Shareholder wealth effects of equity issues in emerging markets: evidence from rights offerings in Greece.
An important difference between these two methods of raising equity capital is the possibility of wealth transfers from new to old shareholders, arising from the information asymmetry between management and outside investors. In contrast to general cash offers, in rights issues the new shares are acquired by existing shareholders. Thus, to the extent that all current shareholders exercise their preemptive rights, the wealth transfer effect (described by Myers and Majluf, 1984) becomes irrelevant. Consequently, any stock price effects associated with announcements of rights issues cannot be attributed to this information effect. The ability to isolate this effect makes rights issues an ideal sample for further examination and understanding of the stock price reaction to announcements of equity offerings.
While prior studies, relying on US data, have shown that announcements of general cash offers are associated with a significant price reduction of 3%, on average, of the firms' outstanding common stock, the evidence regarding rights issues is mixed.(2) For example, Nelson (1965), examining all rights offerings in the US for the period 1946-1957, finds that share prices six months after rights offerings are not significantly different from prices six months prior to the offerings. Scholes (1972), investigating US rights issues for the period 1926-1966, finds that prices of shares generally rise before the issue, fall by 0.3% during the month of the issue, but do not change after the issue. Smith (1977), using monthly returns, finds no significant excess returns during the month of a rights offer. White and Lusztig (1980) and Hansen (1988) document a negative reaction to rights offers announcements. Kothare (1991), investigating US rights offerings for the period 1970-1987, reports negative (statistically significant) announcement period abnormal returns. Eckbo and Masulis (1992), examining rights offerings in the US for the period 1963-1981, report negative (marginally significant) announcement period abnormal returns. Marsh (1979) analyzes rights issues in the UK and finds large positive abnormal returns prior to the announcement of the issue, but a statistically insignificant setback in the months surrounding the issue itself. Loderer and Zimmermann (1988) investigate rights issues in Switzerland using monthly stock returns and report insignificant average abnormal returns. Kang (1990), examining rights issues in Korea, finds a significant stock price increase during the period surrounding the announcement of a rights issue.
There is a key difference between the studies on general cash offers and those on rights issues. Studies of general cash offers rely mainly on US data, while studies of rights issues cover various countries. Interestingly, the study from Korea reports strikingly different findings from those of the US and UK studies. This disparity suggests that the mixed evidence on rights offerings may reflect the different institutional characteristics of the associated countries, making further research potentially useful. For several reasons, presented in the next section, Greece is an appropriate country for further investigation of the valuation effects of rights issues.
This study, based on an emerging European capital market with different institutional characteristics from most countries covered by prior studies, contributes to the literature because: 1) it examines investors' reaction to equity offerings, without the complication of information asymmetries, 2) it examines rights issues in a different institutional framework, and 3) it enriches the limited empirical evidence on emerging European capital markets.
The remainder of this paper is organized as follows. Section I provides a brief description of the institutional characteristics of Greek capital markets as pertinent to rights offerings. Section II describes the data. Section III contains the methodology used to measure the announcement effect of rights issues on share prices and presents the corresponding findings. Section IV provides further evidence based on cross-sectional analysis. Section V presents a summary and conclusions.
I. Institutional Characteristics of Greek Capital Markets
The institutional environment of the Greek capital markets in the period 1981-1990 was different from that of most countries covered by prior studies. The major areas which may have a role in explaining stock price reactions to announcements of rights issues are: 1) the secondary market for rights; 2) the firms' ownership structure; 3) the size of the corporate bond market; and 4) the information disclosure requirements.
A. Secondary Market for Rights
For all practical purposes there did not exist an organized secondary market for rights in Greece during the period of this study. The absence of an active market for rights suggests that an equity issue may not dilute the proportional ownership of existing shareholders. On the other hand, an active market for rights, such as the one in the US, brings about a reduction in the fractional ownership of those shareholders who do not subscribe to the issue by selling their rights. As a result, rights issues in the absence of an active market for rights may be associated with less negative information effects relative to the latter case.
B. Ownership Structure
In many publicly traded firms in Greece, the majority of the shares are owned either by the state or by a few members of a single family, resulting in low ownership diffusion. This characteristic implies that, in rights issues, shareholders of firms with low ownership diffusion must make larger capital contributions in order to sustain their percentage ownership. Therefore, shareholders are expected to approve the rights issue only if the firm's prospects are promising, implying that announcements of rights issues are perceived as conveying good news to the capital markets.
C. Size of the Corporate Bond Market
Until 1987 no corporate bonds were listed on the Athens Stock Exchange (ASE). The first bonds were introduced in 1988, but they experienced no trading activity. The first "active" bonds were those issued by the state-owned National Mortgage Bank in 1989. The capital raised through publicly traded corporate bonds in 1988-1990 represented only 0.02% (in 1988) to 0.27% (in 1990) of the total market value of all securities listed on the ASE during that period.(3) Despite the small size of the corporate bond market, the debt ratio of Greek firms in the corresponding period was relatively high. The debt-asset ratio approached 70%, but most debt represented short and intermediate-term bank credit.(4) Therefore, the main source of long-term capital in the period 1981-1990 was the equity market, which in the case of already listed firms took the form of rights issues only. Thus, rights offers provided a good opportunity for the capital markets to reevaluate the long-term prospects of the issuing firms.
D. Information Disclosure Requirements
During the period of this study, firms listed on the ASE were required to publish an audited annual balance sheet and income statement within the six-month period following the statements' closing date. They were also required to produce brief interim (semiannual) statements within a four-month period following the statements' closing date. However, due to light penalties, most firms did not comply with the latter requirement. In addition, Greek firms were not required to produce consolidated financial statements, nor were they bound, until 1989, by a standardized accounting system.
Basic listing requirements were set up in 1985, along with the introduction of the prospectus for all public offerings (initial and seasoned). The prospectus is available to interested investors at the beginning of the subscription period. Although the exact type of information that must appear in the prospectus was not specified until 1992, firms have generally disclosed information about their operations (past, current, and prospective), the identity of their major shareholders, and all capital changes since their establishment. In addition, they have disclosed some information on the new issue, including the number of shares issued, issue price, total capital raised, and purpose of the issue (but only in very general terms).
The lax financial reporting requirements, along with the absence of an active corporate bond market and the lack of a well-organized securities analysts industry, suggests that corporate decisions may not have been announced in a timely fashion, thus preventing the appropriate price adjustments.(5) In this environment equity offerings became a major credible outlet for the release of important corporate information into the capital markets. The announcement of a rights issue thus prompted investors to focus more on the firm and reappraise its value.
II. Sample and Data Description
The initial sample used in this study included all announcements of common stock rights offerings by firms listed in the Athens Stock Exchange (ASE) during the period 1981-1990. The sample was identified through a search of ASE publications, the daily press (all major newspapers were considered), and the Database of Daily Athens Stock Exchange Security Returns.(6) The announcement day investigated is the day of the first public announcement in the press. To ensure that this was the first day that the information became public, the announcement was confirmed or corrected by reviewing each firm's official records at the ASE.
The announcement is a press release by the board of directors calling a general meeting of shareholders to approve the rights offering proposal. All proposals included in the sample were subsequently approved by the shareholders.
Greek corporations issue common and/or nonvoting preferred shares. Both types may take either registered or bearer form, depending on the type of issuing corporation. According to the law, banks, insurance, leasing, shipping, and media firms must issue registered shares. All other firms can choose either type of security.
During the period under investigation, 61 listed firms made a total of 111 new share offerings, which can be classified into three groups: 73 common stock issues, 4 preferred stock issues, and 34 contemporaneous issues of common and preferred stock. Daily returns for these securities were taken from the Database of Daily Athens Stock Exchange Security Returns, making this study one of the first empirical studies to use this database.(7)
For each rights issue, the daily stock returns were obtained for 211 trading days surrounding the announcement date (i.e., t = -200 to t = +10). These 211 days are divided into two groups, the estimation period from t = -200 to t = -51 and the prediction period from t = -50 to t = +10, with t = 0 corresponding to the date of the initial public announcement. To limit problems from infrequent trading, the final sample includes only firms having at least 30 daily trading returns during the estimation period. In addition, in order to avoid mixing one-day returns with multiple-day returns, the return following a missing return observation is also excluded.(8) Thus, the final sample contains a total of 59 [TABULAR DATA FOR TABLE 1 OMITTED] rights issue announcement dates, of which 16 offers represent equity offerings of state-owned firms.(9) Of the final sample, 38 issues represent common stock rights offerings, 4 represent preferred offerings, and the remaining 17 offers are contemporaneous rights offerings of common and preferred stock. The analysis here focuses on the common stock price reaction only.
Table 1 presents summary statistics (mean, standard deviation, median, maximum, minimum) of pertinent variables for the sample of 38 rights issues of common stock (Panel A), and the 17 rights issues of common and preferred stock (Panel B).
As shown in Panel A, the new common shares issued represent, on average, 43.3% of common shares outstanding with a minimum of 4% and a maximum of 100%. The average offer price as a percentage of the market price one month prior to the announcement of the stock offering is 62.6% with a minimum of 19.7% and a maximum of 116.8%. The corresponding average for the sample of Swiss rights, reported in Loderer and Zimmermann (1988), is 39.8% with a minimum of 2.5% and a maximum of 85.8%. The corresponding average for US rights offerings is 92% for the period 1973-1983 (Loderer and Zimmermann, 1988) and 87.6% for the period 1963-1981 (Eckbo and Masulis, 1992). It seems therefore, that European rights issues are associated with larger discounts relative to their US counterparts. The value of the additional shares as a percentage of the value of outstanding shares is on average 58.9% with a minimum of 3.1% and a maximum of 191.7%. The corresponding average for Swiss rights issues is 17% (Loderer and Zimmermann, 1988). For the US, Mikkelson and Partch (1986) report an average of 15.1%. Also, Eckbo and Masulis (1992) report that in seasoned common stock offers (including firm-commitment underwritten offers and rights offers) of US industrial firms, the amount offered relative to the market value of equity is, on average, 23.8%. Therefore, when Greek-listed firms issue additional equity, they do so on a larger scale than firms in the US and Switzerland. Similar conclusions can be drawn from Panel B, as well.
III. Estimation of Abnormal Stock Returns
To examine the effects of rights issues on stock returns, the event study methodology, as described by Brown and Warner (1980, 1985), is used. Accordingly, for each security i the market-adjusted model is used to calculate an abnormal return (prediction error) [e.sub.it] for the event day t as follows.(10)
[e.sub.it] = [R.sub.it] - [R.sub.mt] (1)
where [R.sub.it] is the return on security i for the event day i and [R.sub.mt] is the market return on the event day t. The index used for the market portfolio returns is a value-weighted index constructed by Travlos (1992).
Also, a binomial-sign test is used to examine whether the proportion of positive event day returns is greater than expected under the null hypothesis. Since it is a test of location, it is not affected by outlier returns in either a positive or negative direction (Doukas and Travlos, 1988).
A. Findings: Stock Price Reaction to Announcements of Rights Issues
Table 2 presents daily average abnormal returns and cumulative average abnormal returns for the overall sample, for the period of t = -10 to t = +10 days relative to the announcement day (t = 0).(11) Column 1 lists the event time relative to the announcement day in terms of trading days. Column 2 presents the number (N) of firms with valid returns on each event day. Column 3 lists the number of negative/positive abnormal returns. Column 4 presents the daily average abnormal returns (AR) for each event day. Column 5 contains the cumulative daily average abnormal returns (CAR). Column 6 contains the two-tailed t-statistics for the average abnormal returns (t(AR)). We also present cumulative daily average abnormal returns (CAR) for two, six, and eleven days, and the corresponding two-tailed t-statistics and the two-tailed Z-values for the proportion of positive abnormal returns on the event days - 1 and 0.
As shown in Table 2, the average abnormal return on the announcement day (t = 0) is 2.45%, which is statistically significant (t = 5.09) at the 0.01 level. In addition, the Z-value for the number of positive abnormal returns on the announcement day is statistically significant at the 0.01 level (Z = 2.89).
Since the exact time of the press release is not known, either its publication day (in the daily press) or the day before might be the relevant announcement day, suggesting that the use of a two-day cumulative abnormal return (CAR) is appropriate. The two-day CAR is 3.97%, which is statistically significant at the 0.01 level (t = 4.12).
Also, the six-day and 11-day CARs are 8.23% and 11.52%, respectively, which are statistically significant at the 0.01 level (t = 2.85) and 0.05 level (t = 2.17), respectively. The statistically significant abnormal returns observed before the announcement day might be due to leakage of the board of directors' decision to propose the rights issue. Alternatively, since trading on insider information was not prohibited during the sample period, these abnormal returns may reflect insiders' trading activities. After the announcement day, the abnormal returns are, in general, statistically insignificant.(12)
Based on these findings, we cannot accept the null hypothesis of zero abnormal returns. Thus, we conclude that announcements of equity rights issues are associated with a positive effect on share prices. The positive abnormal returns and cumulative abnormal returns confirm the findings reported by Kang (1990) and partially those by Loderer and Zimmermann (1988), but they are in direct contrast to the findings reported by the other studies.
Table 2. Abnormal Common Stock Returns from Equity Rights Issues of Firms Listed on the Athens Stock Exchange in 1981-1990 This table shows the number of negative and positive observations, daily average abnormal returns (AR), cumulative average abnormal returns (CAR), and t-statistics of AR (t(AR)) for each day in the period from 10 days before through 10 days after the announcement date of equity rights issues. Negative/ Period N Positive AR CAR t(AR) -10 47 25/22 0.00398 0.00398 0.82617 -9 45 26/19 0.00646 0.01044 1.34262 -8 47 17/30 0.02109 0.03153 4.37934(***) -7 46 25/21 0.00308 0.03461 0.64060 -6 47 28/19 -0.00168 0.03293 -0.34912 -5 48 26/22 -0.00028 0.03265 -0.05871 -4 50 24/26 0.00430 0.03695 0.89265 -3 47 18/29 0.02905 0.06599 6.03211(***) -2 44 23/21 0.00954 0.07553 1.98074(*) -1 43 18/25 0.01513 0.09066 3.14282(***) 0 48 14/34 0.02451 0.11518 5.09117(***) 1 50 26/24 -0.00131 0.11387 -0.27144 2 49 28/21 -0.00545 0.10842 -1.13211 3 41 21/20 0.00161 0.11003 0.33487 4 42 22/20 0.00127 0.11130 0.26365 5 47 29/18 -0.00700 0.10430 -1.45384 6 44 20/24 0.00230 0.10661 0.47837 7 44 19/25 0.01095 0.11755 2.27357(**) 8 46 28/18 -0.00522 0.11234 -1.08368 9 42 24/18 -0.00370 0.10863 -0.76881 10 41 14/27 0.01532 0.12396 3.18205(***) Cumulative Average Abnormal Returns (CAR): t-Statistic (CAR) (-1,0) = 0.03965 4.11699(***) (CAR) (-5,0) = 0.08225 2.84680(***) (CAR) (-10,0) = 0.11518 2.17458(**) Z-Value Number of Positive AR Period -1: 1.06749 Number of Positive AR Period 0: 2.88675(***) *** Significant at the 0.01 level. ** Significant at the 0.05 level. * Significant at the 0.10 level.
IV. Cross-Sectional Analysis
In order to thoroughly analyze the abnormal returns associated with rights issues announcements, we perform a cross-sectional analysis of the abnormal returns. This analysis helps discriminate among various hypotheses and identify factors that explain the abnormal stock returns.
A. Hypotheses Tested
Miller and Rock (1985) predict a negative stock price reaction to equity issues because they are perceived as releasing negative information about the firm's cash flows.(13) In contrast to the negative information effect, equity issues can also be interpreted as favorable news about the firm's investment opportunities. In particular, since the additional capital must be committed by the existing shareholders, rights issues attest to the shareholders' confidence in their own firm's future. That is, rights issues are perceived as a signal that the firm has discovered new positive-net-present-value projects, causing a positive reevaluation of the firm's shares. In addition, to the extent that larger rights offerings are associated with larger-net-present-value projects, the stock price reaction should be related to the size of the offerings.
Available empirical evidence provides support for this hypothesis. McConnell and Muscarella (1985) find that stock prices rise about 1% at the announcement of corporate investment increases. Masulis and Korwar (1986) conjecture that the negative stock price reaction to equity offering announcements can be partially offset by contemporaneous announcements of capital expenditure increases.
Price Pressure Effects
The price pressure argument assumes that, at any given instant, the demand curve for a firm's shares is downward-sloping and that an increased supply of shares decreases their price (see Asquith and Mullins (1986), Loderer and Zimmermann (1988)). However, this prediction conflicts with the classical view, which holds that in capital markets, near perfect substitutes for a firm's securities are always available. According to the latter hypothesis, the issuing firm's shares and their substitutes must sell at the same price, implying a horizontal demand curve for a firm's stock (Scholes (1972), Marsh (1979), Hess and Frost (1982)).
The price pressure hypothesis has been tested extensively, mainly by investigating the relation between price and quantity changes in new equity offerings. The evidence on this matter is mixed, with Scholes (1972), Marsh (1979), and Hess and Frost (1982) finding no negative relation, while Asquith and Mullins (1986), Masulis and Korwar (1986), and Loderer and Zimmermann (1988) documenting a negative relation. This approach has been challenged, however, by some later studies (see Loderer and Zimmermann (1988) and Loderer, Cooney, and Van Drunen (1991)) which point out that if the demand functions for different stocks are not identical and firms do not face the same initial price-quantity combinations, downward-sloping demand curves can generate almost any cross-sectional relation between changes in stock price and the number of shares outstanding.
To address this problem, Loderer, Cooney, and Van Drunen (1991) test the relation between price changes and possible determinants of demand elasticities in a sample of primary stock offerings by US-regulated firms for the period 1969-1982. Following Morton's (1987) theoretical model, it is expected that announcement period abnormal stock returns are inversely related to both firm size and return variance, and positively related to the investor base holding the stock. Loderer, Cooney, and Van Drunen report a negative stock price reaction to announcements of stock offerings, with no evidence that the decline is the result of adverse information about future cash flows, and some evidence (inconclusive) that the offer announcement effects are related to Merton's (1987) determinants of price elasticities.
Wealth Redistribution Effects
An unexpected issue of new equity reduces the corporate financial leverage ratio, making the debt less risky. Consequently, the market value of the debt increases at the expense of the shareholders. Masulis and Korwar (1986) report regression results that show a negative relationship between the abnormal returns on the announcement day and the leverage change associated with the issue.
As discussed in Section II, equity offerings in Greece are characterized by a substantial dispersion of the ratio of issue price to closing market price one month prior to the announcement day. This raises the question of whether the issue price has any valuation implications. Recently, Michaely and Shaw (1995) show that auditor reputation conveys information about the risk of the issue.
According to the prevailing view, the issue price in rights offerings is irrelevant since any mispricing of the new shares is reflected in the preemptive right value (see e.g., Brealey and Myers, 1991). On the other hand, Heinkel and Schwartz (1986) argue that the issue price in uninsured rights offerings may convey a signal about firm quality. In their model, offer failure is costly and a rights issuer who privately expects the share price to fall during the rights offer period prefers to self-insure by selecting a relatively low offer price in relation to the current market price. As a result, market participants infer the issuer's private information from the size of the offer price. Thus, Heinkel and Schwartz (1986) suggest that the issue price is a source of valuable information. They also predict that lower issue prices will cause larger downward adjustments in the stock's price. Loderer and Zimmermann (1988) find some evidence that offer prices convey this type of information.
The cross-sectional analysis is performed by estimating the regressions:
[CAR.sub.i] (-t,0) = [a.sub.0] + [a.sub.1][INVEST.sub.i] + [a.sub.2][SIZE.sub.i] + [a.sub.3][VAR.sub.i] + [a.sub.4][OFFER.sub.i] + [a.sub.5][DTOA.sub.i] + [a.sub.6][CONTR.sub.i] + [a.sub.7][LIQUID.sub.i] + [a.sub.8][MARKET.sub.i] + [a.sub.9][S-P.sub.i] + [a.sub.10]B-[R.sub.i] + [e.sub.i] (2)
[CAR.sub.i] (-t,0) is [CAR.sub.i] (-10,0) that is, the eleven-day (t = - 10 to t = 0) cumulative abnormal stock return (CAR) for sample firm i;(14)
INVEST is a measure of the additional capital committed by existing shareholders relative to the firm's outstanding equity capital;
SIZE is a measure of the firm's size;
VAR is the return variance of firm i;
OFFER is the offer price as a fraction of firm i's stock price;
DTOA is a dummy variable based on the firm's debt-to-assets ratio;
CONTR is a measure of the firm's ownership diffusion;
LIQUID is a measure of the firm's stock liquidity;
MARKET is a measure of the stock market's performance prior to the issue;
S-P is a dummy variable separating the state-owned firms from the non-state-owned firms;
B-R is a dummy variable separating the registered from the bearer shares; and is an error term with the usual OLS properties.
The first five variables are introduced in order to test the hypotheses outlined in Section A, while the last five variables capture additional determinants of abnormal returns. Their relation to the abnormal returns is as follows:
INVEST: This variable is used to test information effects. This variable is computed as INVEST = EI/EO, where EI is the new equity issued and EO is the equity capital outstanding one month prior to the announcement day. It represents the additional capital, relative to existing capital, committed by the shareholders.(15) The amount of the new capital may represent an unanticipated deficit in internal cash flow and thus elicit a negative price response ([a.sub.1] [less than] 0), as predicted by Miller and Rock (1985). Alternatively, this variable can capture information effects arising from revised expectations of investment opportunities. Considering that the firm will contribute additional capital to positive-NPV projects only, the larger the INVEST, the greater the stock price increase, implying that [a.sub.1] [greater than] 0.
SIZE: The SIZE variable is used to test the price pressure hypothesis. The firm's size is measured in absolute terms (natural logarithm of the product of the number of common shares outstanding times the share price one month prior to the announcement day). As pointed out by Loderer, Cooney, and Van Drunen (1991), greater compensation (i.e., a larger expected return) is required if investors have already tied up a substantial portion of their wealth to hold the stock of a large firm, implying that [a.sub.2] [less than] 0.
VAR: This is an additional variable for testing the price pressure hypothesis, and is equal to the daily common stock return variance over event days -200 to -51. According to Merton (1987) and Loderer, Cooney, and Van Drunen (1991), a higher return variance increases the compensation that risk averse investors require to hold more shares of the stock, implying a price decline for additional equity offerings. The size of the decline is directly related to the required increase in the expected return, suggesting [a.sub.3] [less than] 0.
OFFER: This variable is used to test the signaling effects caused by the size of the offer price. It is equal to the ratio of the offer price to the closing market price one month prior to the announcement day. As is in the case of Switzerland (and contrary to the US), the offer price is disclosed together with the size of the new issue on the announcement day. The lower the OFFER, the more negative the signal, implying [a.sub.4] [greater than] 0.
DTOA: This variable captures the wealth redistribution effects brought about by the rights issue. The DTOA is a dummy variable based on the debt-to-assets ratio. That is, DTOA = 0 for firms with a debt-assets ratio below the sample median debt-assets ratio, while DTOA = 1 for firms with a debt-assets ratio above the median. The debt of firms in the latter group carries a higher chance of default; thus, it gains more, relative to the former group, when the firm raises equity (and hence improves its financial leverage position). This implies that for issues of the second group, there is a stronger wealth transfer effect from shareholders to bondholders, suggesting a negative sign for [a.sub.5].
CONTR: This variable captures the information effects associated with the diffusion of ownership and is measured as the number of shareholders prior to the offering. By this definition of share ownership diffusion in rights issues, a small number of shareholders implies both a large fractional ownership and a large contribution of new capital by existing shareholders. Their decision to commit more personal wealth is considered to be good news about the firm's prospects. The smaller the number of individual contributors (i.e., low share ownership diffusion), the stronger the information effect should be, implying [a.sub.6] [less than] 0. Alternatively, as pointed out by Demsetz and Lehn (1985), the ownership concentration may tend to reduce potential problems within the firm and make asymmetric information less of a problem, implying also a negative sign for [a.sub.6].
LIQUID: This variable is a proxy of stock liquidity and is measured by the average daily trading volume of each sample firm's stock in the year prior to the offering, divided by the number of shares outstanding over the corresponding period. As suggested by Amihud and Mendelson (1986), investors will be willing to invest large amounts in illiquid assets only if appropriately compensated with a higher expected return. To the extent that the required discount increases with the size of the offering, stock prices should decline. In addition, equity offerings of liquid assets are predicted to have a smaller price effect relative to equity offerings of illiquid assets, implying [a.sub.7][greater than] 0.
MARKET: This variable controls the market conditions effect. It represents the cumulative average return of the stock market index over the period t = -50 to t = -1. Choe, Masulis, and Nanda (1993) show that announcements of equity issues convey less adverse information about equity values in periods with more promising investment opportunities. Consistent with their results, we should find a positive relation between the growth of the stock market index and the abnormal stock rerum on the announcement day, implying [a.sub.8] [greater than] 0.
S-P: This is a dummy variable dividing the sample firms into two groups, state-owned finns (S-P = 0) and non-state-owned firms (S-P = 1). In state-owned firms the continuous exposure to various political institutions, regulatory bodies, and the media is likely to result in partial anticipation of the offers, suggesting less of a surprise relative to the offers by non-state-owned firms, implying [a.sub.9] [greater than] 0.
B-R: This is a dummy variable separating registered shares (B-R = 0) from bearer shares (B-R = 1). Bearer shares are more liquid than registered shares, suggesting a smaller price decline relative to registered shares, thus [a.sub.10] [greater than] 0.
C. Test Results
Coefficient estimates for the regression model, where the dependent variable is the eleven-day cumulative average abnormal stock return CAR (-10,0) are presented in Table 3.
The regression model explains 26.21% of the variation in the dependent variable. The coefficient estimate of INVEST is positive and statistically significant at the 0.01 level. This finding cannot be explained by the Miller and Rock (1985) asymmetric information hypothesis. Instead, it provides empirical support to the hypothesis that rights offerings on the ASE are perceived as good news about the firms' future investment opportunities. That is, investors believe that firms issue the additional equity in order to finance new positive-NPV projects.
The coefficient of the variable SIZE is statistically insignificant at conventional levels. In addition, the coefficient of the variable VAR is positive and statistically significant at the 0.05 level, but it has the opposite sign from that predicted by the price pressure hypothesis. These two coefficients, therefore, provide no support to the price pressure hypothesis. These findings confirm the results reported by Scholes (1972), Marsh (1979), and Hess and Frost (1982), but not those reported by Asquith and Mullins (1986), Loderer and Zimmermann (1988), and Loderer, Cooney, and Van Drunen (1991).
The variable OFFER has a statistically insignificant coefficient. This implies that in rights issues shareholders are indifferent to the level of the subscription price. This finding provides no support to the theoretical argument of Heinkel and Schwartz (1986), and is inconsistent with the results reported in [TABULAR DATA FOR TABLE 3 OMITTED] Loderer and Zimmermann (1988). Similarly, the coefficient of the variable DTOA is statistically insignificant at conventional levels, providing no support to the wealth redistribution hypothesis.
The coefficient of the variable CONTR is negative and significant at the 0.10 level. Thus, firms with larger ownership concentration (i.e., fewer shareholders) are associated with higher abnormal returns. This finding supports the view that the market considers the large amount of additional capital committed by the few current shareholders as a strong positive signal about the firm's prospects. It is also consistent with the view, suggested by Demsetz and Lehn (1985), that equity offers by firms with large ownership concentrations are associated with less severe agency and asymmetric information effects.
The coefficient of the variables LIQUID is statistically insignificant at any conventional level. The variable MARKET has a positive coefficient, which is statistically significant at the 0.05 level. This finding is consistent with the evidence provided by Choe, Masulis, and Nanda (1993) and supports their argument that in rising stock markets, equity issues are associated with smaller adverse-selection effects. Regarding the two dummy variables S-P and B-R, their coefficients are statistically insignificant at any conventional level.
V. Summary and Conclusions
This paper investigates the response of stock prices to the announcement of rights offerings in Greece, and explores factors explaining the associated price reaction. The motivation for this study stems from the need to analyze the valuation effects of equity offerings in capital markets with entirely different institutional settings from those of the countries covered by most of the prior studies.
In Greece, the lack of an active secondary market for rights suggests large current-shareholder participation in the issue, thus reducing significantly the adverse-selection costs of the type described in Myers and Majluf (1984). Furthermore, rights offerings in Greece lead to much larger increases in the firms' capital relative to the corresponding increases reported in earlier studies. This has two important implications. First, rights offers substantially increase the supply of shares of common stock, making this sample ideal for testing the price pressure hypothesis. Second, they are associated with large infusions of capital, providing, therefore, a good sample to test for asymmetric information effects of the type analyzed in Miller and Rock (1985).
Furthermore, stock ownership in Greece is concentrated in the hands of a few shareholders, making both the agency and adverse information effects less severe relative to countries analyzed in prior studies. In addition, the lax financial reporting requirements (at least for the sample period), along with the lack of an active corporate bond market, make equity offerings the major credible outlet for the release of important information into the capital markets. In such an institutional setting, announcements of rights offerings are likely to be perceived as hard evidence of the firms' positive prospects, implying a positive stock-price reaction.
Contrary to the prediction of the Miller and Rock (1985) hypothesis, rights issues in Greece are associated with statistically significant positive abnormal stock returns, indicating that rights issue announcements are considered good news. This finding agrees with the evidence on rights issues in Korea, reported by Kang (1990), but differs from the evidence reported in prior studies for countries with developed capital markets.
I also furnish evidence that the abnormal returns are associated with the amount of capital raised relative to the existing capital, the degree of ownership concentration, and the market conditions prevailing prior to the rights issue announcement. These findings provide further support for the view that rights issues in Greece are perceived as conveying positive information about the firms' outlook.
The evidence, based on the coefficients of the variables that measure the stock return variance and the size of the issuing firm, provides no support to the price pressure hypothesis. In addition, the stock-price reaction is not related to the level of the subscription price, providing no support to the signaling effect analyzed in Heinkel and Schwartz (1986). Also, abnormal returns are not related to the issuing firms' debt-assets ratio, providing no supporting evidence for the existence of wealth redistribution effects. Nor do the findings reveal any relation between abnormal returns and the liquidity of the firms' stock, the type of firm (state-owned versus non-state-owned), or the type of stock (registered versus bearer).
These findings are important because they help explain why the stock price reaction to equity offerings varies according to issuing method, being negative for general cash offers and less negative to positive for rights issues. My results also offer a possible explanation for why abnormal returns associated with rights issue announcements vary across countries, being negative or zero in countries with developed capital markets and large ownership dispersion, but positive in countries with less developed capital markets and large ownership concentration.
I have benefited from long discussions with George J. Papaioannou and Nickolaos G. Travlos, detailed reports by two anonymous referees for this journal, and comments from Gikas Hardouvelis and Manolis Tsiritakis. This paper is based on my Ph.D. dissertation at the University of Piraeus, Greece. The assistance of my committee (P. Athanasopoulos, N. Traylos, G. Hardouvelis, G. Karathanassis, G. Diakogiannis, M. Glezakos, A. Merikas, M. Tsiritakis) is greatly appreciated. The Commercial Bank of Greece and Nickolaos G. Travlos generously provided the stock return data. Titan Cement Co. generously funded the acquisition of the S.A.S. program. Opinions expressed here do not represent those of the National Bank of Greece.
1 Firms listed on the Athens Stock Exchange (ASE) must use the rights issue method (except in some rare cases stipulated by law). However, they must use the general cash offer when they go public and list their shares on the ASE.
2 See Asquith and Mullins (1986), Kalay and Shimrat (1987), Masulis and Korwar (1986), and Mikkelson and Partch (1986). A review of the literature is provided by Smith (1986).
3 See Athens Stock Exchange Fact Book (1995).
4 See Provopoulos and Papadimitriou (1995).
5 Financial reporting requirements have become stricter since 1991, with the introduction of a new law (Law 1969/1991) which requires consolidated interim and annual financial statements and imposes stiff penalties for non-complying firms.
6 Travlos (1992) has developed a comprehensive databank of daily stock returns for the Athens Stock Exchange. In it, the stock prices have been adjusted so that they reflect dividend payments and capitalization changes.
7 See Travlos (1992).
8 For the estimation period (maximum observations 150) sample finns have, on average, 119 returns (minimum 48 and maximum 150). After eliminating the observations following the missing values, the average is 103 returns (minimum 31 and maximum 150). Similarly, for the period t=-50 to t=0 (maximum observations 51), the corresponding averages are 44 and 40. Interestingly, for the period t=-10 to t=0 (maximum observations 11), the associated averages rise to 10 and 9, suggesting either leakage of or trading on insider information.
9 The distribution of the 59 rights issues per year is (respective number of rights issues are in parentheses): 1985 (1), 1986 (2), 1987 (11), 1988 (7), 1989 (7), 1990 (31).
10 The choice of the market-adjusted model, instead of the risk-adjusted model, is due to the associated problems in estimating beta coefficients in a thin stock market such as the ASE. (See Papaioannou and Phillipatos (1982), Glezakos (1987), and Karathanassis and Philippas (1993).) Results obtained from applying the later methodology are qualitatively similar to those reported here.
11 Two other groups were also investigated: the sample of 38 common stock rights issues, and the sample of 31 "uncontaminated" common share rights offerings (i.e., a sample excluding those affected by announcements of other concurrent events, such as stock splits, stock dividends, earnings, etc.). The results from these samples are qualitatively similar to those presented here.
12 From a detailed investigation, the statistically significant average abnormal returns in days +7 and +10 resulted from extraordinarily high abnormal returns of specific stocks. For day +7, there are two outliers with associated abnormal returns of 11.68% and 17.95%, respectively. For day +10, there are four outliers with associated abnormal returns of 10.53%, 29.99%, 10.27%, and 13.80%, respectively.
13 Two additional models, Leland and Pyle (1977) and Myers and Maljuf (1984), also predict a negative stock reaction to equity offerings. In the former model, this effect is associated with the reduction of the original owners' fractional ownership of the firm. In the latter model, the negative price reaction is due to the potential wealth transfer from new to existing shareholders. In rights issues, the preemptive right of the existing shareholders keeps their fractional ownership intact, making these two models not applicable here.
14 The choice of this window stems from the existence (as shown in Table 2) of significant abnormal returns in several event days during the period t = -10 to t = 0.
15 Different versions of this variable were used where EI included only the new common stock issued. Also, in some versions, both EI and EO included common stock only. The results were similar to those reported here.
Amihud, Y. and H. Mendelson, 1986, "Asset Pricing and the Bid-Ask Spread," Journal of Financial Economics (December), 223-249.
Asquith, P. and D.W. Mullins, Jr., 1986, "Equity Issues and Offering Dilution," Journal of Financial Economics (January/February), 61-89.
Athens Stock Exchange Fact Book, ASE, 1995.
Brealey, R.A. and S.C. Myers, 1991, Principles of Corporate Finance, New York, NY, McGraw-Hill.
Brown, S.J. and J.B. Warner, 1980, "Measuring Security Price Performance," Journal of Financial Economics (June), 205-258.
Brown, S.J. and J.B. Warner, 1985, "Using Daily Stock Returns: The Case of Event Studies," Journal of Financial Economics (March), 3-31.
Choe, H., R.W. Masulis, and V. Nanda, 1993, "Common Stock Offerings Across the Business Cycle," Journal of Empirical Finance (???), 3-31.
Demsetz, H. and K. Lehn, 1985, "The Structure of Corporate Ownership: Causes and Consequences," Journal of Political Economy (December), 1155-1177.
Doukas, J. and N.G. Traylos, 1988, "The Effect of Corporate Multinationalism on Shareholder's Wealth: Evidence From International Acquisitions," Journal of Finance (December), 1161-1175.
Eckbo, B.E. and R.W. Masulis, 1992, "Adverse Selection and the Rights Offer Paradox," Journal of Financial Economics (December), 293-332.
Glezakos, M., 1987, "The Athens Stock Exchange: The Effect of Firm Specific Factors on The Stock Returns," University of Bath, England, Unpublished Dissertation.
Hansen, R.S., 1988, "The Demise of the Rights Issues," Review of Financial Studies (Fall), 289-309.
Heinkel, R. and E.S. Schwartz, 1986, "Rights Versus Underwritten Offerings: An Asymmetric Information Approach," Journal of Finance (March), 1-18.
Hess, A.C. and P.A. Frost, 1982, "Tests for Price Effects of New Issues of Seasoned Securities," Journal of Finance (March), 11-25.
Kalay, A. and A. Shimrat, 1987, "Firm Value and Seasoned Equity Issues, Price Pressure, Wealth Redistribution, or Negative Information," Journal of Financial Economics (September), 109-126.
Kang, H., 1990, "Effects of Seasoned Equity Offerings in Korea on Shareholder's Wealth," Pacific-Basin Capital Markets Research, S.G. Rhee and R.P. Chang, eds., North Holland, Elsevier Science Publishers B.V., 265-282.
Karathanassis, G. and N. Philippas, 1993, "Heteroscedasticity in the Market Model: Some Evidence from the Athens Stock Exchange," Managerial & Decision Economics (November/December), 563-567.
Kothare, M., 1991, "Equity Financing: An Investigation of the Rights Offer Anomaly," University of Texas, Working Paper (November).
Leland, H. and D. Pyle, 1977, "Informational Asymmetries, Financial Structure, and Financial Intermediation," Journal of Finance (June), 371-387.
Loderer, C., J.W. Cooney, and L.D. Van Drunen, 1991, "The Price Elasticity of Demand for Common Stock," Journal of Finance (June), 621-651.
Loderer, C. and H. Zimmermann, 1988, "Stock Offerings in a Different Institutional Setting: The Swiss Case," Journal of Banking and Finance (September), 353-378.
Marsh, P.R., 1979, "Equity Rights Issues and the Efficiency of the UK Stock Market," Journal of Finance (September), 839-862.
Masulis, R.W. and A.N. Korwar, 1986, "Seasoned Equity Offerings: An Empirical Investigation," Journal of Financial Economics (January/February), 91-118.
McConnell, J.J. and C.J. Muscarella, 1985, "Corporate Capital Expenditure Decisions and the Market Value of the Firm," Journal of Financial Economics (September), 399-422.
Merton, R.C., 1987, "A Simple Model of Capital Market Equilibrium With Incomplete Information," Journal of Finance (July), 483-510.
Michealy, R. and W.H. Shaw, 1995, "Does the Choice of Auditor Convey Quality in an Initial Public Offering?" Financial Management (Winter), 15-31.
Mikkelson, W.H. and M.M. Partch, 1986, "Valuation Effects of Security Offerings and the Issuance Process," Journal of Financial Economics (January/February), 31-60.
Miller, M. and K. Rock, 1985, "Dividend Policy Under Asymmetric Information," Journal of Finance (September), 1031-1051.
Myers, S. and N. Majluf, 1984, "Corporate Financing and Investment Decisions When Firms Have Information That Investors Do Not Have," Journal of Financial Economics (June), 187-221.
Nelson, J.R., 1965, "Price Effects in Rights Offerings," Journal of Finance (June), 650-657.
Papaioannou, G. and G. Phillipatos, 1982, "The Intervaling Effect Under Non-Synchronous Trading and Price Adjustment Lags in the Athens Stock Exchange," Spoudai (October-December), 627-639.
Provopoulos, G. and P. Papadimitriou, 1995, "The Bank and Non-Bank Debt of Greek Industrial Firms in the period 1989-1992," The Greek Financial System: Trends and Prospects, G. Provopoulos, ed., Foundation for Economic and Industrial Studies, Athens, 165-190.
Scholes, M.S., 1972, "Market for Securities: Substitution Versus Price Pressure and the Effects of Information on Share Prices," Journal of Business (April), 179-211.
Smith, C.W., Jr., 1977, "Alternative Methods for Raising Capital: Rights Versus Underwritten Offerings," Journal of Financial Economics (March), 273-307.
Smith, C.W., Jr., 1986, "Investment Banking and the Capital Acquisition Process," Journal of Financial Economics (December), 3-29.
Travlos, N.G., 1992, Athens Stock Exchange Data Bank of Risk and Returns of Stocks for the Period 1981-1990, Boston College, Working Paper.
White, R.W., and P.A. Lusztig, 1980, "The Price Effects of Rights Offerings," Journal of Financial and Quantitative Analysis (March), 25-40.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Special Issue: European Corporate Finance|
|Author:||Tsangarakis, Nickolaos V.|
|Date:||Sep 22, 1996|
|Previous Article:||Factors affecting the value of the stock voting right: evidence from the Swiss equity market.|
|Next Article:||Trade credit, product quality, and intragroup trade: some European evidence.|