The Effect of Corporate Governance on the Valuation of Book Value and Earnings during the Asian Financial Crisis

Transcription

1 The Effect of Corporate Governance on the Valuation of Book Value and Earnings during the Asian Financial Crisis Paqua Y. Davis-Friday* Department of Accountancy Universy of Notre Dame 386 Mendoza College of Business Notre Dame, IN Li Li Eng Oklahoma State Universy Tulsa 700 N. Greenwood Ave Tulsa, OK Chao-Shin Liu Department of Accountancy Universy of Notre Dame 387 Mendoza College of Business Notre Dame, IN December 2002 *Corresponding Author. The authors are grateful for helpful comments from participants at the 2002 American Accounting Association Annual Meeting. Paqua Davis-Friday and Chao-Shin Liu are grateful for research support from Ernst & Young, LLP and Delote & Touche, LLP, respectively.

2 The Effect of Corporate Governance on the Valuation of Book Value and Earnings during the Asian Financial Crisis Abstract This paper examines the value relevance of earnings and book values in four Asian countries, Indonesia, South Korea, Malaysia and Thailand, in the period surrounding the Asian financial crisis. Specifically, we examine the impact of the economic environment on the value relevance of book value and earnings, controlling for the qualy of financial reporting and corporate governance mechanisms. Our results indicate that the value relevance of earnings in Indonesia and Thailand was significantly reduced during the Asian financial crisis while the value relevance of book values increased. In Malaysia, the value relevance of both earnings and book value decreased during the crisis. In Korea, neher book value nor earnings was significantly impacted by the crisis. Our results indicate that the level of certain corporate governance mechanisms and financial reporting qualy have an impact on how the crisis affected the value relevance of earnings and book values. Specifically, the value relevance of book values decreases when the rule of law is lower, when the level of ownership concentration is higher, and when the qualy of aud reports is lower. Finally, our results indicate that Korea s tax-based accounting standards help to migate the effect of the financial crisis on the value relevance of book values, but not the value relevance of earnings.

3 I. INTRODUCTION This paper examines the effect of the Asian financial crisis on the value relevance of earnings and book values in four Asian countries, Indonesia, Korea, Malaysia and Thailand. In addion, we investigate the role of various corporate governance mechanisms and the qualy of financial reporting in determining the extent of the effects of the crisis on the valuation of equy book values and earnings. Barth, Beaver and Landsman (1998) find evidence that the relative value relevance of equy book value and net income shifts as the financial health of the firm decreases. We extend their analysis of bankrupt firms to a setting where the overall economy is in financial crisis to examine the interaction between country-specific corporate governance mechanisms and market participants assessment of financial accounting information. Bushman and Smh (2001, 240) suggest that cross-country designs represent a powerful setting for investigating issues relating to the economic effects of financial accounting information and corporate governance because of significant cross-country differences in both financial accounting regimes and economic performance. In addion, vast cross-country differences in the legal protection of investors rights, communication networks, and other instutional characteristics enable researchers to explore how the economic effects of financial accounting information vary wh other factors (Bushman and Smh 2001, 240). Bushman and Smh (2001, 241) argue that future research on the connection between the governance use and capal markets use of financial accounting information is important for developing a more complete understanding of the effects of financial accounting information on economic performance. While governance research typically focuses on a particular governance mechanism in isolation, a more complete understanding requires an explic recognion of the interactions across governance mechanisms (Bushman and Smh 2001, 286). Therefore, we 1

4 identify several country-specific instutional factors that influence the total economic effects of financial accounting information, as well as factors that influence the economic effects of financial accounting information through s governance role. Specifically, we investigate the effects of shareholder rights, credor rights, the rule of law, ownership concentration and the qualy of financial reporting on the relation between earnings and book values before and during the crisis. We base our research on a sample of 158 firms from Indonesia, 217 firms from Korea, 271 firms from Malaysia, and 389 firms from Thailand. We examine the value relevance of equy book values and earnings one year prior to the crisis (1996) and during the crisis (1997). The financial statement and stock price data are collected from Datastream International. We include measures of the levels of investor protection from LaPorta et al. (1998) and assessments of the qualy of financial reporting from Saudagaran and Diga (1997a). Our results indicate that the value relevance of earnings in Indonesia and Thailand was significantly reduced during the Asian financial crisis while the value relevance of book values increased. In Malaysia, the value relevance of both earnings and book value decreased during the crisis. In Korea, neher book value nor earnings was significantly impacted by the crisis. The effect of the crisis on the value relevance of book values is impacted by the level of both corporate governance mechanisms and the qualy of financial reporting. Specifically, the value relevance of book value appears to decrease when the rule of law is low, ownership concentration is high and the qualy of aud reports is low. The decrease in the value relevance of book value is migated when accounting standards are based on the tax code. However, none of these factors intervene to reduce the effect of the crisis on the value relevance of earnings. 2

5 The remainder of the paper is organized as follows. The next section discusses related prior research and accounting and corporate governance characteristics in the countries in our sample. The third section contains the hypotheses. The fourth section describes the research design and sample, and the fifth section discusses the results. The final section draws conclusions and summarizes the paper s contribution. II. BACKGROUND AND PRIOR RESEARCH Barth, Beaver, and Landsman (1998) examine the relative value relevance of equy book value and earnings in a set of firms that are facing bankruptcy. They argue that a firm s equy book value reflects liquidation value and net income reflects unrecognized net assets. As the financial health of a company decreases, liquidation value effects dominate unrecognized net asset valuation effects. Hence the incremental explanatory power of a firm s equy book value increases and the incremental explanatory power of net income decreases when the financial health of the firm deteriorates. The Asian financial crisis resulted in the deterioration of the financial health of affected companies. The incidence of negative earnings in the four Asian countries increased as a result of the financial crisis, and negative earnings are normally perceived to contain large transory components. We investigate whether the results from the Barth, Beaver and Landsman (1998) analysis of bankrupt firms extends to a setting where firms are operating in a financial crisis. Two studies examine the firm-specific effects of the Asian financial crisis in two countries severely affected by currency devaluations and stock market declines. Graham, King and Bailes (2000) document a decline in the value relevance of earnings in Thailand following the devaluation of the Thai Baht, but the incremental value relevance of book values increased. 3

6 Ho, Liu and Sohn (2001) find evidence that the value relevance of Korean earnings declines during the crisis and s importance is replaced by cash flows from operations as opposed to book value of equy. Their results hold for a sample of firms that report posive earnings throughout their sample period ( ), and also after controlling for the amount of foreign exchange translation gains and losses included in earnings and book value. These studies suggest that countries may be affected differentially by an economic crisis resulting from currency devaluations and stock market declines. 1 Prior research finds differences in the value relevance of accounting measures between countries that may be attributed to differences in accounting systems. Harris, Lang and Möeller (1994) find that the earnings coefficient in Germany is larger than that in the Uned States, consistent wh a more conservative measurement approach in Germany. In examining the relation between stock prices and accounting information in six Asian countries, Graham and King (1998) find systematic differences in the value relevance of accounting numbers as well as in the relative explanatory power of book value and earnings that appear to be the result of the degree of conservatism across the countries. Ali and Hwang (2000) find that value relevance of accounting data is lower for Continental model countries than for Brish-American model countries, and also when tax rules influence financial accounting measurements. In contrast, Ball, Robin and Wu (2000) find no differences in the property of timeliness of accounting income between countries that have code-law or common-law origins. 1 Mton (2002) reports that the countries in our sample experienced stock returns ranging from -52.1% in Thailand to- 79.0% in Malaysia and currency depreciation ranging from -34.5% in Korea to -78.0% in Indonesia during the crisis. 4

7 Domestic Financial Reporting Environments We now provide some background information regarding the accounting environment of the countries in this study. The development of domestic accounting standards in Indonesia, South Korea, Malaysia and Thailand has been influenced in part by their historical allegiances. Indonesia was a Dutch colony, and early Indonesian Accounting Principles (PAI) were based on Grady s (1965) Inventory of GAAP for Business Enterprises. In the late 1980s, Indonesia issued several statements of accounting principles that were based on U.S. GAAP. In September 1994, Indonesia adopted 21 International Accounting Standards (IAS) that were renamed Indonesian Financial Accounting Standards and made mandatory for all publicly listed companies (Saudagaran and Diga 2000, 8). The regulatory framework of Korean accounting and financial reporting has been influenced by U.S. laws and practices since the Second World War. Korean accounting standards also take account of IAS. However, they are strongly influenced by tax laws (Nobes and Parker 1995, 316). Accounting in Korea has a continental European influence modified by a U.S. influence (Nobes and Parker 1995, 319). Malaysia was a Brish colony and has historically looked to the U.K. in setting s domestic accounting standards. After the International Accounting Standards Commtee s (IASC) formation in 1973, Malaysia was one of the earliest countries in the Association of Southeast Asian Nations (ASEAN) to adopt IASC standards. In Malaysia, most IAS are adopted as Approved Accounting Standards (Saudagaran and Diga 2000, 8). Accounting standards in Thailand are based generally on U.S. GAAP but also incorporate concepts from the U.K. and Germany (especially the concept of prudence). More recently, 5

8 Thailand also gradually began to adopt IAS promulgated by the IASC. Presently, 17 of 23 Thai accounting standards are based on IAS (Saudagaran and Diga 2000, 8). Through the adoption of IAS, the ASEAN countries of Indonesia, Malaysia and Thailand are now moving towards a global harmonization of accounting policies. However, there exist differences in the perceived qualy of the regulatory environment between the ASEAN countries (Saudagaran and Diga 1997b). Further, there exist differences in the characteristics of financial reporting between the four countries in our paper (Saudagaran and Diga 1997a). The countries differ wh regard to disclosure levels, the reliabily of financial reports, the perceived qualy of reports, the level of enforcement of standards, and the comprehensiveness of aud reports. We examine the value relevance of book value and earnings in these four Asian countries that are moving towards harmonized accounting practices through the adoption of IAS, but yet still have differences in the enforcement of these standards. The relation between market value and book value and earnings is examined taking into account these country differences. Prior studies show that there are significant differences in the information content and timeliness of earnings between countries (e.g., Alford et al. 1993; Ball, Kothari and Robin 2000). However, a proper valuation model can greatly migate accounting diversy problems in international valuations (Frankel and Lee 1996). Frankel and Lee (1996) find that the Edwards-Bell-Ohlson (EBO) model is useful in explaining cross-sectional stock prices in different countries. The model that we use in our paper is based on Barth, Beaver and Landsman (1998), which is an adaptation of the EBO model. Corporate Governance The countries also differ wh respect to their corporate governance mechanisms. La Porta et al. (2000) classify Indonesia and South Korea as civil law countries, and Malaysia and 6

9 Thailand as common law countries. Minory shareholder rights are highest in Malaysia and lowest in Indonesia (Johnson et al. 2000). Anti-director rights are similar in Indonesia, Korea and Thailand and higher in Malaysia (Johnson et al. 2000; LaPorta et al. 1998). There is clearly a link between corporate governance and financial accounting (Sloan 2001). Bushman and Smh (2001) propose that research should explo cross-country differences in financial reporting and governance regimes. Hence, we examine whether differences in corporate governance impact the relative value relevance of the book value of equy and earnings of firms in these countries during the financial crisis. III. HYPOTHESES We hypothesize that the Asian crisis, which led to deterioration in Asian economies, will differentially impact the value relevance of book values and net income in Indonesia, Korea, Malaysia, and Thailand. We suggest that the impact of the crisis on firms in these countries will be a function of the corporate governance mechanisms and the qualy of financial reporting whin the country. Our first hypothesis regarding the overall impact of the crisis on the value relevance of book values and earnings relates to the results from Barth, Beaver and Landsman (1998) who document that the relative explanatory power of book value and net income is a function of financial health. That is, as financial health decreases, the incremental explanatory power of book value increases while the incremental explanatory power of net income decreases. Our first hypothesis, stated in alternative form is: H1a: The value relevance of equy book value increases while the value relevance of net income decreases as the economic environment deteriorates. 7

10 Bushman and Smh (2001) suggest that we may expect to see firms shift towards the use of more costly performance measures when accounting information has limed usefulness. For example, in countries where the accounting and legal systems provide relatively poor investor protection from managerial opportunism, there is a substution toward costly monoring by large shareholders. Therefore, we might expect corporate governance to affect the valuerelevance of accounting information. This leads to our second hypothesis: H2a: The extent of the effect of the Asian crisis on the value relevance of book value and earnings is a function of the level of shareholder and credor protections. Saudagaran and Diga (1997b) state that there exist differences in the perceived qualy of the regulatory environment among the ASEAN countries. We would expect to find valuation differences in countries that have higher qualy financial reporting and disclosure (i.e., IAS) and high qualy aud functions as demonstrated by the qualy of aud reports. H3a: The extent of the effect of the Asian crisis on the value relevance of book values and earnings is a function of the qualy of the country s financial reporting environment. IV. RESEARCH DESIGN AND METHODOLOGY Methodology We employ a methodology similar to that in Barth, Beaver and Landsman (1998). The market value of equy is regressed on book value of equy and net income. We include indicator variables and several corporate governance characteristics and measures of financial reporting qualy (i.e., accounting standards and aud reports) to examine the effect of the economic environment (the effect of the crisis) on the valuation of earnings and book values. 8

11 Inially, we estimate the following regression to ascertain the effect of the crisis on the earnings and book value coefficients: MVS = α * α1i + α 2EPS + α 3BVS + α 4I * EPS + α 5I BVS ε (1) where MVS is the market value of equy per share of company i on March 31 at year t+1; I is an indicator variable equal to zero in 1996 and one in 1997 (the year of the Asian financial crisis); EPS is earnings per share of company i over the period ending on December 31 in year t; and BVS is book value per share of company i at December 31 in year t. 2 In order to investigate the role of country-specific shareholder protections in the valuation of earnings and book values, we employ the corporate governance metrics from LaPorta et al. (1998). LaPorta et al. (1998) use multiple sources to assess the level of several dimensions of investors protections through various corporate governance mechanisms. Because is not clear how the absolute values of the variables relate to capal market valuations, we examine the effect of these variables in relative terms by categorizing the countries into high and low levels of protection for each of the variables. Table 1 reports the values of all of the unique corporate governance measures from LaPorta et al. (1998) for the four countries. Indonesia, Korea and Thailand have similar scores for shareholder rights while Malaysia has relatively higher levels of shareholder rights protections. Therefore, Indonesia, Korea and Thailand are classified as having lower levels of protection while Malaysia is classified in the higher protection level. The second measure that we investigate is the level of credor rights protection. The variable includes factors such as the existence of restrictions for going into reorganization and the level of legal reserve required as a percentage of capal. Here, we find that Indonesia and 2 We also estimate the regressions using the year end values for the market value of equy per share. The results are essentially the same as those reported. 9

12 Malaysia have relatively higher levels of protection and Korea and Thailand are classified together in the lower level. The third corporate governance measure attempts to capture the level of enforcement of investor protection laws. This variable includes factors like the efficiency of the judicial system, the risk of appropriation and the level of perceived corruption in the country. Indonesia and Thailand are considered to have lower levels of the rule of law and Korea and Malaysia are classified in the higher level. The final governance variable is the level of ownership concentration of public companies. This variable measures the ownership stake of the three largest shareholders in the ten largest (market capalization) publicly traded companies in the country. LaPorta et al. (1998) report that companies in countries wh poor investor protection usually have more concentrated ownership. The value of this variable is similar for Indonesia, Malaysia and Thailand (at or above the median) and lower for Korea. We also examine the role of the qualy of financial reporting (i.e., accounting standards and aud reports) in migating the effects of the economic crisis on the valuation of equy book values and earnings. We use two measures to capture the qualy of financial reporting. The first is the origin of the country s accounting standards. The objective of International Accounting Standards (IAS) is to generate financial information that is useful in determining a firm s value, i.e., transparent financial information (International Accounting Standards Commtee 1999). On the other hand, accounting standards that are primarily based on the tax code are not geared toward investors, but regulators. Therefore, to the extent countries accounting standards are more aligned wh IAS, they are more likely to communicate the underlying economic value of firms to investors. Three of our sample countries, Indonesia, Malaysia and Thailand are classified as IAS countries while Korea s accounting standards are classified as tax-based. 10

13 The second measure, from Saudagaran and Diga (1997a), captures the qualy of aud reports, which provide a means for outsiders to monor firms. Indonesia and Malaysia are classified as having relatively lower qualy aud reports and Korea and Thailand relatively higher level reports. All of the variables discussed here are summarized in table 1. [Insert table 1 about here] We estimate four regressions to determine whether the strength of a country s corporate governance mechanisms had any effect on the change in the value relevance of earnings and book values during the Asian financial crisis. The general model takes the following form: MVS = β + β I + β BVS β EPS β I * EPS β I * BVS 3 + β CG * I * EPS 7 + β CG * I * BVS β SIZE + ε 8 (2) where CG, which represents one of the four corporate governance measures (i.e., shareholder rights, credor rights, rule of law and ownership concentration), is an indicator variable taking on the value of one for countries wh a low level of investor protection and zero otherwise; and all other variables are as defined in equation (1). Our final analysis investigates whether the qualy of a country s financial reporting environment, based on the use of IAS or the tax code as a basis for accounting standards and the qualy of aud reports, had any impact on the change in the value relevance of earnings and book values during the crisis: MVS = λ + λ I + λ BVS λ EPS λ I * EPS λ I * BVS 3 + λ AD * I * EPS 7 + λ AD * I * BVS λ SIZE + ε 8 (3) where AD, which represents one of the two financial reporting measures (i.e., IAS vs. Tax and Aud Report Qualy), is an indicator variable taking on the value of one for countries wh accounting standards based on the tax code (Korea) and zero for those based on International 11

14 Accounting Standards (Indonesia, Malaysia and Thailand) or one for those wh relatively lower qualy aud reports and zero otherwise. We also include firm size (measured as the natural logarhm of the book value of total assets) as a control variable. Sample The sample consists of firms from Indonesia, Korea, Malaysia and Thailand for which all required data are available. The market value of equy, book value of equy and net income data in this paper are obtained from Datastream Research Service. Datastream is a database of economic, company and financial data. The sample period is 1996 and The sample consists of 1035 firms (across two years); 158 from Indonesia, 217 from Korea, 271 from Malaysia and 389 from Thailand. Table 2 provides descriptive statistics of the book value of equy, market value of equy and earnings per share of the sample firms across countries. The values are all in U.S. dollars per share. The Korean sample has the highest average book value of equy ($21.93) and market value of equy ($26.27) in The sample from Thailand has the second highest book value of equy ($3.67) but has a lower market value ($1.90). This would seem to indicate that the book value of the Thailand sample is overstated. The Korean sample has the highest earnings per share in 1996 ($0.58) and the Indonesian sample has the lowest ($0.08). In 1997, as the Asian economic crisis took hold of the economies of the countries included in the sample, all four countries suffered reduced profably and lower market and book values. Korea continues to have the highest average market value and book value of equy per share ($8.73 and $10.30, respectively), but the amounts are less than half their values in The mean and median differences are statistically significantly smaller in 1997 than in

15 Three of the four countries (Indonesia, Korea and Thailand) move from profabily to losses in The mean and median values of all of the variables, except for the average book value per share in Indonesia, are statistically significantly lower in 1997 than in [Insert table 2 about here] Table 3, panel A provides summary statistics of ROE (return on equy), E/P (earningsto-price ratio) and B/M (book-to-market ratio) across countries, and Panel B presents the p- values of Wilcoxon rank sum tests of differences in median ratios between countries. ROE is lowest in Korea in 1996 and This is consistent wh the fact that Korea allows asset revaluations and has relatively higher equy book values than the other countries. The median ROE of Korea is significantly lower than that in Indonesia, Malaysia and Thailand as indicated by the pairwise comparisons. The E/P ratio of Korea is the second lowest among the countries in Pairwise comparisons show that the E/P ratio of Korea is significantly lower than that in Indonesia in both years, Malaysia in 1997 and Thailand in The results for the B/M ratio are not consistent wh those of the other ratios. The median B/M ratio in Korea is the second highest in 1996, and the highest in This suggests a significantly less conservative balance sheet valuation in Korea, which allows asset revaluations, than in Indonesia, Malaysia and Thailand in [Insert table 3 about here] V. RESULTS In table 4, we present the results from examining the effect of the Asian financial crisis on the relative value relevance of book value of equy and net income. The increment to the intercept, α1, which captures the effect of the crisis year, is negative and significant for all 13

16 country estimations except Indonesia. In the case of Indonesia, the coefficient is posive, but is not significantly different than zero. The coefficients on earnings per share (α2) and book value per share (α3) are posive and significant in all cases, except for the book value of equy in the Indonesia estimation. It is possible that book values are not closely tied to market values in Indonesia so that the crisis did not have an impact on the valuation. We hypothesize that the incremental explanatory power of equy book value increases and the incremental explanatory power of net income decreases as the economic environment deteriorates. That is, we predict α4 to be negative and α5 to be posive. The results for the deterioration in the explanatory power of income hold for Indonesia (α4 = -4.44, t = -5.45), Malaysia (α4 = -4.04, t = -1.93) and Thailand (α4 = -4.52, t = ). This implies that the incremental explanatory power of income has declined for Indonesia, Malaysia and Thailand from 1996 to The results for the increase in the explanatory power of book value hold for Thailand (α5 = 0.06, t = 3.15) and marginally for Indonesia (α5 = 0.07, t = 1.68). Contrary to hypothesis H1a, the explanatory power of book value has also declined for Malaysia (α5 = -1.46, t = -2.24). Based on the data reported in table 1 regarding corporate governance mechanisms and financial reporting qualy, there is no reason to believe that the valuation of both book values and earnings in Malaysia should significantly decrease during the crisis. The only evidence we can find is in Mton (2002) which indicates that Malaysia had more negative stock returns during the crisis and larger market-to-book ratios prior to the crisis than the other countries included in our sample. This would suggest that the Malaysian stock market was relatively overvalued prior to the crisis. Both coefficients for Korea are negative, but insignificant. Ho, Liu and Sohn (2001) also document negative coefficients. However, both coefficients are significant in their estimation 14

17 (table 6, regression 7). A different sample composion and power of the tests may contribute to the difference in results. 3 Our evidence suggests that the economic crisis had no effect on the explanatory power of book value of equy and net income of firms in Korea. Overall, the results are consistent wh the findings by Graham, King and Bailes (2000) regarding the effect of the Asian financial crisis on Thailand. The results in Barth, Beaver and Landsman (1998) are supported in the case of Indonesia and Thailand in that the valuation of book values increases when firms are in financial crisis. Our results from Korea and Malaysia are inconsistent wh the Barth, Beaver and Landsman (1998). Therefore, the results from the examination of U.S. firms in financial distress do not appear to generalize to firms in all countries experiencing an economy-wide financial crisis. [Insert table 4 about here] Descriptive data on the degree to which the very existence and operation of control mechanisms vary across countries holds enormous potential. Such data not only would provide useful insights into how the mechanisms by which financial accounting potentially can serve a governance role vary around the world, but also would provide data for testing complementaries between high-qualy financial accounting regimes and high-qualy control mechanisms in promoting economic efficiency (Bushman and Smh 2001, 289). In table 5 we examine the impact of the economic environment, corporate governance, and the qualy of financial reporting on the value relevance of book value and earnings. The results from the shareholder rights estimation are presented in panel A. The estimation is pooled across the countries wh indicator and interaction variables capturing the effects of the economic 3 Ho, Liu and Sohn (2001) use data for 429 sample firms collected directly from the Korean Stock Exchange over four sample years. The larger sample size increases the power of their test and makes easier to find significant results. 15

18 crisis and the interactive effect of the governance variable given that the financial crisis has occurred. 4 The results indicate that the crisis reduced market values overall (β1 = -2.25, t = -4.73) and that larger firms exhibed lower returns (β8 = -0.45, t = -3.28). However, the valuation of book values and earnings remains unchanged during the crisis. While the coefficients on the interaction between the crisis and book values and earnings are negative, they are insignificant (β3 = -0.54, t = and β6 = -0.12, t = -0.04). Finally, appears that the levels of shareholder rights do not have an impact on the valuations of book values and earnings given that the crisis occurred (β4 and β7 are insignificant). Panel B of table 5 reports the results for credor rights. The pattern of results are similar except that the coefficient on the valuation of book values during the crisis is marginally significantly lower (γ3 = -0.56, t = -1.70) at the nine percent level two-tailed. Panels C and D of table 5 report the results regarding the rule of law and level of ownership concentration. In both estimations we find that market values are lower during the crisis, and size is negatively related to market values. The book values are valued less for countries wh lower levels of the rule of law or higher levels of concentrated ownership (δ4 = -0.47, t = and θ4 = -0.47, t = -4.32). Although earnings are valued lower (δ6 = -0.91, t = and θ6 = -0.90, t = -3.60), corporate governance variables do not have a significant impact. Overall, our corporate governance results indicate that the levels of shareholder and credor rights do not have a significant impact on how the crisis affects the value relevance of earnings and book values. However, in countries where book values are negatively impacted by 4 We are very grateful to an anonymous reviewer for suggesting this estimation and explanation. 16

19 the crisis, the decrease in the value relevance of book values is greater when the level of the rule of law is lower and the level of ownership concentration is higher. Panel E of table 5 reports the results of our analysis of the effect of accounting standards on the impact of the crisis. Our results indicate that market values and the valuations of book values (λ3 = -0.53, t = -4.89) are lower during the crisis, but not earnings (λ6 = -0.40, t = -0.50). Addionally, having tax-based accounting standards helps to migate the effect of the financial crisis on the value relevance of book values, but not on earnings. Ho, Liu and Sohn (2001, 207) report that Korea allows asset revaluations. This practice results in book values of net assets that are closer to market value and helps the book value to fare better during the crisis. Specifically, Korea s market-adjusted book values fare better during the crisis than those in Indonesia, Malaysia and Thailand (λ4 = 0.47, t = 4.32). Finally, in the last estimation including measures of the qualy of aud reports, the results are similar except the incremental value relevance of earnings is lower during the crisis (φ6 = -0.86, t = -3.44). As in the accounting standards estimation, we find that aud report qualy does not affect the valuation of earnings (φ7 = 1.07, t = 0.42), but lower levels of aud report qualy are weakly associated wh lower valuation levels for book values (φ4 = -0.46, t = -1.42). 5,6 Taken together, appears that accounting standards and the qualy of aud reports interact wh the effects of the crisis on book values, but not on earnings. 5 We also conduct analyses to investigate whether the changes in the value relevance of book value and earnings interact wh firm size. In untabulated results we find that the size variable is insignificant and all of the other results remain the same. 6 We also employ the Barth, Beaver and Landsman approach (1998, 25) to identify 15 industry groups and rerun the analyses reported in table 5. Due to missing industry information in the Datastream database, the analyses only include 1074 observations. However, the inferences remain the same. Addionally, when we exclude the 103 firms from regulated industries (i.e., financial instutions, insurance and real estate companies), the inferences remain the same. 17

20 [Insert table 5 about here] VI. SUMMARY AND CONCLUSION The purpose of this is research is to assess the changes in the value relevance of earnings and book values in the four countries most affected by the Asian financial crisis, Indonesia, Korea, Malaysia, and Thailand. We investigate the role of the countries corporate governance mechanisms and financial reporting qualy in determining the extent of the effect of the crisis on the value relevance of earnings and book values. Consistent wh prior research, we find that the value relevance of earnings and book values changes during the Asian financial crisis in Indonesia, Malaysia and Thailand. Our findings provide partial support for our hypotheses that the extent of the change in value relevance is related to the countries corporate governance mechanisms and financial reporting qualy. Overall, we provide evidence that the extent of shifts in the value-relevance of book values is not just a function of firm-specific financial health and macroeconomic environment. We demonstrate that accounting standards, aud report qualy, rule of law and ownership concentration also play a role in determining the relation between stock prices and book values. 18

21 REFERENCES Alford, A., J. Jones, R. Leftwich, and M. Zmijewski The relative informativeness of accounting disclosures in different countries. Journal of Accounting Research 31 (Supplement): Ali, A. and L. Hwang Country-specific factors related to financial reporting and the value relevance of accounting data. Journal of Accounting Research 38 (Spring): Ball, R., S.P. Kothari, and A. Robin The effect of international instutional factors on properties of accounting earnings. Journal of Accounting and Economics 29: Ball, R., A. Robin, and J.S. Wu Incentives versus standards: Properties of accounting income in four East Asian countries, and implications for acceptance of IAS. Working paper, Universy of Rochester and Rochester Instute of Technology. Barth, M., W. Beaver, and W. Landsman Relative valuation roles of equy book value and net income as a function of financial health. Journal of Accounting and Economics 25 (1): Bushman, R. and A. Smh Financial accounting information and corporate governance. Journal of Accounting and Economics 32: Frankel, R. and C. Lee Accounting diversy and international valuation. Working paper, Universy of Michigan and Cornell Universy. Grady, P Inventory of Generally Accepted Accounting Principles for Business Enterprises. Accounting Research Study No. 7. New York: AICPA. Graham, R.C. and R.D. King 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(4): Graham, R., R. King and J. Bailes The value relevance of accounting information during a financial crisis: Thailand and the 1997 decline in the value of the baht. Journal of International Financial Management and Accounting 11(2): Harris, T., M. Lang and H.P. Möller The value relevance of German accounting measures: An empirical analysis. Journal of Accounting Research 32(2): Ho, L-C., C-S Liu, and P. Sohn The value relevance of accounting information around the 1997 Asian financial crisis - The case of South Korea. Asia Pacific Journal of Accounting and Economics (December): International Accounting Standards Commtee International Accounting Standards. Johnson, S., P. Boone, A. Breach and E. Friedman Corporate governance in the Asian financial crisis. Journal of Financial Economics 58:

22 LaPorta, R., F. Lopez-de-Silanes, A. Shleifer and R. Vishny Law and finance. Journal of Polical Economy 106: LaPorta, R., F. Lopez-de-Silanes, A. Shleifer and R. Vishny Agency problems and dividend policies around the world. Journal of Finance 55: Nobes, C. and R. Parker Comparative International Accounting. Prentice-Hall: NY. Saudagaran, S.M., and J.G. Diga. 1997a. Financial reporting in emerging capal markets: Characteristics and policy issues. Accounting Horizons 11(2): Saudagaran, S.M., and J.G. Diga. 1997b. Accounting regulation in ASEAN: A choice between the global and regional paradigms of harmonization. Journal of International Financial Management and Accounting 8(1): Saudagaran, S.M., and J.G. Diga The instutional environment of financial reporting regulation in ASEAN. The International Journal of Accounting 35(1): Sloan, R Financial accounting and corporate governance: A discussion. Journal of Accounting and Economics 32:

23 Table 1 Corporate Governance and Financial Reporting Measures Shareholder Rights Credor Rights Compose Scores Rule of Law (Enforcement) Mean Ownership (3 largest shareholders) Accounting Standards (IAS vs. Tax) Aud Report Qualy Indonesia Korea Malaysia Thailand Notes: The Shareholder Rights variable is from LaPorta et al. (1998, 1130). It is the compose score from the following six discrete measures which take on a value of one if the protection is in the Law and zero otherwise (1) one shareone vote; (2) shareholders are allowed to mail their proxy vote to the firm; (3) shareholders are not required to depos their shares prior to the general shareholders' meeting; (4) cumulative voting or proportional representation of minories on the board of directors is allowed; (5) an oppressed minories mechanism is in place; (6) there is a preemptive right to new issues. Addionally, the variable includes two continuous measures: (1) the minimum percentage of share capal that entles a shareholder to call for an extraordinary shareholders' meeting; plus (2) an anti-director rights index. The total score ranges from zero to twelve (most rights). The Credor Rights variable is from LaPorta et al. (1998, 1136). It is the compose score from the following four discrete measures which take on a value of one if the protection is in the Law and zero otherwise: (1) no automatic stay on assets; (2) secured credors are paid first; (3) restrictions for going into reorganization; (4) management does not stay in reorganization. Addionally, there is one continuous measure: the level of legal reserve required as a percentage of capal. The Rule of Law variable is from LaPorta et al. (1998, 1142). It is the compose score from the following five measures which take on values ranging from one to ten: (1) efficiency of the judicial system; (2) an assessment of the law and order tradion in the country produced by the country risk rating agency International Country Risk; (3) corruption; (4) risk of appropriation; and (5) risk of contract repudiation. The Ownership variable is from LaPorta et al. (1998). It consists of the combined ownership stake of the three largest shareholders in the ten largest (market capalization) nonfinancial, domestic, totally private, publicly traded companies in each country in the sample. The Accounting Standards variable is from Saudagaran and Diga (1997a). It is an indicator variable taking on the value of one for countries wh accounting standards based on the tax code (Korea) and zero for those based on International Accounting Standards (Indonesia, Malaysia and Thailand). The Aud Report Qualy variable is from Saudagaran and Diga (1997a). It indicates that Korea and Thailand have aud reports of above average qualy and those from Indonesia and Malaysia are average or below. 21

24 Table 2 Descriptive Statistics Paired T-Test Variable Country N Mean Std Dev Median Mean Std Dev Median T (p-value) MVS Indonesia Korea Malaysia Thailand BVS Indonesia (0.59) Korea Malaysia Thailand EPS Indonesia Korea Malaysia Thailand Variable Definions: MVS is market value of equy per share; BVS is book value per share; EPS is earnings per share. Wilcoxon Rank Sum Test Z (p-value)

25 Table 3 Distribution of Financial Ratios Panel A: Distribution of Median ROE, E/P, and B/M across Countries Indonesia Korea Malaysia Thailand Observations ROE E/P B/M Panel B: P-values from Wilcoxon Rank Sum of Tests of Median Differences in Financial Ratios Between Countries Indonesia - Korea Indonesia - Malaysia Indonesia - Thailand Korea - Malaysia Korea - Thailand Malaysia - Thailand ROE E/P B/M Variable Definions: ROE = Return on Equy (Net Income/Book Value of Equy); E/P = Earnings to Price ratio (Earnings per share/market Value of Equy per share); B/M = Book to Market ratio (Book Value per share/ Market Value of Equy per share). 23

26 Table 4 Pooled regression of market value of equy on book value of equy and net income, using fixed-effects estimation wh fixed year effects MVS = α + α I + α EPS + α BVS + α I I * BVS + ε * EPS + α 5 α0 α1 α2 α3 α4 α5 Adj. R 2 Obs. Indonesia Coefficient (t-statistic) 0.22 (3.81) 0.08 (1.06) 9.25 (17.31) 0.01 (0.41) (-5.45) 0.07 (1.68) p-value Korea Coefficient (t-statistic) (9.14) (-4.47) 0.99 (2.96) 0.49 (9.09) (-1.09) (-0.71) p-value Malaysia Coefficient (t-statistic) 1.20 (3.57) (-2.13) 3.96 (2.07) 2.59 (5.73) (-1.93) (-2.24) p-value Thailand Coefficient (t-statistic) 1.09 (11.88) (-3.36) 5.13 (21.90) 0.02 (2.10) (-16.46) 0.06 (3.15) p-value Pooled Coefficient (t-statistic) 3.32 (10.47) (-5.20) 1.03 (5.94) 0.60 (29.11) (-3.35) (-2.45) p-value Variable Definions: MVS is market value of equy per share of company i at time t; I is an indicator variable equal to zero in 1996 and one in 1997; BVS is book value per share of company i at time t; EPS is earnings per share of company i over the period ending at time t; 24

27 Table 5 Pooled regression of market value of equy on book value of equy and net income, controlling for corporate governance and accounting standards Panel A: Shareholder Rights MVS = β + β I + β BVS β EPS β I * EPS β I * BVS 3 + β SR * I * EPS 7 + β SR * I * BVS β SIZE + ε 8 Coefficient (t-statistic) β 0 β 1 β 2 β 3 β 4 β 5 β 6 β 7 β 8 Adj (5.19) (-4.73) 0.61 (29.30) (-0.55) 0.44 (0.45) 1.00 (5.76) (-0.04) (-0.26) (-3.28) R 2 Obs p-value Panel B: Credor Rights MVS = γ + γ I + γ BVS γ EPS γ I * EPS γ I * BVS 3 + γ CR * I * EPS 7 + γ CR * I * BVS γ SIZE + ε 8 Coefficient (t-statistic) γ 0 γ 1 γ 2 γ 3 γ 4 γ 5 γ 6 γ 7 γ 8 Adj (5.13) (-4.86) 0.61 (29.30) (-1.70) 0.46 (1.42) 1.00 (5.77) 0.21 (0.08) (-0.42) (-3.21) R 2 Obs p-value Panel C: Rule of Law MVS = δ + δ I + δ BVS δ EPS δ I * EPS δ I * BVS 3 + δ RL * I * EPS 7 + δ RL * I * BVS δ SIZE + ε 8 Coefficient (t-statistic) δ 0 δ 1 δ 2 δ 3 δ 4 δ 5 δ 6 δ 7 δ 8 Adj (4.70) (-4.93) 0.61 (29.34) (-1.13) (-4.26) 1.01 (5.81) (-3.62) 0.46 (0.54) (-2.79) R 2 Obs p-value

28 Table 5 (Continued) Pooled regression of market value of equy on book value of equy and net income, controlling for corporate governance and accounting standards Panel D: Ownership Concentration MVS = θ + θ I + θ BVS θ EPS θ I * EPS θ I * BVS 3 + θ CO * I * EPS 7 + θ CO * I * BVS θ SIZE + ε θ 8 Coefficient (t-statistic) θ 0 θ 1 θ 2 θ 3 θ 4 θ 5 θ 6 θ 7 θ 8 Adj (4.69) (-4.73) 0.61 (29.34) (-1.18) (-4.32) 1.01 (5.81) (-3.60) 0.50 (0.62) (-2.78) R 2 Obs p-value Panel E: Accounting Standards MVS = λ + λ I + λ BVS λ EPS λ I * EPS λ I * BVS 3 + λ AC * I * EPS 7 + λ AC * I * BVS λ SIZE + ε 8 Coefficient (t-statistic) λ 0 λ 1 λ 2 λ 3 λ 4 λ 5 λ 6 λ 7 λ 8 Adj (4.69) (-4.73) 0.61 (29.34) (-4.89) 0.47 (4.32) 1.01 (5.81) (-0.50) (-0.62) (-2.78) R 2 Obs p-value Panel F: Aud Reports MVS = φ + φ I + φ BVS φ EPS φ I * EPS φ I * BVS 3 + φ AR * I * EPS 7 + φ AR * I * BVS φ SIZE + ε 8 Coefficient (t-statistic) φ 0 φ 1 φ 2 φ 3 φ 4 φ 5 φ 6 φ 7 φ 8 Adj (5.13) (-4.86) 0.61 (29.30) (2.27) (-1.42) 1.00 (5.77) (-3.44) 1.07 (0.42) (-3.21) R 2 Obs p-value

29 Table 5 (Continued) Variable Definions: MVS is market value of equy per share of company i at time t; I is an indicator variable equal to zero in 1996 and one in 1997; BVS is book value per share of company i at time t; EPS is earnings per share of company i over the period ending at time t; SR is an indicator variable taking on the value of one for countries wh a low level of minory shareholders rights (Indonesia, Korea and Thailand) and zero otherwise; CR is an indicator variable taking on the value of one for countries wh a low level of credor rights (Korea and Thailand) and zero otherwise; RL is an indicator variable taking on the value of one for countries wh the lowest rules of law and enforcement (Indonesia and Thailand) and zero otherwise; CO is an indicator variable taking on the value of one for countries wh a high level of concentrated ownership (Indonesia, Malaysia and Thailand) and zero otherwise; AC is an indicator variable taking on the value of one for countries wh tax-based accounting standards (Korea) and zero otherwise; AR is an indicator variable taking on the value of one for countries wh low qualy aud reports (Indonesia and Malaysia) and zero otherwise. Size is the natural logarhm of the book value of total assets. More information about the measurement of the corporate governance and accounting variables is provided in table 1. 27