Economics (1994) vol. 18, pp. 131–155. Firth, M., “The impact of size, stock market listing, and auditors on voluntary disclosure in annual reports,” Accounting and Business Research (1979) vol. 9, pp. 273–280. Gray, S., G. Meek and C. Roberts, “International capital market pressures and voluntary annual report disclosures by U.S. and U.K. multinationals,” Journal of International Financial Management and Accounting (1995) vol. 6, pp. 43–68. Hossain, M., M. Perera and A. Rahman, “Voluntary disclosure in the annual reports of New Zealand companies,” Journal of International Financial Management and Accounting (1995) vol. 6, pp. 69–87. Hossain, M., L. Tan and M. Adams, “Voluntary disclosure in an emerging capital market: Some empirical evidence from companies listed on the Kuala Lumpur stock exchange,” The International Journal of Accounting (1994) vol. 29, pp. 334–351. Jensen, M. and W. Meckling, “Theory of the firm: managerial behavior, agency costs and ownership structure,” Journal of Financial Economics (1976) vol. 3, pp. 305–360. Lang, M. and R. Lundholm, “Cross-sectional determinants of analyst ratings of corporate disclosures,” Journal of Accounting Research (1993) vol. 31, pp. 246–271. Lang, M. and R. Lundholm, “Corporate disclosure policy and analyst behavior,” The Accounting Review (1996) vol. 71, pp. 467–492. © Blackwell Publishers Ltd. 1998.
242
Pascal Dumontier and Bernard Raffournier
Meeks, G., C. Roberts and S. Gray, “Factors influencing voluntary annual report disclosures by U.S., U.K. and continental European multinational corporations,” Journal of International Business Studies (1995) pp. 555–572. Myers, S., “Determinants of corporate borrowing,” Journal of Financial Economics (1977) vol. 4, pp. 147–175. Nobes, C., International Classification of Financial Reporting (Beckenham, UK: Croom Helm, 1984), 144p. Nobes, C., A study of the International Accounting Standards Committee (London: Coopers & Lybrand, 1994), 35p. Prémont, P. and A. Dionne, “La propension à présenter des états vérifiés en contexte d’émission d’actions,” Finéco (1993) vol. 3, pp. 177–189. Raffournier, B., “Switzerland,” in European Accounting Guide, D. Alexander and S. Archer, eds. (San Diego: Harcourt Brace, 1995a), pp. 1249–1317. Raffournier, B., “The determinants of voluntary financial disclosure by Swiss listed companies,” The European Accounting Review (1995b) vol. 4, pp. 261–280. Singhvi, S. and H. Desai, “An empirical analysis of the quality of corporate financial disclosure,” The Accounting Review (1971) vol. 46, pp. 129–138. Wallace, R. and K. Naser, “Firm-specific determinants of the comprehensiveness of mandatory disclosure in the corporate annual reports of firms listed on the stock exchange of Hong Kong,” Journal of Accounting and Public Policy (1995) vol. 14, pp. 311–368. Watts, R. and J. Zimmerman, Positive accounting theory (Englewood Cliffs: Prentice-Hall, 1986), 388p.
© Blackwell Publishers Ltd. 1998.
Why Swiss Firms Comply Voluntarily with IAS 243 Appendix 1. Correspondence between IAS and ARR IAS IAS 1 – IAS 2 – IAS 4 – IAS 5 –
Disclosure of accounting policies Inventories Depreciation accounting Information to be disclosed in financial statements
IAS 7 – IAS 8 –
Cash flow statements Net profit or loss for the period, fundamental errors and changes in accounting policies Research and development costs
IAS 9 –
IAS 10 – Contingencies and events occurring after the balance sheet date IAS 11 – Construction contracts IAS 12 – Accounting for taxes on income IAS 13 – Presentation of current assets and current liabilities IAS 14 – Reporting financial information by segment IAS 15 – Information reflecting the effects of changing prices IAS 16 – Property, plant and equipment IAS 17 – Accounting for leases IAS 18 – Revenue IAS 19 – Retirement benefit costs IAS 20 – Accounting for government grants and disclosure of government assistance IAS 21 – The effects of changes in foreign exchange rates IAS 22 – Business combinations IAS 23 – Borrowing costs IAS 24 – Related party disclosures IAS 25 – Accounting for investments IAS 26 – Accounting and reporting by retirement benefit plans IAS 27 – Consolidated financial statements and accounting for investments in subsidiaries IAS 28 – Accounting for investments in associates IAS 29 – Financial reporting in hyperinflationary economies IAS 30 – Disclosures in the financial statements of banks and similar financial institutions IAS 31 – Financial reporting of interests in joint ventures IAS 32 – Financial instruments: Disclosure and presentation No standard No standard No standard ARR 1: ARR 2: ARR 3: ARR 4: ARR 5: ARR 6: ARR 7: ARR 8: ARR 9: ARR 10: ARR 11: ARR 12:
ARR ARR 3 No recommendation No recommendation ARR 1, ARR 7 and ARR 8 ARR 6 No recommendation Partially covered by ARR 9 No recommendation No recommendation ARR 11 ARR 7 No recommendation No recommendation No recommendation No recommendation No recommendation No recommendation No recommendation ARR 4 No recommendation No recommendation No recommendation No recommendation No recommendation ARR 2 and ARR 5 No recommendation No recommendation No recommendation No recommendation No recommendation ARR 9 ARR 10 ARR 12
Components of individual company accounts and consolidated financial statements Consolidated financial statements Generally accepted accounting principles The translation of financial statements expressed in foreign currencies for consolidation purpose Valuation directives for consolidated financial statements Funds flow statement Presentation and format of the consolidated balance sheet and income statement Notes to the consolidated financial statements Intangibles Consolidated financial statements of insurance companies Taxes in consolidated financial statements Interim reporting © Blackwell Publishers Ltd. 1998.
244
Pascal Dumontier and Bernard Raffournier
Appendix 2. Some major differences between IAS and ARR IAS Accounting policies Deviations from consistency in presentation and valuation Consolidated financial statements Criteria for inclusion in consolidation Exclusion of subsidiaries whose activities are dissimilar from those of the other enterprises within the group Elimination of intercompany profits Treatment of goodwill
Translation of foreign financial statements Translation methods allowed
Choice of translation methods Funds flow statement Basis of funds flow statement
© Blackwell Publishers Ltd. 1998.
ARR
prohibited (IAS 1)
allowed (ARR 3)
control (IAS 27) prohibited (IAS 27)
not specified (ARR 2) allowed ARR 2)
mandatory (IAS 27) recognised as an asset (IAS 22)
discretionary (ARR 2) recognised as an asset or adjusted against shareholders’ interests (ARR 2)
closing rate temporal (IAS 21)
closing rate temporal monetary/non monetary (ARR 4) free (ARR 4)
regulated (IAS 21) cash and cash equivalents (IAS 7)
cash or net cash or net monetary working capital or net financial position (ARR 6)
Why Swiss Firms Comply Voluntarily with IAS 245 Appendix 3. Correlation matrix of independent variables AIP AIP BIG INT1 INT2 LIST LASSETS LSALES LTDTA MRET ROA ROE1 ROE2 TDTA DIFF
1.00000 –0.11212 –0.25820 –0.30504 –0.00098 0.20280 –0.02101 0.33921 –0.06981 –0.07550 0.11686 0.10771 0.14824 –0.07317
BIG6 1.00000 0.15475 0.13181 0.09782 0.25057 0.32800 –0.20231 0.00598 –0.00084 0.03854 0.06430 0.00417 0.06970
LTDTA MRET LTDTA MRET ROA ROE1 ROE2 TDTA DIFF
1.00000 –0.37195 –0.23968 –0.20220 0.06357 0.71011 –0.04502
1.00000 0.53025 0.45832 0.22220 –0.30135 0.04849
INT1
INT2
LIST
LASSETS LSALES
1.00000 0.68719 1.00000 0.23268 0.41672 1.00000 0.18981 0.23714 0.57334 1.00000 0.23630 0.27399 0.56681 0.87669 1.00000 –0.02629 –0.25058 –0.13501 –0.05983 –0.26031 0.13008 0.22628 0.25419 0.25284 0.26481 0.25053 0.14620 0.11291 –0.05494 0.06718 0.04908 0.02740 0.09128 0.11162 0.16689 0.13554 –0.05310 0.05981 0.05131 0.09029 0.02686 –0.23405 0.01266 0.12999 0.02765 0.29429 0.19457 0.15000 0.17626 0.23418
ROA
ROE1
ROE2
1.00000 0.60444 1.00000 0.61761 0.79119 1.00000 –0.24039 –0.16295 0.23042 0.10217 –0.04659 –0.02528
TDTA
DIFF
1.00000 0.00878
1.00000
© Blackwell Publishers Ltd. 1998.
