The Relative Value Relevance of Earnings and Book Value in Malaysia and Singapore.

Transcription

1 The Relative Value Relevance of Earnings and Book Value in Malaysia and Singapore. Dalilawati Zainal, Muhd Kamil Ibrahim, Khairul Anuar Kamarudin and Jagjit Kaur Draft: Last updated on 28 February, 2005 Abstract This study examines the relative value relevance of earnings and book value in two Asian countries with similar accounting systems, Malaysia and Singapore. Based on the similarities of accounting systems between the two countries, this study hypothesizes that book value is more value relevant than earnings in determining market value of equity in Malaysia and Singapore. Results from Wald chi-square and OLS regression tests provide evidence confirming the hypothesis. This study also reveals that earnings sign, firm size and macroeconomic events affect the relative value relevance in both countries under study, whereby the results seem to be driven by firms reporting negative earnings and large firm only. The results are not robust for firms reporting positive earnings, small firms and individual year samples (represent macroeconomic events). Overall, this study provides evidence concerning the relative value relevance of earnings and book value, being two important summary measures of financial statements, from Asian region. 1

2 1.0 Introduction The value relevance of accounting numbers (earnings and book value) has been a major focus of accounting research since the late 1960s. While earlier research focused on the value relevance of earnings in explaining stock returns (Ball and Brown, 1968; Beaver, Clark and Wright, 1979; Bowen, 1981; Lipe, 1986), recent studies have included book value, as another important variable in determining market value of equity, in addition to earnings (Ohlson, 1995; Collins, Maydew and Weiss, 1997; Francis and Schipper, 1999; Ely and Waymire, 1999; Ibrahim, Mohd Said, Abd Latif and Abd Shukur, 2004). Ohlson (1995) showed that earnings and book value are value relevant and they complement each other in explaining market value of equity. Furthermore, Collins et al. (1997), Francis and Schipper (1999) and Ely and Waymire (1999), who examined changes in the value relevance of earnings and book value for the US firms, reported a decline in the value relevance of earnings, having been replaced by an increase in the value relevance of book value. Collins et al. reported that increased reporting of losses, nonrecurring items and economic importance of unreported intangible assets explain much of the shift in value relevance from earnings to book value. Recently, Ibrahim et al. (2004) revealed that Ohlson valuation model is more suitable than the balance sheet identity model in explaining association between purchased goodwill and market valuation for Malaysian firms. Since most of prior research concentrated in the US only, researchers have now shifted their focus to examine the value relevance of accounting numbers in other countries such as Japan, the European and Asian countries owing to the increased importance of accounting numbers in global market following the economic globalization (King and Langli, 1998; Graham and King, 2000; Arce and Mora, 2002; Black and White, 2003). These studies concluded that diversity in institutions (political, economic, legal, taxation, financial and professional system) and cultures across the globe has resulted in differences in accounting systems, which in turns lead to the differences in the relative value relevance of earnings and book value across countries. While King and Langli (1998), Graham and King (2000), Arce and Mora (2002) and Black and White (2003) have provided evidence on the effect of difference in accounting systems on the relative value relevance of earnings and book value, this study provides evidence on the effect of similarity in accounting systems on the relative value relevance of earnings and book value in two Asian countries, Malaysia and Singapore. Given the similarities of accounting systems between Malaysia and Singapore (Nair and Frank, 1980; Gray, 1988; MacGregor, Hossain and Yap, 1997; Fan and Wong, 2002; Deesomsak, Paudyal and Pescetto, 2004), both countries are expected to have a similar relative value relevance of earnings and book value in explaining market value of equity. Specifically, this study aims to determine whether earnings or book value are more value relevant in explaining market value of equity in Malaysia and Singapore. In addition, this study also examines whether the factors such as earnings sign, firm size and macroeconomic events affect the relative value relevance of earnings and book value in explaining market value of equity in the two countries. The rest of this paper is divided into the following sections: literature review (2.0), hypotheses development (3.0), research method (4.0), results (5.0) and conclusions (6.0). 2

3 2.0 Literature Review 2.1 The Relative Value Relevance of Earnings and Book Value across Countries There has been a growing number of literature documenting differences in the relative value relevance of earnings and book value across countries (King and Langli, 1998; Graham and King, 2000; Arce and Mora, 2002; Black and White, 2003). King and Langli (1998) examined the differences in the value relevance of earnings and book value in three European countries (Germany, the UK and Norway) using a fifteen-year sample period ( ). These countries provide useful examples of two accounting systems: the UK (the Anglo-Saxon system), Germany (the Continental system) and Norway (lies somewhere between the two systems). Countries under the Anglo-Saxon system give greater focus on equity holders (market-orientated), implement common law legal system, have a separate accounting-tax system and regulated by the professional bodies. While countries under the Continental system give greater emphasis on debt holders (creditororientated), implement code-based legal system, have a strong tax influence on accounting and the presence of governmental rather than professional regulatory bodies. When stock price is regressed on earnings and book value, King and Langli found that the UK provides the highest value relevance of earnings and book value (0.718), followed by Norway (0.591) and Germany (0.428). In examining the relative value relevance, King and Langli decomposed the basic valuation model into two separate models that consider the individual value relevance of earnings and book value, whereby stock price is regressed separately on earnings and book value. Results from two separate regressions revealed that earnings are more value relevant than book value in the UK, while book value is more value relevant than earnings in Germany and Norway. Graham and King (2000) examined the value relevance of earnings and book value in six Asian countries (Indonesia, Malaysia, Thailand, the Philippines, Taiwan and Korea). Graham and King observed that Korea (0.683) and the Philippines (0.680) have higher explanatory power of earnings and book value compared to Thailand (0.397), Indonesia (0.308), Malaysia (0.277) and Taiwan (0.169). These differences are generally consistent with differences in accounting practices across countries. However, since Korean accounting practice is strongly influenced by tax law, Graham and King did not expect a high value relevance of earnings and book value in Korea. Overall, their study provided evidence on the differences in the value relevance of earnings and book value across different countries in Asian region. Graham and King also found that book value has greater value relevance than earnings in all six Asian countries under study. Arce and Mora (2002) examined the differences in the relative value relevance of earnings and book value across eight European countries that represent the most important capital markets in Europe. The countries are classified into two groups: the Continental system (Belgium, France, Germany, Italy, Switzerland and Spain) and the Anglo-Saxon system (the UK and the Netherlands). Generally, their findings indicated that earnings are more value relevant than book value in the Anglo-Saxon (market-orientated) countries, while book value is more value relevant than earnings in the Continental (creditor-orientated) countries, consistent with King and Langli (1998). Black and White (2003) investigated the effect of differences in accounting systems in Germany (the Continental system), Japan (lies somewhere between the two systems) and the 3

4 US (the Anglo-Saxon system) on the relative value relevance of earnings and book value across countries for over thirteen-year ( ). Findings from regression test indicated that earnings are more value relevant than book value in the US, while book value is more value relevant than earnings in Germany and Japan. Black and White also examined the effect of earnings sign, firm size and macroeconomic events on the relative value relevance. They observed that the results on the relative value relevance for Germany are robust, as book value is more value relevant than earnings for both positive and negative earnings sample, in all years with sufficient sample sizes, and for all size quartiles. For Japan, the results are mixed between the Japan-Parent Only and Japan-Consolidated sample. While in the US, the results seem to be driven by positive earnings sample, certain years and large firms only. Therefore, their findings suggested that earnings sign, firm size and macroeconomic events affect the relative value relevance of earnings and book value in Japan and the US only. Generally, differences in accounting system may lead to a difference in the relative value relevance of earnings and book value across countries. While the Anglo-Saxon countries place greater emphasis on earnings compared to book value, the Continental and mixed (lie between the Anglo-Saxon and the Continental system) countries and place greater emphasis on book value compared to earnings. 2.2 Accounting System in Malaysia and Singapore There has been much similarity of accounting systems between Malaysia and Singapore (Nair and Frank, 1980; Gray, 1988; MacGregor, Hossain and Yap, 1997; Fan and Wong, 2002; Deesomsak, Paudyal and Pescetto, 2004). Nair and Frank (1980) categorized Malaysia and Singapore under the same group (the British Commonwealth model) that shares some common attributes in accounting practices. According to Deesomsak et al. (2004), both countries are geographically close and share common recent British colonization history. Being the old British Empire, most aspects of accounting system in the two countries are inherited from the British (Anglo-Saxon) system. For example, both countries adopt common law and market-based financial system, adhere to the true and fair view presentation of financial reports, and emphasize on public disclosure of information and external audit. Both countries also have professional accounting bodies and standard setting bodies independent of government. MacGregor et al. (1997) noted that Malaysia and Singapore show very marked similarities in structures of their regulatory frameworks, form of their corporate reports, culture influence and general attitude towards professionalism in accounting. Both countries have similar Companies Acts as the manner in which government bodies regulate the accounting profession and similar professional membership recognition and requirements. In addition, both countries have almost identical accounting standards based on the International Accounting Standards (IAS). Since both countries have adopted International Accounting Standards without significant alteration, it is difficult to detect any significant differences in financial accounting practices and hence values. Even though most aspects of accounting systems in Malaysia and Singapore are inherited from the British system, Gray (1988) noted that there is a difference in culture between the Asian countries (Malaysia and Singapore) and the UK. Gray noted that accountants in the UK exercise an individual professional judgment and maintenance of professional selfregulation, while accountants in Asian countries emphasize on compliance with prescriptive legal requirements and statutory control, similar to the Continental countries. 4

5 Deesomsak et al. (2004) found that type of ownership structure is another important factor that differentiates between Malaysia and Singapore with the UK. While the UK has a public ownership structure (high earnings informativeness), Malaysia and Singapore have more concentrated type of ownership structure (low earnings informativeness), similar to the Continental countries. Fan and Wong (2002), who examined the informativeness of accounting earnings in East Asian countries including Malaysia and Singapore, also observed lower earnings informativeness in Malaysia and Singapore as a result of the concentrated type of corporate ownership structure in the countries. In summary, both countries share common attributes of accounting practices similar to the UK (Anglo-Saxon system). Yet, there are several aspects of accounting system that differentiate Malaysia and Singapore with the Anglo-Saxon system such as cultures (Gray, 1988) and types of corporate ownership (Fan and Wong, 2002; Deesomsak et al., 2004), which place Malaysia and Singapore similar to those countries in the Continental system. The description of accounting system in Malaysia and Singapore shows that accounting system in both countries lie between the Anglo-Saxon and the Continental system, similar to Norway (King and Langli, 1998) and Japan (Black and White, 2003). 3.0 Hypotheses Development This study examines the relative value relevance of earnings and book value in two Asian countries with similar accounting systems, Malaysia and Singapore. King and Langli (1998) and Black and White (2003), who examined the relative value relevance of earnings and book value in Norway and Japan (countries that lie between the Anglo-Saxon and the Continental system) respectively, found that book value has greater value relevance than earnings in those countries. Since accounting systems in Malaysia and Singapore also lie between the Anglo-Saxon and the Continental system, book value is expected to be more value relevance than earnings in both countries. Therefore, the following hypothesis is developed: HI: The relative value relevance of book value is greater than earnings in Malaysia and Singapore. Apart from examining the relative value relevance, this study also includes an investigation on the effect of earnings sign, firm size and macroeconomic events on the relative value relevance of earnings and book value in Malaysia and Singapore. Previous literature revealed several factors that may affect the relative value relevance of earnings and book value, for example, earnings sign, firm size and macroeconomic events (Hayn, 1995; Collins et al., 1997; Black, 2003; Black and White, 2003). According to Hayn (1995), the value relevance of earnings depends on the sign of earnings. Hayn reported that negative earnings (loss) are less value relevant than positive earnings (profit) in determining the future prospect of a firm based on the finding that loss is more weakly associated with returns than profits. Since investors can always liquidate the firm rather than suffer from indefinite losses, investors perceive loss as temporary, thus lead to the lower value relevance of negative earnings compared to positive earnings. Similarly, Collins et al. (1997) in their research also reported a lower value relevance of negative earnings compared to positive earnings. Collins et al. revealed that the relative value relevance of earnings is lower than book value when earnings are negative, while the relative value relevance of earnings is higher than book value when earnings are positive. 5

6 In examining the effect of firm size on the value relevance of earning and book value, Black (2003), who examined the usefulness of financial statement components in firm valuation, revealed that the value relevance of accounting numbers differs across firm life cycle. While both earnings and book value are insignificant in determining firm valuation for start-up (small) firms, they are significant in determining firm valuation for mature (large) firms. For growth (medium size) firms, book value is significant, but not earnings. Their findings revealed that firm size affects the value relevance of earnings and book value in the US. Similarly, Black and White (2003) examined the effect of earnings sign, firm size and macroeconomic events on the relative value relevance of earnings and book value in three countries, the US, Japan and Germany. Their findings revealed that these three factors affect the relative value relevance of earnings and book value in the US and Japan. Based on findings from previous literature (Hayn, 1995; Collins et al., 1997; Black, 2003; Black and White, 2003), this study hypothesizes: H2: Earnings sign affects the relative value relevance of earnings and book value in Malaysia and Singapore. H3: Firm size affects the relative value relevance of earnings and book value in Malaysia and Singapore. H4: Macroeconomic events affect the relative value relevance of earnings and book value in Malaysia and Singapore. 3.0 Research Method 3.1 Research Design Generally, this study was conducted on firms listed on the main board of the Kuala Lumpur Stock Exchange (KLSE), or currently known as Bursa Malaysia and Singapore Stock Exchange (SSE). The population comprised all firms listed on the main board of the KLSE and SSE across six-year period from 1998 to To examine the relative value relevance of earnings and book value in Malaysia and Singapore, financial statements from six accounting periods ( ) were used. Data for this study were extracted from the Thomson Financial Datastream over six-year period from 1998 to This study requires four variables from financial statements: market and book value of equity, earnings before extraordinary items and total assets. Definition of variables used in the study is summarized in Table 3.3. Table 3.1: Definition of Variables Used in the Study Variables Descriptions Symbol Market Value of Equity Market capitalization= MVE Ordinary shares outstanding x Share price Book Value of Equity Common shareholders' fund BVE Earnings Earnings before extraordinary items EARN Total Assets Total fixed and current assets TA This study uses earnings before extraordinary items rather than net income or earnings after extraordinary items. This is because earnings before extraordinary items are better captured 6

7 the permanent component of earnings as compared to net income or earnings after extraordinary items (Arce and Mora, 2002). Net income or earnings after extraordinary items incorporate transitory components such as extraordinary items and non-recurring items that could mislead investors when valuing shares. 3.2 Samplings Sampling criteria as set by Black and White (2003) are used in selecting sample for this study. First, firms that are listed under finance sector are excluded because of their unique features and business activities. Then, a firm-year is included as observation if all such variables (market and book value of equity, earnings and total assets) are available for a given fiscal year. Firm-year with any missing variables is excluded. Next, firm-years with negative book values are deleted as these firms are likely to be in financial distress and might have a different relationship between their accounting numbers and market value of equity than healthy firms. By restricting the sample to firm-years with positive book value only, any effect from the financial distress firms on the result of the study (if any) can be avoided or at least, reduced. Finally, any value of variables (earnings before extraordinary items, book value and market value) that positioned at top and bottom 1 percent are removed to eliminate extreme observations, which are not representative of the samples as a whole. These restrictions on the sample will allow the model to fit better than it would fit unrestricted data. That is, the explanatory power of earnings and book value in the sample is likely to be greater than for an unrestricted sample. Second, the samples across two countries will be more homogeneous, and the effects of different business cycles in the two countries will be reduced. Table 3.2: Sample Selection and Size Sample Selection Firm-years Thomson Financial Datastream (excluding firms in finance sector) Missing observations of market value of equity (MVE), book value of (269) equity (BVE), earnings (EARN) and total assets (TA) Negative book value (113) Top and bottom 1% of MVE, BVE and EARN * (126) Sample Size 3850 Represented by: Malaysia 2406 Singapore 1444 Sample Size 3850 *Note: The total for all variables deleted is approximately 2% Table 3.1 summarizes the sample selection and size used for the study. After excluding firms in finance sectors, firms with missing observation of variables and negative book value, and firms positioned at the top and bottom one percent of market and book value of equity and earnings, the final sample size for this study is 3850 firm-year observations. 7

8 These observations are represented by Malaysia and Singapore with 2406 observations and 1444 firm-year observations respectively. 3.3 Analysis Of Data This study uses the following model to test the hypotheses: MVE jt = β 0 + β 1 EARN jt + β 2 BVE jt + e jt (1) Where: MVE j,t EARN j,t BVE j,t = Market value of equity for firm j at the end of fiscal year t = Earnings before extraordinary items for firm j for fiscal year t = Book value of equity for firm j at the end of fiscal year t Theoretically derived by Ohlson (1995), this model has been used in numerous accounting studies in examining the value relevance of accounting numbers (Collins et al., 1997; Francis and Schipper, 1999; Ely and Waymire, 1999; Ibrahim et al., 2004c). The analysis of the study is based on cross-sectional regressions of earnings and book value of equity (independent variables) on market value of equity (dependent variable). This study employed a methodology, which is derived and used by Biddle et al. (1995) in assessing the relative information content of net income, cash flows and net sales for stock returns. Following Biddle et al. (1995), this study used the Wald chi-square statistic (adjusted for heteroskedasticity) to compare the ability of two independent variables in equation (1) to explain a dependent variable under the null hypothesis that the explanatory power of both variables is equal. Thus, rejection of the null hypothesis indicates that one of the independent variables has greater explanatory power than the other. Then, the variable with greater explanatory power is derived by comparing the r-squares of the two competing independent variables in the following equations: MVE jt = β 0 + β 1 EARN jt + e jt MVE jt = β 0 + β 1 BVE jt + e jt (1a) (1b) In examining the relative value relevance of earnings and book value in Malaysia and Singapore (H1), the Wald chi-square test (adjusted for heteroskedasticity) is performed on equation (1) for each country sample. Chi-square test is chosen because of the categorical (non-parametric) data used in this study. Then, the r-squares derived from the Ordinary- Least-Square (OLS) regression tests are compared for equation (1a) and (1b) for each country sample. These tests are performed using the MICROFIT 4.0 software package. One potential econometric problem when estimating cross-sectional valuation models is the problem of heteroskedastic disturbance, which arise from the fact that large or small firms tend to produce large or small disturbance. This issue has been discussed by many researchers (Biddle et al., 1995; Black and White, 2003; Ibrahim et al., 2004). According to Gujarati (1995), if heteroskedasticity is present, then the usual OLS estimators, although unbiased, no longer exhibit minimum variance among all linear unbiased estimators. In short, they are no longer the best linear unbiased estimator. In order to overcome the heteroskedasticity problem, a procedure known as heteroskedasticity-consistent covariance matrix estimators (HCCME) is established to obtain consistent estimates of the variances and covariances of OLS estimators even if there is heteroskedasticity (White, 1980). White s heteroskedasticity-corrected standard errors are 8

9 available with the MICROFIT software package as a standard output. Then, the Wald test of restriction based on adjusted covariance matrix (adjusted for heteroskedasticity) is performed to arrive at chi-square statistic. These procedures are conducted to overcome the heteroskedasticity problem in this study, thus produce better and unbiased results. Apart from examining the relative value relevance of earnings and book value in Malaysia and Singapore, this study also examine the effect of earnings sign (Hayn, 1995; Collins et al., 1997; Black and White, 2003), firm size (Black, 2003; Black and White, 2003) and macroeconomic events (Black and White, 2003) on the relative value relevance. Following Black and White (2003), each country sample is divided into positive earnings and negative earnings samples, four size quartiles (measured by total assets), individual year sample, to examine the effect of earnings sign (H2), firm size (H3) and macroeconomic events on the relative value relevance. The same statistical tests as being used to test H1 are applied to examine the effect of earnings sign (H2), firm size (H3) and macroeconomic events (H4) on the relative value relevance of earnings and book value in the two countries. 4.0 Results This section discusses results of the study, which include descriptive statistics, test on the relative value relevance of earnings and book value and the effect of earnings sign, firms size and macroeconomic events on the relative value relevance. 4.1 Descriptive Statistics Table 4.1: Descriptive Statistics for Selected Variables Variables N Mean Standard Deviation Panel A: Malaysia Market Value ,205 1,609,690 Earnings , ,794 Book Value , ,267 Assets ,298,461 2,611,378 Panel B: Singapore Market Value ,165 2,320,309 Earnings , ,433 Book Value ,659 1,531,239 Assets ,830,369 5,163,023 Notes: N is the number of firm-years from 1998 to 2003 for Malaysia and Singapore. Market Value is the market value of equity at the end of fiscal year. Earnings are earnings before extraordinary items for the fiscal year. Book Value is the book value of common / ordinary equity at the end of fiscal year. Assets are total assets at the end of the fiscal year. All figures for Malaysia and Singapore are in Ringgit Malaysia (RM). 9

10 Results of descriptive statistics are shown in Table 4.1. The table shows the mean and standard deviation for selected variables: market and book value of equity, earnings and assets, by country. The data indicates that sample size used for Malaysia is larger than Singapore. However, the average (mean) market and book value of equity, earnings and assets are much larger for Singaporean sample compared to Malaysian sample. The results signify the difference in size of capital market between Malaysia and Singapore. While Malaysia is known as emerging capital market, Singapore is already well-known as one of the Tiger Economy in Asia with more established capital market (MacGregor et al., 1997; Deesomsak et al., 2004). Apart from descriptive statistics, this study also compares differences in means for variables using t-test statistics. Table 4.2 shows t-test result for differences in means for variables: market value and book value of equity, earnings and assets between Malaysia and Singapore. The results indicate that there is a significant difference in the average size of market and book value of equity and assets (p-value: 0.000), consistent with MacGregor et al. (1997). However, there is no significant difference in the average size of earnings between Malaysia and Singapore (p-value > 0.05). These findings signify that Malaysia is more efficient in terms of assets management compared to Singapore. Given the same level of resources, Malaysia is able to produce more profit compared to Singapore based on the year 1998 to 2003 data. Table 4.2: T-test Results for Mean Differences across Country Samples: Malaysia versus Singapore Variables t-statistic p-value Market Value Earnings Book Value Assets To examine the effect of firm size on the value relevance of earnings and book value in the two countries, each country sample is split into quartiles based on total assets. Section 4.4 presents the result of the test. 4.2 The Relative Value Relevance of Earnings and Book Value in Malaysia and Singapore This study hypothesizes that the relative value relevance of book value is greater than earnings in Malaysia and Singapore. Table 4.3 presents the Wald chi-square test results for the relative value relevance of earnings and book value in Malaysia and Singapore. For Malaysian sample, Wald chi-square statistic is significant (p-value < 0.05), which indicates a significant difference in the explanatory power of earnings and book value in explaining market value of equity. Results from OLS regression test on equation (1a) and (1b) reveal that book value of equity (0.611) has higher r-square than earnings (0.476), thus indicates the greater value relevance of book value compared to earnings in explaining market value of equity in Malaysia. 10

11 Table 4.3: Wald Chi-square Test Result for the Relative Value Relevance of Earnings and Book Value MVE jt = β 0 + β 1 EARN jt + β 2 BVE jt + e jt (1) MVE jt = β 0 + β 1 EARN jt + e jt MVE jt = β 0 + β 1 BVE jt + e jt (1a) (1b) Country N R 2 Earnings Equation (1a) R 2 Book Value Equation (1b) 2 Test p-value Statistics Equation (1) Malaysia Singapore Similar results are found for Singaporean sample. There is a significant difference in the explanatory power of earnings and book value and the r-square of book value (0.488) is greater than earnings (0.338). The results indicate that book value is more value relevant than earnings in explaining market value of equity in Singapore. In summary, results from Wald chi-square and OLS regression tests indicate that the relative value relevance of book value is greater than earnings in Malaysia and Singapore at significant level (p-value < 0.05), which provide evidence confirming H1. Both countries have a similar relative value relevance of earnings and book value, consistent with the similarity of accounting systems between the two countries as highlighted by MacGregor et al. (1997). The findings also provide support for King and Langli (1998) and Black and White (2003), who found a greater relative value relevance of book value compared to earnings in Norway and Japan (lies between the Anglo-Saxon and the Continental system) respectively. 4.3 Sensitivity Analysis Previous literature has identified earnings sign, firm size and macroeconomic events as three factors that may affect the relative value relevance of earnings and book value. Therefore, this study examines the effect of the three factors on the relative value relevance of earnings and book value in Malaysia and Singapore Earnings Sign This study hypothesizes that earnings sign affects the relative value relevance of earnings and book value in Malaysia and Singapore. Table 4.4 presents the Wald chi-square test results for earnings sign effect on the relative value relevance of earnings and book value in Malaysia and Singapore. Results from Wald chi-square test reveal that there is a significant difference in the explanatory power of earnings and book value in explaining market value of equity in both countries when sample of each country is separated into positive and negative earnings samples (p-value < 0.05). In Malaysian sample, result shows that the explanatory power of book value (0.547) is greater than earnings (0.270 when earnings are negative. However, when earnings are 11

12 positive, earnings (0.693) have greater explanatory power than book value (0.630) in explaining market value of equity. Table 4.4: Wald Chi-square Test Results for Earnings Sign Effect on the Relative Value Relevance of Earnings and Book Value MVE jt = β 0 + β 1 EARN jt + β 2 BVE jt + e jt (1) MVE jt = β 0 + β 1 EARN jt + e jt MVE jt = β 0 + β 1 BVE jt + e jt (1a) (1b) Country Earnings Sign N R 2 Earnings Equation (1a) R 2 Book Value Equation (1b) 2 Test Statistics Equation (1) p-value Malaysia Positive earnings Negative earnings Singapore Positive earnings Negative earnings Similarly, in Singaporean sample, findings also indicate that book value (0.687) has greater explanatory power than earnings (0.614) in explaining market value of equity when earnings are negative. However, when earnings are positive, earnings (0.623) are more value relevant than book value (0.470) in explaining market value of equity. In summary, both Malaysia and Singapore have shown the effect of earnings sign on the relative value relevance of earnings and book value in explaining market value of equity, thus, provide support for H2. Findings indicate that firms with positive earnings have higher value relevance of earnings compared to book value. However, firms with negative earnings have lower value relevance of earnings compared to book value. The results of this study are consistent with findings by Hayn (1995), Collins et al. (1997) and Black and White (2003). Collins et al. (1997) and Black and White (2003) revealed that the relative value relevance of earnings is lower than book value when earnings are negative, while the relative value relevance of earnings is higher than book value when earnings are positive in the US. Similarly, Hayn (1995) suggested a lower value relevance of negative earnings as a result of its transitory nature. According to Hayn, negative earnings are not expected to persist into the future, thus, convey less information about future cash flow of a firm, which in turns lead to the lower value relevance of the negative earnings. Since transitory (negative) earnings are less value relevant, book value becomes relatively more value relevant compared to earnings when earnings are negative. 12

13 Apart from that, the findings on earnings sign effect also indicate that the result on the higher value relevance of book value as compared to earnings in Malaysia and Singapore (H1) is supported for the negative earnings sample only. H1 is not supported for firms reporting positive earnings Firm Size Results from descriptive statistics and t-test statistics have shown the difference in firm size between Malaysia and Singapore (Refer Table 4.1 and 4.2). Much prior literature also provided evidence on the differences in the value relevance of earnings and book value across firm size (Black, 2003; Black and White, 2003). Therefore, this study hypothesizes that firm size affects the relative value relevance of earnings and book value in Malaysia and Singapore. In order to test for the effect of firm size on the relative value relevance of earnings and book value of equity in Malaysia and Singapore, each country sample is classified into four quartiles, measured by total assets. The first quartile represents small firms, second and third quartile represent medium-size firm, and fourth quartile represents large firm, as measured by total assets. In classifying the sample into four quartiles, total asset values for each country sample is arranged in ascending order, then, the 25 th, 50 th and 75 th percentile values are identified. Then, the total assets values are classified into first quartile (1 st - 25 th percentile), second quartile (26 th - 50 th percentile), third quartile (51 st - 75 th percentile) and fourth quartile (76 th th percentile). Table 4.5: Wald Chi-square Test Results for Firm Size Effect on the Relative Value Relevance of Earnings and Book Value MVE jt = β 0 + β 1 EARN jt + β 2 BVE jt + e jt (1) MVE jt = β 0 + β 1 EARN jt + e jt MVE jt = β 0 + β 1 BVE jt + e jt (1a) (1b) Country Size Quartile N R 2 Earnings Equation (1a) R 2 Book Value Equation (1b) 2 Test Statistics Equation (1) p-value Malaysia 1st quartile nd quartile rd quartile th quartile Singapore 1st quartile nd quartile rd quartile th quartile

14 Table 4.5 presents the Wald chi-square test results for firm size effect on the relative value relevance of earnings and book value in Malaysia and Singapore. For Malaysian sample, there is a significant difference in the relative value relevance of earnings and book value in second and fourth quartile (p-value < 0.05). While earnings are more value relevant than book value for firms in the second quartile, book value is more value relevant than earnings for firms in the fourth quartile. However, there is no statistical difference in the value relevance of earnings and book value for firms in the first and third quartile (p-value > 0.05). For Singaporean sample, there is a significant difference in the relative value relevance of earnings and book value in the third and fourth quartile (p-value < 0.05). While earnings are more value relevant than book value for firms in the third quartile, book value is more value relevant than earnings for firms in the fourth quartile. The first two quartiles show no statistical difference in the value relevance of earnings and book value (p-value > 0.05). In summary, firm size affects the relative value relevance of earnings and book value in Malaysia and Singapore. Findings indicate that the relative value relevance of book value is greater than earnings for large firm (fourth quartile) in both countries. For medium size firms (second and third quartiles), the results show either higher value relevance of earnings or no significant difference in the value relevance of earnings and book value. However, there is no significant difference in the value relevance of earnings and book value for small firms (first quartile) in both countries under study. These findings provide support for H3. The findings also reveal that H1 is supported for large firm only in both countries under study. The results of the study provide support for Black (2003), who found a difference in the value relevance of accounting numbers across firm life cycle. These findings are consistent with Black and White (2003), who observed a significant difference in the relative value relevance of earnings and book value for different size quartiles in the US and Japan Macroeconomic Events This study also examines whether macroeconomic events, such as changes in accounting system and economic condition may affect the relative value relevance of earnings and book value in Malaysia and Singapore. The dynamic nature of accounting systems, whereby the systems are evolving and changing over time due to the promulgation of new accounting standards, and changes in economic condition such as recession and prosperity, could affect the relative value relevance of earnings and book value in a country. This study hypothesizes that macroeconomic events affect the relative value relevance of earnings and book value in Malaysia and Singapore. In order to test the effect of macroeconomic events on the value relevance of earnings and book value, tests are performed on each country sample by year. Year is used as a proxy for macroeconomic events. Table 4.6 presents the Wald chi-square test results for macroeconomic events effect on the relative value relevance of earnings and book value in Malaysia and Singapore. Table 4.6, Panel A provides test results for Malaysia. Findings from Wald chi-square tests reveal that there is no statistical difference in the relative value relevance of earnings and book value for all six years from 1998 to 2003 (p-value > 0.05). 14

15 Table 4.6: Wald Chi-square Test Results for Macroeconomic Events Effect on the Relative Value Relevance of Earnings and Book Value MVE jt = β 0 + β 1 EARN jt + β 2 BVE jt + e jt (1) MVE jt = β 0 + β 1 EARN jt + e jt MVE jt = β 0 + β 1 BVE jt + e jt (1a) (1b) Year N R 2 Earnings Equation (1a) R 2 Book Value Equation (1b) 2 Test Statistics Equation (1) p-value Panel A: Malaysia Panel B: Singapore Table 4.6, Panel B provides test results for Singapore. Results from Wald chi-square tests reveal that there is no significant difference in the relative value relevance of earnings and book value for all year from 2000 to 2003 (p-value > 0.05). While the relative value relevance of book value is greater than earnings in the year 1998, the relative value relevance of earnings is greater than book value in the year However, since the sample size for the two years ( ) is smaller than other years, the results may not representative for all samples for Singapore. Generally, there is no significant difference in the relative value relevance of earnings and book value for all the six year ( ) in Malaysia and Singapore, which implies no change in macroeconomic events during the year 1998 to

16 However, the results from individual year sample and pooled sample seem to be different. While the individual year sample reveal that there is no significant difference in the relative value relevance of earnings and book value in both countries for all years under study (pvalue > 0.05), the pooled sample ( ) indicate that there is a significant difference in the relative value relevance of the two variables (p-value < 0.05). The differences between the two results signify that macroeconomic events affect the relative value relevance of earnings and book value in Malaysia and Singapore. Therefore, H4 is supported and consistent with findings by Black and White (2003), who found that macroeconomic events affect the relative value relevance of earnings and book value in the US and Japan. In summary, the results of the study indicate that the relative value relevance of book value is greater than earnings in both countries under study, Malaysia and Singapore, which support H1. In sensitivity analysis, findings indicate that earnings sign, firm size and macroeconomic events affect the relative value relevance of earnings and book value in Malaysia and Singapore, thus provide support for H2, H3 and H4. The finding on the relative value relevance (H1) seems to be driven by firms reporting negative earnings and large firm only. However, the findings (H1) are not robust for firms reporting positive earnings, small firms and individual year sample (represent macroeconomic events), as earnings sign, firm size and macroeconomic events affect the relative value relevance of earnings and book value. 5.0 Conclusions This study examines the relative value relevance of earnings and book value in two Asian countries with similar accounting system, Malaysia and Singapore. Findings indicate that book value has greater value relevance than earnings in both countries, consistent with King and Langli (1998) and Black and White (2003), who found a greater value relevance of book value as compared to earnings in Norway and Japan (lies between the Anglo-Saxon and the Continental models) respectively. This study also found that earnings sign, firm size and macroeconomic events have effects on the relative value relevance of earnings and book value in the two countries. The results on the relative value relevance seem to be driven by firms reporting negative earnings and large firms only. However, the results are not supported for firms reporting positive earnings, small firms and individual year samples. Overall, this study provides evidence concerning the relative value relevance of earnings and book value, being two important summary measures of financial statements from Asian region. This study contributes to the limited literature on the value relevance of accounting numbers in Asian countries, particularly in Malaysia and Singapore. Apart from that, findings of this study may contribute to the international accounting literature, whereby the information on the relative usefulness of accounting numbers across countries may assist investors in making investment decisions at international level. Further, it may facilitate the effort of convergence of international accounting practice through formulation of a better quality of international accounting standard (IAS) that suit international needs. 6.0 References: Arce, M., & Mora, A. (2002). Empirical evidence of the effect of European accounting differences on the stock market valuation of earnings and book value. The European Accounting Review, 11,

17 Ball, R., & Brown, P. (1968). An empirical evaluation of accounting income numbers. Journal of Accounting Research, 6, Beaver, W. H., Clark, R., & Wright, W. (1979). The association between unsystematic security returns and the magnitude of earnings forecast errors. Journal of Accounting Research, 17, Biddle, G., Seow, G., & Siegel, A. F. (1995). Relative versus incremental information content. Contemporary Accounting Research, 12, Black, E. L. (2003). Usefulness of financial statement components in valuation: an examination of start-up and growth firms. Venture Capital, 5, Black, E. L., & White, J. J. (2003). An international comparison of income statement and balance sheet information: Germany, Japan and the US. European Accounting Review, 12, Bowen, R. E. (1981). Valuation of earnings components in the electric industry. The Accounting Review, 56, Collins, D. W., Maydew, E. L., & Weiss, I. S. (1997). Changes in the value relevance of earnings and book values over the past forty years. Journal of Accounting and Economics, 24, Deesomsak, R., Paudyal, K., & Pescetto, G. (2004). The determinants of capital structure: evidence from the Asia Pacific region. Journal of Multinational Financial Management, 14, Ely, K., & Waymire, G. (1999). Accounting standard-setting organizations and earnings relevance: longitudinal evidence from NYSE common stocks, Journal of Accounting Research, 37, Fan, J. P. H., & Wong, T. J. (2002). Corporate ownership structure and the informativeness of accounting earnings in East Asia. Journal of Accounting and Economics, 33, Francis, J., & Schipper, K. (1999). Have financial statement lost their relevance? Journal of Accounting Research, 37, Graham, R. C., & King, R. D. (2000). Accounting practices and the market valuation of accounting numbers: evidence from Indonesia, Korea, Malaysia, the Philippines, Taiwan and Thailand. The International Journal of Accounting, 35, Gray, S. J. (1988). Towards a theory of cultural influence on the development of accounting systems internationally. Abacus, 24, Gujarati, D. N. (1995). Basic Econometrics (3 rd ed.). New York: McGraw-Hill Inc Hayn, C. (1995). The information content of losses. Journal of Accounting and Economics, 20,

18 Ibrahim, M. K., Kamaruddin, K. A., & Mohamed Zain, M. (2004). Financial Reporting in Malaysia: Some Empirical Evidence, Malaysia: Utusan Publications. Ibrahim, M. K., Mohd Said, M., Abd Latif, R., & Abd Shukur, Z. (2004b). Value-relevance of accounting numbers: an empirical investigation of purchased goodwill. in Financial Reporting in Malaysia: Some Empirical Evidence, Malaysia: Utusan Publications. King, R. D., & Langli, J. C. (1998). Accounting diversity and firm valuation. International Journal of Accounting, 33, Lipe, R. C. (1986). The information contained in the components of earnings. Journal of Accounting Research, 24, MacGregor, A., Hossain, M., & Yap, K. (1997). Accounting in Malaysia & Singapore: culture s lack of consequences? in Accounting in the Asia-Pacific Region. Singapore: John Wiley & Sons, Nair, R. D., & Frank, W. G. (1980). The impact of disclosure and measurement practices on international accounting classifications. The Accounting Review, 55, Ohlson, J. A. (1995). Earnings, book values and dividend in equity valuation. Contemporary Accounting Research, 11, White, H. (1980). A heteroskedasticity-consistent covariance estimator and a direct test for heteroskedasticity. Econometrica,