Disappearing Dividends in the Thai Capital Market: Changing Firm Characteristics or Lower Propensity to Pay

Transcription

1 Journal of Economic and Social Policy Volume 1 Issue 1 Enterprising Finance Article Disappearing Dividends in the Thai Capital Market: Changing Firm Characteristics or Lower Propensity to Pay Malinee Ronapat Southern Cross University Michael Evans Southern Cross University Follow this and additional works at: Recommended Citation Ronapat, Malinee and Evans, Michael (2) "Disappearing Dividends in the Thai Capital Market: Changing Firm Characteristics or Lower Propensity to Pay," Journal of Economic and Social Policy: Vol 1 : Iss 1, Article 7 Available at: epublications@scu is an electronic repository administered by Southern Cross University Library Its goal is to capture and preserve the intellectual output of Southern Cross University authors and researchers, and to increase visibility and impact through open access to researchers around the world For further information please contact epubs@scueduau

2 Disappearing Dividends in the Thai Capital Market: Changing Firm Characteristics or Lower Propensity to Pay This article is available in Journal of Economic and Social Policy:

3 Ronapat and Evans: Disappearing Dividends in the Thai Capital Market: Changing Firm Vol 2, no 7, February, pp Disappearing Dividends in the Thai Capital Market: Changing Firm Characteristics or Lower Propensity to Pay Malinee Ronapat Doctor of Business Administration Graduate College of Management Southern Cross University and Associate Professor Michael Evans Director Graduate College of Management Southern Cross University Abstract The number of firms in the Stock Exchange of Thailand that pay cash dividends has declined from 42 percent in 1 to 44 percent in 22 This decline is partly due to the changing characteristics of listed firms, which are increasingly becoming non-payers with high growth, small size and high earnings, but no dividends This paper investigates the correlation between dividend yield, different market ratios and the characteristics described by Fama and French (21), in the Thai capital market Key Words dividends, capital market, investment Introduction Dividend, as it is the only cash flow to investors from a firm, has long been an interesting topic for discussion and research Dividend is widely accepted as a tool for analysing and forecasting a firm's performance, characteristics and values Investors have different preferences on the type of firms they want to invest in Like Fama and French (21), this paper looked at the dividend payment activity of listed firms and classified them according to their profitability, investment opportunities and size, and identified each group's characteristics using summary statistics and logit regression We used these characteristics to predict the dividend payment activity of listed firms in the Thai capital market, and to show changes in the characteristics of listed firms and their propensity to pay dividends over time The findings suggest that the characteristics of firms that paid dividends changed slightly before the Asian economic crisis in 17, a lot during the crisis and returned to normal after the crisis Interestingly, listed firms had a lower propensity to pay dividends even before the crisis, but this largely declined after the crisis The findings of this research confirmed the results of Fama and French (21) with regards to the characteristics and propensity to pay dividends of listed firms in a Published by epublications@scu, 2 1

4 Journal of Economic and Social Policy, Vol 1, Iss 1 [2], Art 7 developing country (Thailand) This study can be used as an early warning of the changes in characteristics and propensity to pay dividends of listed firms, as well as a mechanism to categorise them The Global Phenomenon of Disappearing Dividends In the US, dividends are generally taxed at a higher rate than capital gains Therefore, firms presume that dividends are less valuable than capital gains (Fama & French 21) Consequently, firms that pay dividends are disadvantaged because they experience higher costs of equity than firms that do not pay dividends, or finance their activities with equity (Fama & French 21) Even though dividends are a high-cost method of obtaining finance, some firms continue to pay dividends Some firms claim that they need to compensate and be honest to their loyal stockholders (Petty et al 2) Fama and French (21) investigated the issue of disappearing dividends in the US This study focused on the period from 17 to 1 and found that the percentage of firms paying dividends in the US declined sharply after 17 In 17, 2 percent of listed firms paid dividends and this figure rose to percent in 17 However, by 1 only 2 percent of US listed firms paid dividends The percentage of firms paying dividends decreased by more than percent between 17 and 1 Fama and French (21) provided two explanations for this outcome Firstly, the characteristics of newly listed firms changed after 17 At present, many listed firms experience characteristics such as low profitability and strong growth, and have never made dividend payments (Fama & French 21) Secondly, Fama and French (21) studied the characteristics of listed firms over time and found that fewer firms appear to pay dividends than in the past Fama and French (21) entitled this trend 'the declining propensity to pay dividends' Fama and French (21) used summary statistics (or descriptive statistics), such as frequencies and means, to categorise the firms listed (excluding financials and utilities firms) on three US stock markets (AMEX, NASDAQ and NYSE) Three characteristics were found to affect the level of dividend payments, namely, profitability, investment opportunities and a firm's size Logit regression analysis was used to confirm the significance of these factors Earlier researchers such as Tobin (1), Gordon (12, 1), Higgins (172), Fazzari, Hubbard and Petersen (1), Fama and French (17) and Aivazian, Booth and Cleary (2) focused on the relationship between internally generated funds, profitability and investment Ratios were used to assess the level of profitability, investment, growth, size or the liquidity of firms to standardise the measurement of these variables Ratios which are often used by researchers include the debt ratio (total liabilities divided by total assets), the M/B ratio (market to book value), ROA (net income divided by total assets), ROE (net income divided by shareholders' equity), total assets, and the growth in assets (Fazzari, Hubbard, and Petersen 1; DeAngelo and DeAngelo 1; Fama and French 1, 17; James, Fama and French 2, Aivazian, Booth and Cleary 2) 2

5 Ronapat and Evans: Disappearing Dividends in the Thai Capital Market: Changing Firm In relation to a firm's profitability, Fama and French (21) used E t /A t (the ratio of aggregate earnings before interest to aggregate assets) and Y t /BE t (aggregate common stock earnings over aggregate book equity) After calculating the values and the average for the two ratios, the average values of ratios for payers of dividends, nonpayers and former payers were calculated using statistical software (Fama & French 21) Fama and French (21) found that the profitability of the firms (as measured by E t /A t and Y t /BE t ratios) varies with the type of firm (dividends payers, former payers and non-payers) Firms that distribute dividends were found to have higher profitability than non-payers (Fama & French 21) As mentioned earlier, the level of investment opportunities was seen as a factor which influences the dividend policies of firms which are listed in the US Fama and French (21) claimed that investment opportunities vary across dividend groups (payers, non-payers: former-payers and never paid firms) In this way, the investment opportunities of firms listed on the Thai stock market will be measured by using V t /A t (assets growth rates, which is the ratio of the aggregate market value to the aggregate book value of assets), dat/at (the rate of change of total assets, or the firms' assets growth rate) and RD t /A t (research and development, or R&D) expenditures following Fama and French's (21) research paper on 'disappearing dividends' Summary statistics (descriptive statistics) and logit regression will be used to analyse the significance of the investment opportunities of firms which are listed in Thailand Fama and French (21) concluded that firms which have not paid dividends also have better growth opportunities than payers and former payers Firms that have never paid dividends also grow faster than those which have not paid dividends Former payers of dividends have poor investment opportunities, because their internal source of finance has been depleted by the payment of dividends (Fama & French 21) The average assets size (A t ) and debt ratios (total liabilities to total assets) of firms listed on the US stock markets has influenced firm size (Fama & French 21) Summary statistics were used to calculate the values of the average assets size ratio and logit regression was applied to confirm this characteristic Fama and French (21) found that dividends payers are much larger than former payers and nonpayers, in the case of US listed firms The propensity to pay dividends is measured by applying the average coefficient to the value of each firm's characteristic to form a formula on profitability, investment opportunities and size (Fama & French 21) A base period (or years) was set and the ratios of a variety of groups of firms were compared in relation to their propensity to pay dividends in different time periods This process of analysis and comparison was undertaken with the aid of logit regression (Fama & French 21) Fama and French (21) stated that listed firms have a lower propensity to pay dividends and this resulted in lower payments and a decline in the number of payers To measure change in the characteristics of listed firms, the sample was divided into four groups: (1) dividends payers, (2) non-payers () former-payers, and (4) never paid (Fama & French 21) The three important characteristics of listed firms, namely, (1) profitability, (2) investment opportunities, and () size, were measured and a logit regression analysis was used to test for any change in firm characteristics, Published by epublications@scu, 2

6 Journal of Economic and Social Policy, Vol 1, Iss 1 [2], Art 7 or divergence between the expected number of payers and the actual number of payers (Fama & French 21) Fama and French (21) found that listed firms have changed their characteristics over time and this has resulted in lower dividend payment in the US stock markets Fama and French (21) concluded that firms which have never paid dividends are more profitable and experience stronger growth opportunities than former payers Consequently, it can be argued that low earnings, strong investment and growth opportunities, and small size are characteristics of firms which have never paid dividends (Fama & French 21) In addition, former payers tend to be distressed firms (Fama & French 21) Finally, payers of dividends are large firms with high earnings and low investment opportunities (Fama & French 21) Fama and French (21) concluded that the characteristics and their propensity to pay dividends of listed firms have changed The change has resulted in a disappearance of dividends and a reduction in the number of payers (Fama & French 21) A study by Benito and Young (21) supported the findings of Fama and French (21) in relation to the decline in the payment of dividends This study used secondary data from the UK stock market (FTSE) which was collected during the period 174 to 1 It was found that the proportion of firms which do not pay dividends rose from 14 percent in 174 to 22 percent in 1 (Benito and Young 21) The study also found that the proportion of firms that were not paying dividends tended to increase during recessions Benito and Young (21) suggested that the increase in the proportion of firms omitting dividends was largely due to an increase in the proportion of firms that have never paid dividends In addition, firms which paid high dividends more than doubled their volume of payments to shareholders between 177 and 1 (Benito and Young 21) The study also found that low cash flows, high leverage and greater investment opportunities are associated with an increase in the propensity to omit dividends in the UK (Benito and Young 21) This finding is consistent with the conclusions of Fama and French (21) Benito and Young (21) suggested that non-payers tend to be small firms with strong investment opportunities This indicates that firms which do not pay dividends have positive prospects for growth and are not distressed The decision to avoid the payment of dividends was regarded as evidence of financial distress In short, Benito and Young (21) concluded that the characteristics of firms and the propensity to pay dividends have a large impact on the payment of dividends in the UK (Benito and Young 21) The two previous studies concentrated on the payment of dividends by firms listed on stock markets of developed countries Aivazian, Booth and Cleary (2) investigated eight developing countries, including South Korea, India, Malaysia, Thailand, Zimbabwe, Jordan, Pakistan and Turkey This study used secondary data collected between 11 and 1, and descriptive statistics and t-statistics to compare these emerging markets with the US The findings indicate that firm profitability, growth and investment opportunities affect the payment of dividends of emerging markets in a similar fashion to the US (Aivazian, Booth and Cleary 2) They added that a high ROE (which represent a firm's profitability) results in the payment of high 4

7 Ronapat and Evans: Disappearing Dividends in the Thai Capital Market: Changing Firm dividends, while high debt ratios are associated with a low payment of dividends However, the 'sensitivity to these factors varies across countries and emerging market firms tend to pay higher dividends than the US firms and are more sensitive to some of the variables' (Aivazian, Booth and Cleary 2, p 7) These researchers, therefore, concluded that firms in emerging markets respond to the same variables as those in the US (Aivazian, Booth and Cleary 2) However, the study did not shed light on why the payment of dividends is in decline (Aivazian, Booth and Cleary 2) Many researchers have attempted to explain the phenomenon of disappearing dividends Yardeni (2) stated that the payment of dividends was important in the US prior to 12, but has become less important over the past 2 years (Yardeni 2) The researcher claimed that 'if the US Congress eliminates the double taxation of dividends, then dividends should matter more again' (Yardeni 2, p 14) Mr Allan Greenspan (present Chairman of the Federal Reserve Bank, US) observed that shareholders' obsession with earnings is a relatively new phenomenon: Prior to the past several decades, earnings forecasts were not nearly so important a factor in assessing the value of corporations In fact, I do not recall price-to-earnings ratios as a prominent statistic in the 1s Instead, investors tended to value stocks on the basis of their dividend yields (Yardeni 2, p 1) The phenomenon of disappearing dividends remains a puzzle (Black 17; Fama and French 21; Benito & Young 21; Aivazian, Booth & Cleary 2) Fama and French's (21) findings have added to the knowledge of dividends, although more research is needed to explain the pattern of dividends, payers' characteristics and the propensity to pay dividends in developing capital markets such as Thailand This leads to an investigation of the relationship between dividends and a given set of characteristics of listed firms and other important market factors which may explain why firms pay, or avoid paying dividends As stated earlier, the analysis of Fama and French (21) focused on firms which are listed in the US, and decline in the number of payers and the propensity to pay dividends was evident However, more analysis is needed on whether this is occurring in foreign markets (Ferris, Sen & Yui 22), particularly when the characteristics of firms and investors differ from those of the US The issue of whether decline in the payment of dividends is an 'international phenomenon' needs more research (Ferris, Sen & Yui 22) This study will respond to this need by focusing on the emerging capital markets such as Thailand Time Trends in Cash Dividends Dividend receipts are taxed at a higher rate than capital gains, making dividends less valuable in the eyes of investors (Fama & French 21) This is true in the case of Thailand as well Dividend receipt is taxed at 1 percent with a tax credit of /7 at the end of the year Thus, the dividend receipt's real tax rate is at 71 percent However, capital gain is currently tax exempt (SEC 22) Firms paying dividends are presumed to have a competitive disadvantage (such as higher tax rate and lower investment opportunities) as they have a higher cost of equity when compared to Published by epublications@scu, 2

8 Journal of Economic and Social Policy, Vol 1, Iss 1 [2], Art 7 firms that do not pay dividends In spite of this, some firms are still paying dividends and investors and researchers are at a loss to explain this (Fama & French 21) This section focuses on the time trends in cash dividends of the firm sample selected, including the movement in number of 'payers' and 'non-payers' over time The sample firms from 1 to 22 are classified as payers or non-payers according to their dividend payment actions in the past (t-1) and in the present (t) Table 1: Counts and Percent of SET firms in Different Dividend Groups

9 Ronapat and Evans: Disappearing Dividends in the Thai Capital Market: Changing Firm 1 Co unt s of SE T fir ms All Fir ms Ne w List s Per cen t of SE T fir ms Pay ers No n- Pay er For me r- Pay er Nev er- Pai d Ne w List s Ne w List s tha t Pay Published by epublications@scu, 2 7

10 Journal of Economic and Social Policy, Vol 1, Iss 1 [2], Art 7 Patterns in the payment of dividends are explored from 1 to 22 with statistics from the SIMS database In 1, 42 percent of listed firms paid dividends (table 1) The percentage of payers rose to its peak of percent in 12 and fell gradually to 71 percent in 17 The percentage of payers fell to 24 percent in 1 This figure was recorded a year after the Asian crisis However, the percentage of firms paying dividends recovered slightly in 1 (2 percent) and rose to 44 percent by 22 However, the percentage of firms paying dividends in 22 (44 percent) was well below the 1 level (42 percent) Figure 1 shows the total number of firms on the Stock Exchange of Thailand (SET) each year and the number of firms that: (1) pay dividends; (2) do not pay dividends; () formerly paid dividends; and (4) have never paid dividends (SET 1, 11, 12, 1, 14, 1, 1, 17, 1, 1, 2a, 2b, 21a, 21b, 22a, 22b) Number of SET firms in different dividend groups All firms Payers Non-Payer Former Payers Never-Paid Num ber of SET firm s Year Figure 1: Source: Developed for this research The Number of SET Firms in Each Group Figure 2 shows the percentage of the total number of firms in the four dividend groups

11 Ronapat and Evans: Disappearing Dividends in the Thai Capital Market: Changing Firm Percent of all SET firms in different dividend groups Payers Non-Payer Former-Payer Never-Paid Percent Year Figure 2: Source: Developed for this research Percentage of All SET Firms in Different Dividend Groups Table 1 indicates that the number of listed firms increased sharply from 2 in 1 to almost 4 in 1 (pre-crisis) Table 1 also indicates that the number of listed firms in the sample declined to below 4 immediately after the crisis This figure has recovered slightly and stood at just below 4 firms in 22 However, the current number of listed firms is still below the figure for 1 and 17 The number of firms paying dividends increased from almost 2 in 1 to greater than firms in 1 However, the rate of increase in the number of payers has declined The number of firms which pay dividends has partly recovered although it was still below 2 in 22 There are four possible explanations for the decline in the number of firms paying dividends in the short term and the long term Firstly, existing and new listed firms exhibit a lower propensity to pay dividends, or are less willing to pay dividends Secondly, firms that have recently listed with the characteristics of non-payers tend to avoid paying dividends Thirdly, the characteristics of mature firms appear to have changed and many of these firms have ceased paying dividends Fourthly, many firms have been distressed since the crisis (both new lists and existing firms), and have ceased paying dividends This will be discussed more in detail below The number of firms which did not pay dividends before the crisis (the figure was less than 1) was consistently less than the number of payers However, the number of firms which did not pay dividends rose sharply from 112 in 17, to a peak of 2 in 1 This figure has declined slowly since 1, although it remained slightly above 2 in 22 Fama and French (21) suggested that firms which do not pay dividends can be classified into the two groups, former payers and firms which have never paid dividends Former payers are firms that do not pay dividends in the present year (time t) but paid dividends in a previous year Figure 1 indicates that the number of former payers (before and after the crisis) is similar to the number of non-payers In addition, the number of firms which have never paid dividends was relatively small in Published by epublications@scu, 2

12 Journal of Economic and Social Policy, Vol 1, Iss 1 [2], Art 7 1 The number of non-payers and former payers increased sharply in 17 and peaked in 1 (figure 1) This increase in the number of non-payers and former payers can be attributed to the Asian Economic Crisis Indeed, a strong decline in the number of firms which pay dividends also appears to result from the crisis The number of non-payers declined from about in 1 to about 2 in 22 In addition, the number of former payers declined from 2 in 1 to about 1 in 22 A trend that has emerged since the crisis is that the number of firms that have never paid dividends accounts for an increasingly larger share of the non-payer group (figure 2) It appears that: (1) firms that were listed between 17 and 22 are more likely to have never paid dividends, and (2) most of Thailand's listed firms tend to pay dividends if they are capable of doing so Therefore, most of the non-payers (former payers are the majority of this group) were forced to cease paying dividends due to financial distress flowing from the crisis This view is consistent with the findings of Fama and French (21) Former payers are likely to be distressed and have not paid dividends in time t However, it remains to be seen whether non-payers and former payers will pay dividends in the future This issue will be addressed later As stated earlier, the percentage of firms which paid dividends decreased from 42 percent in 1 to 44 percent in 22 Before the Asian Economic Crisis (1-1), the percentage of payers declined slightly from 42 percent in 1 to 1 percent in 1 However, the percentage of payers fell sharply from 71 percent to 24 percent during 17 and 1 In 1, the percentage of payers increased slightly to 2 percent Indeed, the percentage of payers has continued to rise and stood at 44 percent in 22 This figure is about 2 percent higher than the figure recorded during the crisis, although it remains about 4 percent lower than 1 The trends in the percentage of payers indicate: (1) in the long term, the payment of dividends appears to be disappearing because the percentage of listed payers has declined; (2) in the short term (the pre-crisis period), the percentage of payers declined slightly; this outcome may be an early indicator of the financial collapse which occurred in 17; and () in the short term (post-crisis period), the percentage of firms paying dividends has rebounded from the low point recorded during the crisis and may rise to pre-crisis levels in the future The decline in the percentage of dividend payers may also result from growth in the non-payer group due to new listings Before the crisis, the number of firms in most sectors expanded by more than 1 percent per annum due to new listings (table 1) However, the percentage of new listings declined to less than 1 percent of the total number of listed firms during the crisis and currently stands at about 2 percent Table 1 indicates that some newly listed firms paid dividends, although the percentage of these firms fluctuated between 1 and 22 The percentage of new firms that paid dividends stood at 22 percent (11), 7 percent (12), 1 percent (1), 2 percent (14), 2 percent (1), 7 percent (1) and percent in 17 After the crisis, less than percent of new lists paid dividends In addition, the percentage of new firms that paid dividends was lower than the total percentage of payers in all years with the exception of 1 and 21 (table 1) Therefore, it appears that new firms are less likely to pay dividends and this partly explains the decline in the percentage of payers between 1 and

13 Ronapat and Evans: Disappearing Dividends in the Thai Capital Market: Changing Firm Fama and French (21) found that the percentage of listed firms which paid dividends decreased sharply from almost 7 percent in 17 (a period of recession) to 2 percent in 1 As stated earlier, Fama and French (21) averaged the figures for the period preceding 17 and explored the decline in the payment of dividends after 17 However, Thailand's capital market is relatively new and has only experienced one major economic downturn (17-1) As stated earlier the percentage of listed payers in Thailand's market has recovered slightly since the crisis The percentage of firms paying dividends in the US has continued to decline Fama and French (21) did not indicate whether there was any change in the percentage of payers in the US market after 1 These researchers presented their data over several time periods and did not investigate trends from one year to the next Therefore, more research is needed to determine whether the decline in the number of payers in the US has been sustained This study presents the pattern of dividends year by year from 1 to 22, and in several time periods (range of years) This analysis is similar to that of Fama and French (21) This study found that the percentage of payers in Thailand's capital market increased from 2 percent in 1 to 4 percent in 22 (figure 2) Table 2 presents the dividend behaviour of payers in year t-1 and non-payers in year t- 1 in time t The table is divided into three parts The first part presents behaviour in year t of firms that paid dividends during the previous year These firms continue to pay dividends, ceased paying or were forced to delist The second part presents the behaviour of firms that were non-payers in time t-1 These firms commenced paying dividends, remained a non-payer, or were delisted from the stock exchange The last part shows the percentage of non-payers that commenced paying dividends in time t The table also indicates if these firms were former payers, or if they have never paid dividends Published by epublications@scu, 2 11

14 Journal of Economic and Social Policy, Vol 1, Iss 1 [2], Art 7 Table 2: What Happens in Year t to SET Firms that Do and Do Not Pay Dividends in Year t-1? What Happens in Year t to Firms that Pay Dividends in Year t Continue to Pay Stop Paying Delist 4 Source: SET (14, 17, What Happens in Year t to Firms that Do Not Pay Dividends in Year t-1 Start Paying Do not Pay Delist Percent of Non-Payers in Year t-1 that Start Paying in Year t-1 2, 22a) All Non-Payers (t-1) Former Payers Never Paid Note: Firms that continue to pay were payers in year t-1 and continue to pay dividends in time t Firms that cease paying dividends are firms that were payers in time t-1, but non-payers in time t Firms that delist are firms that were payers in time t-1 but delisted in time t Numbers shown in percentage Table 2 shows the likelihood that payers and non-payers of the last period (t-1) will continue to pay, commence or cease paying dividends in this year Firms which paid dividends last year (t-1) tend to continue paying dividends in year t Before the crisis, almost percent of the payers continued to pay dividends (table 2) In 17, however, the percentage of payers in the previous year (t-1) that continued to pay (in t) fell to 17 percent and to 4 percent during 1 After the crisis, the percentage of payers in the previous period that continued to pay in the next period increased to 7 percent in 1, percent in 2, 2 percent in 21 and finally 4 percent in 22 Although the number of firms that continue to pay dividends has increased to more than percent, it should be stated that the figures after the crisis, are compared with the lowest percentage of payers that continued to pay dividends in 1 (4 percent) Therefore, after 1, the denominator (the total number of payers in the last period) is less than the figure observed in the pre-crisis period The question which arises is what has occurred to the remaining percent of firms which are firms that ceased paying dividends during the crisis? Firms which paid dividends before the crisis but did not pay during the crisis ceased paying dividends at the rate of 4 percent and delisted at the rate of percent in 1 Before the crisis, the percentage of payers that ceased paying dividends was about 1 percent per annum However, this figure rose to 17 percent in 17 and peaked at 4 percent in 1 In addition, the percentage of payers that delisted in the next period (t) was less than 1 percent per annum before the crisis, but rose to percent in 1 Finally, after the crisis, a greater proportion of firms which paid dividends continued doing so while a lesser proportion ceased paying and delisted This indicates that the market is recovering Firms that did not pay dividends in the last period commenced paying at the rate of 14 percent in 11 and 4 percent in 12 This indicates that the vast majority of firms possessed a high propensity to pay, or resumed paying dividends at this time However, this figure fell to percent in 14 and to less than 1 percent in

15 Ronapat and Evans: Disappearing Dividends in the Thai Capital Market: Changing Firm Finally, after the crisis, this figure increased to above 1 percent, although it is still lower than the percentage of non-payers that commenced paying dividends before the crisis In summary, fewer firms recommenced paying dividends after the crisis Before the crisis, non-payers appeared less willing to commence paying dividends in the next period (decreasing their propensity to pay dividends) The percentage of nonpayers that continued to avoid this payment in the next period also supports this finding (table 2) After 1, a large proportion of non-payers continued to avoid paying dividends, even though it was still several years before the crisis In addition, only percent of non-payers commenced the payment of dividends in 1 This figure increased to 121 percent in 22, indicating that the market was recovering and a high propensity to pay dividends by former non-payers The last part of table 2 indicates that former payers resumed the payment of dividends more frequently than firms that have never paid dividends This view supports the earlier discovery that former payers omit the payment of dividends but do not intend to cease the payment on a permanent basis The descriptive statistical analysis of Fama and French (21) suggests that former payers are distressed firms which ceased paying dividends to preserve their cash The relationship between former payers and distressed will be analysed for Thailand with the aid of a variety of ratios This analysis is presented below A preliminary analysis of the descriptive statistics reveals that the percentage of firms which paid dividends started to decline before the crisis, and fell sharply during the crisis The percentage of payers rose slightly after the crisis, but remains much lower than the level recorded before the crisis It also appears that payers were less likely to continue the payment of dividends during the crisis The analysis suggests that the propensity to pay dividends of existing payers is lower However, this suggestion will be confirmed below As stated earlier, the percentage of non-payers has increased over time and increased sharply immediately after the crisis Most of the non-payers were former payers that were forced to avoid the payment of dividends due to the crisis In addition, it appears that non-payers were less willing to recommence paying dividends This supports the proposed explanation for the observed decline in the percentage of payers; non-payers are less likely to recommence paying dividends Consequently, the total number of payers is smaller and represents a smaller proportion of the total number of firms Many new firms were listed at SET before the crisis and a large proportion of these firms did not pay dividends The percentage of new firms that paid dividends was less than the percentage of listed firms that paid dividends in most years (table 1) Therefore, new lists usually do not pay dividends and the denominator (total number of listed firms) is increasing over time while the numerator (the total number of payers) is becoming smaller, leading to a decline in the percentage of payers The findings of this section have raised several questions which require further investigation: (1) What happened to listed firms in 17 to 1? Why do payers tend to cease paying dividends? What are the characteristics of firms which pay dividends? Has the propensity to pay dividends fallen? Have characteristics of these firms changed? (2) What are new lists' characteristics? Are the characteristics of new lists Published by epublications@scu, 2 1

16 Journal of Economic and Social Policy, Vol 1, Iss 1 [2], Art 7 different from mature firms? Is the decline in percentage of dividend payers due to new listing? Characteristics of Dividend Payers As stated earlier, Fama and French (21) suggested there are three characteristics of firms which pay dividends: (1) profitability (2) investment opportunities and () size Each characteristic will be measured and compared with the aid of a variety of ratios The analysis will compare the changes in these characteristics during the pre-crisis and post-crisis periods by using descriptive statistics In addition, each characteristic will be compared with the findings of the literature Finally, this analysis will test differences in the mean of these characteristics to determine whether they are significantly different Table : Average Values of Six Characteristics Ratios Et/At (percent): The Ratio of Aggregate Earnings Before Interest to Aggregate Assets All Firms Payers Non- Payers Never Paid Former Payers All New Lists Payers Non- Payers Yt/BEt (percent): The Ratio of Aggregate Common Stock Earnings Over Aggregate Book Equity All Firms Payers Non- Payers Never Paid Former Payers All New Lists Payers Non- Payers dat/at (percent): The Growth Rate in Assets All Firms Payers Non- Payers

17 Ronapat and Evans: Disappearing Dividends in the Thai Capital Market: Changing Firm Never Paid Former Payers All New Lists Payers Non- Payers Table : Average Values of Six Characteristics Ratios (Continued) Vt/At: The Ratio of the Aggregate Market Value to the Aggregate Book Value of Assets All Firms Payers Non- Payers Never Paid Former Payers All New Lists Payers Non- Payers Published by epublications@scu, 2 1

18 Journal of Economic and Social Policy, Vol 1, Iss 1 [2], Art 7 At: Total Assets All Firm Payers Non-Pay Never Pa Former P All New Payers Non-Pay Lt/At: The Ratio of Total Liabilities to Total Assets All Firm 7 1 Payers Non-Pay Never Pa Former P All New Payers Non-Pay

19 Ronapat and Evans: Disappearing Dividends in the Thai Capital Market: Changing Firm Table : Average Values of Six Characteristics Ratios (Continued) Et/At (percent): The Ratio of Aggregate Earnings Before Interest to Aggregate Assets All Firms Payers Non-Payers Never Paid Former Payers All New Lists Payers Non-Payers Yt/BEt (percent): The Ratio of Aggregate Common Stock Earnings Over Aggregate Book Equity All Firms Payers Non-Payers Never Paid Former Payers All New Lists Payers Non-Payers Published by epublications@scu, 2 17

20 Journal of Economic and Social Policy, Vol 1, Iss 1 [2], Art 7 Table : Average Values of Six Characteristics Ratios (Continued) dat/at (percent): The Growth Rate in Assets All Firms Payers Non-Payers Never Paid Former Payers All New Lists Payers Non-Payers Vt/At: The Ratio of the Aggregate Market Value to the Aggregate Book Value of Assets All Firms Payers Non-Payers Never Paid Former Payers All New Lists Payers Non-Payers

21 Ronapat and Evans: Disappearing Dividends in the Thai Capital Market: Changing Firm Table : Average Values of Six Characteristics Ratios (Continued) At: Total Assets All Firms Payers Non-Payers Never Paid Former Payers All New Lists Payers Non-Payers Lt/At: The Ratio of Total Liabilities to Total Assets All Firms Payers 4 2 Non-Payers Never Paid Former Payers All New Lists 12 7 Payers Non-Payers Source: Developed for this research Note: At is total asset, BEt is the book value for common equity, MEt is the market value of common equity, Lt is the book liabilities, Vt is the market value of the entire firm, Et is the earnings before interest but after taxes, Yt is the after-tax earnings to common stock and dat is the changes in the assets from time t and t-1 Results for ranges of years are grouped and averaged according to the dividend groups Et/At, Yt/Bet, dat/at are shown in percentages Vt/At and Lt/At are shown in ratio values At is shown in Baht values Published by epublications@scu, 2 1

22 Journal of Economic and Social Policy, Vol 1, Iss 1 [2], Art 7 Profitability Table presents the characteristics of firms in a variety of dividend groups The profitability of the dividend groups is expressed by two ratios '(1) E t /A t (the ratio of aggregate earnings before interest to aggregate assets) (2) Y t /BE t (the ratio of aggregate common stock earnings over aggregate book equity)' (Fama & French 21, p 7) The discussion is divided into two parts, in line with the two ratios for measuring profitability The Ratio of Aggregate Earnings before Interest to Aggregate Assets (E t /A t ) Before the Asian Economic Crisis (11-1) payers, on average, have higher profitability (47 percent) than non-payers (4 percent), firms which have never paid dividends (22 percent) and former payers (2 percent) All samples before the crisis indicate that firms which pay dividends are the most profitable (71 percent) This result is consistent with the findings of Fama and French (21) Therefore, when using the E t /A t ratio for measuring profitability, payers have higher profitability than non-payers When comparing the two groups of non-payers, firms that have never paid dividends were more profitable (22 percent) than former payers (2 percent) During the Asian Economic Crisis almost 7 percent of firms listed in Thailand suffered large losses (SET 22a) Therefore, payers were less profitable during and after the crisis (17-22), although they remained the most profitable ( percent) of the four dividend groups While payers experienced positive profitability, nonpayers experienced losses (-24 percent) Within the non-payer group, firms that have never paid dividends suffered slightly higher losses (-2 percent) than former payers (-1 percent) The average profitability (E t /A t ) of all samples after the crisis was -14 percent This was lower than the profitability of dividend payers, but higher than the figure recorded for non-payers and their sub-groups This finding is consistent with the literature (Fama & French 21) During the period 11 to 21, payers of dividends were the most profitable ( percent) followed by firms that have never paid dividends (4 percent) Non-payers of dividends were more profitable than former payers Non-payers experienced losses of -4 percent, while former payers experienced losses of - percent The all firm average level of profitability was 4 percent Therefore, the E t /A t ratio indicates that payers are more profitable than non-payers during favourable and poor economic conditions Former payers tend to be distressed because they are the least profitable in good times and experience large losses in bad times Hence, these firms often cease paying dividends The descriptive statistics have shown that when non-payers are divided into two sub-groups (never paid and former payers), firms which have never paid dividends perform better than former payers This view is consistent with the findings of La Porta et al (2), Benito and Young (21), DeAngelo, DeAngelo and Skinner (22), Fama and French (21) and Aivazian, Booth and Cleary (2) Generally, firms that were newly listed on the market before the crisis were highly profitable (117 percent) and more profitable than the average of the existing listed 2

23 Ronapat and Evans: Disappearing Dividends in the Thai Capital Market: Changing Firm firms (table ) The profitability of the new lists that paid dividends was also high (111 percent) while the profitability of new lists that did not pay dividends was lower (7 percent) All figures for profitability measured by E t /A t were higher for new lists than for existing listed firms before the crisis period After the crisis, the losses of the new lists increased slightly (-2 percent) when compared to the figure recorded for the existing listed firms However, new lists that paid dividends were highly profitable (71 percent) when compared to all existing listed firms that pay dividends For the whole period of investigation, new lists were relatively profitable (42 percent) and generally more profitable than existing firms during the same period New lists that paid dividends were highly profitable ( percent) and this figure is higher than existing firms which paid dividends Fama and French (21) found that new lists tended to be highly profitable early on, although their profitability declined after the crisis in 17 Therefore, Fama and French (21) concluded that the decline in the percentage of payers after 17 is partly due to the new characteristics of the new lists (low profitability, high investment opportunities and small size) However, in the case of Thailand, a clear trend in the disappearing of dividends is not evident It appears that Thai firms have resumed paying dividends after a crisis Therefore, new lists with higher profitability than mature firms are more common in Thailand There is no clear explanation for this result although it appears that changes in the characteristics of new lists and mature firms are very small and the Thai capital market has not reached its turning point The Ratio of Aggregate Common Stock Earnings over Aggregate Book Equity (Y t /BE t ) Fama and French (21) suggested that E t, or earnings before interest, may not be the relevant tool affecting firms' dividend payment decisions These researchers suggested that Y t, or earnings available for common stocks, could be a better tool Table indicates that the gap between the profitability of payers and non-payers tends to be wider when measured by the Y t /BE t ratio This view is consistent with the findings of Fama and French (21) During the pre-crisis period, dividend payers were more profitable (121 percent) than the other dividend groups When focusing on non-payers, firms that have never paid dividends were more profitable ( percent) than former payers (2 percent) The average level of profitability for all firms was percent, which is lower than that recorded for payers but higher than the other dividend groups This finding is similar to the findings of Fama and French (21) During and after the Asian Economic Crisis, all firms, on average, experienced losses of -22 percent The largest losses were recorded in 17 and 1 (table ) Furthermore, when analysing each dividend group, the data indicate that every dividend group experienced losses Published by epublications@scu, 2 21

24 Journal of Economic and Social Policy, Vol 1, Iss 1 [2], Art 7 In relation to the period 11 to 22, the average Y t /BE t ratio for payers was -21 percent, while firms which have never paid dividends experienced greater losses of - 42 percent Non-payers and the sub-group of former payers experienced very large losses of -14 percent and -42 percent respectively The average level of profitability of all firms was -12 percent When using Y t /BE t as a measure of profitability, payers are more profitable than nonpayers during positive economic conditions This finding is consistent with La Porta et al (2), Benito and Young (21), DeAngelo, DeAngelo and Skinner (22), Fama and French (21) and Aivazian, Booth and Cleary (2) During difficult times, payers experienced smaller losses than non-payers The profitability of new lists was 14 percent before the crisis (using Y t /BE t ) The average profitability of new lists that paid dividends was 147 percent, which is much higher than the level experienced by the existing firms which paid dividends During and after the crisis, new lists made losses of -4 percent However, new lists that paid dividends were highly profitable and this performance was superior to the results achieved by listed firms that pay dividends For the whole period of investigation (1-22), new lists experienced losses of - 11 percent, although the payer groups were more profitable (1 percent) than the returns achieved by existing listed firms that paid dividends (table ) These findings are similar to Fama and French (21) However, Fama and French (21) stated that the characteristics of newly listed firms changed from high to low profitability, after a strong downturn in the US in 17 It should be noted that Fama and French (21) concluded that this outcome is not consistent, over time, for new lists that paid dividends and mature firms that paid dividends Mature firms which pay dividends are less profitable than new listed payers when using both ratios (E t /A t and Y t /BE t ) in the US This finding is similar to the results of this study on the Thai stock market The study focuses on the characteristics of the dividend payers Therefore, it is important to highlight the issue that new lists which pay dividends are more profitable than mature firms that do not pay dividends in Thailand Furthermore, the profitability of new lists has been consistently higher than mature firms over the period of investigation Investment Opportunities The investment opportunities of firms are measured by the three ratios which were recommended by Fama and French (21) (1) da t /A t (growth rate in assets) (2) V t /A t (the ratio of the aggregate market value to the aggregate book value of assets) () RD t /A t (research expenditure) in table However, this study will apply the first two ratios because data on research and development is not available in Thailand This information is not required by SEC and need not be presented on the financial statements of listed firms (SET 2) The two ratios will be calculated and analysed separately in this study 22