University of Amsterdam, Amsterdam Business School MSc Business Economics, Finance track
Master Thesis
Market Reaction to Seasoned Equity Offerings;
Issue Method and Issuer Size effect
Matay Gabraail (6136176) March 2015
Liang Zou
Abstract
In this study I found evidence that the stock market reaction to news of seasoned equity offerings by companies is explained by the issue size. Using a sample of announcements made in the Netherlands between 2000 and 2015 I found a significant negative relation between market reaction and issue size. No significant relation between market reaction and issue method is found in this sample. These results are compatible with most existing literature.
2
Statement of Originality
This document is written by Student Matay Gabraail who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
3
Table of Contents
1. Introduction ... 4 2. Literature Review ... 7 2.1 Theoretical Framework ... 7 2.2 Existing Literature ... 9 3. Methodology ... 144. Data and Descriptive Statistics ... 18
4.1 Data ... 18 4.2 Descriptive Statistics ... 20 5. Results ... 22 6. Conclusion ... 26 7. References ... 28 8. Appendix ... 31
4
1. Introduction
Most companies that enter the market through an initial public offering (IPO) issue equity in the same market in a later stadium, for all kinds of reasons. One third of companies that enters the market through an IPO returns to the market within five years to issue more equity. When they do so, the companies issue on averge three times as much capital as they did with the IPO (Berk and DeMarzo, 2007).
There has been a lot of discussion about the seasoned equity offerings (SEOs) and their implications for investors. Several studies (Asquith and Mullins, 1986; Schipper and Smith, 1986; Masulis and Korwar, 1986; Eckbo and Masulis, 1992; Slovin et al., 1994) found evidence that stock prices fall by 3 percent after companies announce a seasoned equity offering. These studies report that between 70 and 80 percent of companies that announces a seasoned equity offering experience a negative market reaction. Other capital raising instruments, like raising debt in the capital markets, show no significant negative abnormal return for the issuing company.
Researches have tried to find an explanation for this fall in stock prices after announcing seasoned equity offerings. One of the hypotheses that may explain this downfall is set up by the theoretical framework of Myers and Majluf (1984). They state that there is information asymmetry between managers and investors and that a seasoned equity offering signals information (information content hypotheses). In the perception of investors, managers that announce a seasoned equity offering are valuing the company below the market price, since stock prices will fall. A manager that thinks the company is undervalued will not issue additional equity, since the market price will then drop even more below the value of the company known by managers. However, in their analysis Myers and Majluf (1984) assume that the equity issue is targeted at new investors, no findings are reported in the situation of a rights offer where the stock is offered to existing stockholders.
5 An important reason for this difference in underwritten equity offerings and rights offerings is the wealth transfer. In case of a rights offering, where all current investors exercise their right to buy additional shares, there is no wealth transfer effect. The same investors with the same relative investment hold. This arises from the information asymmetry between management and outside investors. Investors assume that management acts in the interest of existing shareholders (Myers and Majluf, 1984).
Burton, et all. (2010) studied the relationship between market reaction and rights offers, but used a sample from the United Kingdom based seasoned equity issue’s. They find that rights offers cause a significant negative market reaction in the announcement period. There is no literature that is compatible with these findings and it is in contrast to the evidence of the information content hypotheses by Myers and Majluf (1984) and Sant and Ferris (1994).
Although there is a lot of evidence of a downfall in stock prices, attempts (Masulis and Korwar, 1986; Mikklesen and Partch, 1986; Sant and Ferris, 1994) to find factors that explain these significant market reactions after announcing a seasoned equity offer are far less successful. In this research I attempt to find an explanation for this downfall. Thereby looking at the issue method and announced issue size.
Reason to research the issue method is the above mentioned wealth transfer effect. Consequently, stock price effects after announcing a rights offer cannot be attributed to the information asymmetry effect, since managers are assumed to act in the interest of existing shareholders. Isolating the information asymmetry effect in case of rights offerings, makes them ideal to see whether it is the information asymmetry that causes share prices to fall after announcing an equity offer (Tsangarakis, 1996).
The reason to research the announced issue size gives insight in the information asymmetry effect. Managers know better than investors, whether a company is overvalued or not. When managers decide to issue equity, since the company is overvalued, the more the announced issue size is the more it is assumed that the company is overvalued (Myers and Majluf, 1984).
6 Therefore, the research question is ‘ Are the issue method and the issue size explaining the negative market reaction when announcing a seasoned equity offering? ’. Based on prior research and the theoretical framework of Myers and Majluf (1984), the first hypotheses is that the announcement of a rights offers do not cause market to react negatively, since managers are assumed to act in the interest of their investors. The second hypotheses is that the announced issue size causes the market to react negatively, since there is asymmetric information and it signals that the more managers want to issue, the more a company is overvalued.
For this study I hand collected a sample of Dutch based companies that announced a seasoned equity offering in the period between January 2000 and December 2014. The reason that this time frame is chosen is because the used database (Bloomberg) collected the first Dutch seasoned equity offering in January 2000 and at the time of this research there were no seasoned equity offerings announced after December 2014.
This study proceeds as follows. Section 2 provides an overview of the previous literature in this area of research. Section 3 describes the used method of this research. Section 4 describes the data and its characteristics. Section 5 shows the results of this study and relates them to the existing literature. Finally, section 6 concludes.
7
2. Literature Review
2.1 Theoretical Framework
Myers and Majluf (1984) provide a theoretical framework for almost all the research that is done in this area. In their study they present a model where the managers have superior information about a certain investment decision. Their model has six very interesting properties.
In the first property Myers and Majluf (1984) state that when a company needs capital, safe securities should be preferred in contrast to more risky securities. For external capital companies should go to bond markets, but equity should be raised by retention. That is, when companies need external financing they should issue debt instead of equity.
The second property in Myers and Majluf (1984) study states that companies should not participate in good investment opportunities if these investment opportunities outstrip operating cash flows and the companies have no ability to issue low risk debt, rather than issue risky securities to finance the investment opportunity. Therefore, stockholders are better off, on average, when the company has sufficient cash (or near-cash) and/or spare debt capacity (financial slack) available to take up opportunities as they appear. As the size of the required equity offering increases the loss in value increases. Therefore, the loss in value increases with an increase in the required investment or a decrease in available financial slack. Whereas, in economic literature the theory states that when the risk about a certain investment opportunity is reduced or when the investment opportunities’ expected NPV is increased the loss decreases.
Their third property is about building up financial slack. Companies should not pay any dividends when the required investments are modest to build up financial slack. The saved cash should be invested in marketable securities or should be reserved to increase borrowing power. In addition, another way to build up financial slack is to issue equity when managers have no or
8 little information advantage; Companies that do not have sufficient financial slack to cover possible investment opportunities in the future should issue equity when managers have no or little information advantage (Myers and Majluf, 1984).
Myers and Majluf (1984) suggest a model where the estimate of asset value by managers is highly correlated with a change in dividends. Managers use their superior information to send a signal to the market. But, the fourth property states that managers should not pay dividends if the funds are not present; companies should not issue equity or other risky security to pay dividends.
Property five of Myers and Majluf (1984) states that when managers issue equity to finance an investment opportunity with superior information in comparison to investors, stock prices will fall, all other things equal. Nevertheless is this investment opportunity undertaken in the interest of the existing stockholders. However, if companies issue safe debt in the bond market to finance an investment opportunity, stock prices will not fall.
The last interesting property in their study Myers and Majluf (1984) that I want to highlight is that in the case of a merger, with a company with no or little financial slack and a company with much financial slack, the value of the combined company increases. However, there are some difficulties in negotiating such a merger. The only way this can succeed is when the company with no or little financial slack can convey their superior information to the potential buyer, otherwise the company with no or little financial slack will be bought in the open market by making offers directly to the stockholders.
The above mentioned properties are dependent on the assumptions made by Myers and Majluf (1984) and may not hold in another context, therefore a lot of research has been done in this area based on the theoretical framework of this article. I summarize these in the next subsection.
9 2.2 Existing Literature
Several studies (Asquith and Mullins, 1986; Schipper and Smith, 1986; Masulis and Korwar, 1986; Eckbo and Masulis, 1992; Slovin et al., 1994) report that an announcement of seasoned equity offerings by companies causes a 3 percent fall of stock prices on average. In all cases, evidence from this literature suggests that between 70 and 80 percent of the sample experience negative abnormal returns when companies report news about a SEO. Whereas announcing other new financing issues, such as straight debt and straight preference there is no significant market reaction.
Although these findings are very consistent, attempts (Masulis and Korwar, 1986; Mikklesen and Partch, 1986; Sant and Ferris, 1994) to find factors that explain these significant market reactions after announcing a seasoned equity offer are far less successful.
In their study Masulis and Korwar (1986) find a significant negative abnormal return after announcing an equity offering which is larger for industrials than for public utilities. However, they find that the market reaction is not affected by the announced purpose of the equity offer.
Whereas, Mikkleson and Partch (1986), who analyze the effect of various types of financing events in their sample with a constant set of companies on stock price effects, find a significant negative abnormal return in stock price on average when the companies in their sample announce an equity offer in common stock or convertible debt. They find no significant price reaction when companies announce an offer of preferred stock, straight debt, private placement of debt and term loans. Credit agreements however have a significant positive abnormal return on stock price.
An important contribution of Mikkleson and Partch (1986) is their analysis of abnormal price returns in stock prices in the announcement period for completely underwritten equity offerings. Their results show no correlation between stock price and either the rating of debt, or the relative size of the offer. Their findings are in line with the predictions of Myers and
10 Majluf (1984) that the significant market reaction during the announcement period is higher for offerings that are intended to finance new capital expenditures than for offerings that are intended to be used to retire existing debt.
Their results are also in line with the pecking order theorem. Where market participants assume that an offer in common stock or convertible debt is a signal of that the market price is too high (Mikkleson and Partch, 1986).
In contrast to both, Masulis and Korwar (1986); Mikklesen and Partch (1986), Sant and Ferris (1994) find that an announcement of an seasoned equity issue that is intended to fund capital expenditures or retire existing debt leads to significant negative abnormal returns. However, an announcement of an seasoned equity issue that is intended to expand employee and management stock options schemes generally leads to a significant increase in stock price. Sant and Ferris (1994) tested three hypotheses to find an explanation for the negative abnormal returns during the announcement period of a seasoned equity offer.
The first hypotheses to be tested to explain the negative market response followed by the announcement of an equity issue is the capital structure. The capital structure of a company is the mix of the long-term debt, specific short-term debt, common equity and preferred equity; how a company finances its overall operations and growth by using different sources of fund (Sant and Ferris, 1994).
The second hypotheses, price pressure argument, is about the supply and demand. It assumes that the demand curve is downward sloping; an increase in the supply of a certain stock decreases their price. This is however in contrast with the classical position. The classical position assumes that there are no arbitrage opportunities and that near perfect substitutes are always available. Therefore, when a company decides to issue equity, their stock price and the stock price of a substitute must be the same. This means that the demand curve for a certain stock is horizontal (Sant and Ferris, 1994).
11 Their last hypotheses to explain the negative market response to an announcement of an equity issue is the information content argument. This argument assumes that there is a difference in information about the value of a certain company known between managers and investors (Sant and Ferris, 1994). This last hypotheses is in line with Myers and Majluf (1984), who state that an equity issue signals information about the value of a certain company that is known by the managers. Therefore, when the market value is underpriced managers will not issue equity, whereas, when the market value is overpriced managers tend to issue equity.
Sant and Ferris (1994) find no significant evidence for the first two hypotheses, capital structure and price pressure. However, they find significant evidence for the third hypotheses, information content. Their study suggests that there are different market reactions to different purposes of an equity issue. They have obtained evidence for a significant relations between abnormal returns and the future growth of cash flows. Therefore, Sant and Ferris (1994) state that the negative abnormal returns are explained by differences in the expectation of future cash flows.
The reason that the difference in the method (rights1 and non-rights) have rarely been
studied in prior articles is that most studies are based on the United States and almost all United States stock issues are fully underwritten2.
However, Eckbo and Masulis (1992) have studied the announcement period market reaction for different equity issue methods. They find negative abnormal returns for all equity issue methods, but not all methods show significant negative market reactions. The underwritten equity offers show the strongest significant negative relation with abnormal
1 A rights issue is an equity offer made to the existing stockholders. With this right existing stockholders have a right (no obligation) to buy additional stocks. (Melissa Beck, associate, in the Capital Markets Group of Morrison & Foerster LLP, 2014).
2 Underwriting is a task of investment banks, where they raise investment capital on behalf of their client. This means that the risk of distributing the securities lies with the investment bank. Should the investment bank not be able to find enough investors, they have to hold the securities their self (Wertheimer, 2006).
12 returns (equal to -3.43 percent). Whereas, rights offers have an insignificant negative relation with abnormal returns (equal to -1.49 percent).
An important reason for this difference in underwritten equity offerings and rights offerings is the wealth transfer. In case of a rights offering, where all current investors exercise their right to buy additional shares, there is no wealth transfer effect. The same investors with the same relative investment hold. This arises from the information asymmetry between management and outside investors. Investors assume that management acts in the interest of existing shareholders (Myers and Majluf, 1984). Consequently, stock price effects after announcing a rights offer cannot be attributed to the information asymmetry effect, since managers are assumed to act in the interest of existing shareholders. Isolating the information asymmetry effect in case of rights offerings, makes them ideal to see whether it is the information asymmetry that causes share prices to fall after announcing an equity offer (Tsangarakis, 1996).
Myers and Majluf (1984) provide a theoretical framework for the study done by Eckbo and Masulis (1992) and also justifies the research done in different equity issue methods; equity issues offered to existing stockholders and the capital market overall. As stated before Myers and Majluf (1984) suggest that an equity issue signals information about the value of a certain company that is known by the managers. Therefore, when the market value is underpriced managers will not issue equity, whereas, when the market value is overpriced managers tend to issue equity. However, in their analysis Myers and Majluf (1984) assume that the equity issue is targeted at new investors, no findings are reported in the situation of a rights offer where the stock is offered to existing stockholders.
Burton, et all. (2010) also studied the relationship between market reaction and rights offers, but used a sample from the United Kingdom based equity issue’s. However, they find that rights offers cause a significant negative market reaction (-6.23 percent) in the announcement period. There is no literature that is compatible with these findings and it is in
13 contrast to the evidence of the information content hypotheses by Myers and Majluf (1984) and Sant and Ferris (1994).
This study aims to find evidence for the relation between the equity offer method and the market reaction and the announced equity offer size and the market reaction. Main attempt is to see whether Dutch based evidence provides support for Myers and Majluf’s (1984) theoretical framework.
14
3. Methodology
From the Corporate Action Calendar in Bloomberg is going to be drawn a sample of Dutch seasoned equity offers. This sample starts in January 2000 and ends in December 2014. Bloomberg lists rights issues and other type of seasoned equity offers in the Corporate Action Calendar.
After all the companies that are suited for this research are collected, for each of these companies I hand collected the return on the announcement day for each of these companies and calculate the expected return. Abnormal returns for each company are then calculated. The following equation is used:
𝐴𝑅𝑖𝑡= 𝑅𝑖𝑡–𝐸(𝑅𝑖𝑡) (1)
where 𝐴𝑅𝑖𝑡 is abnormal return on stock i in time period t, the announcement day; 𝑅𝑖𝑡 is actual
return on stock i in time period t, the announcement day; and 𝐸(𝑅𝑖𝑡) is expected return on
stock i in time period t, the announcement day.
Before I can calculate the abnormal returns of the company stocks on the announcement day, I need to calculate the expected return on the announcement day. Therefore I need the historical Alpha and Beta of each stock. These are hand collected from Bloomberg. Thereafter I can calculate the expected returns by using the conventional market model. The following equation is used:
𝐸(𝑅𝑖𝑡) = 𝛼𝑖 + 𝛽𝑖(𝑅𝑚𝑡) + 𝜀𝑖 (2)
where 𝛼𝑖 is a constant term for stock i; 𝛽𝑖 is the sensitivity (slope coefficient) of the returns on stock i to the returns on the market; 𝑅𝑚𝑡 is the return on the market in time period t, and 𝜀𝑖 is the random error term.
15 The conventional market model parameters are hand collected from Bloomberg using the historical intercept and beta, which means that for every company the beta and intercept are calculated from all the market returns and all the company returns known by Bloomberg. The abnormal returns are calculated for the announcement day for every company that announced an equity issue, regardless of the type of issue (except IPO’s which are not a part of this research). Every Equity issue in the period from January 2000 to December 2014 is collected and there has been made a difference between rights offers and non-rights offers. The period is chosen simply because before January 2000 there were no Dutch rights offers in the data set of Bloomberg, December 2014 is chosen because at the date of collecting data no rights offers were announced in 2015 (known by Bloomberg).
As stated in the literature part Myers and Majluf (1984) suggest that an equity issue signals information about the value of a certain company that is known by the managers. Therefore, when the market value is underpriced managers will not issue equity, whereas, when the market value is overpriced managers tend to issue equity, because it is assumed that managers act in the interest of existing stockholders. However, in their analysis Myers and Majluf (1984) assume that the equity issue is targeted at new investors, no findings are reported in the situation of a rights offer where the stock is offered to existing stockholders.
Eckbo and Masulis (1992) state that there might be an announcement size effect on negative abnormal returns. Which is in line with Myers and Majluf (1984) and Sant and Ferris (1994) who mention the before explained information content or signaling argument, from which can be concluded that in the case of a large overpriced value the managers tend to issue more equity.
Whereas, Burton, et all. (2010) find that rights offers cause a significant negative market reaction (-6.23 percent) in the announcement period. There is no literature that is compatible with these findings and it is in contrast to the evidence of the information content hypotheses by Myers and Majluf (1984) and Sant and Ferris (1994).
16 Therefore, to find whether it is managers that act in interest of existing stockholders and there is an information content argument in the Dutch market, thereby looking at the announcement issue method and announcement issue size in relation to abnormal returns the following regression, which also is used by Burton, et all. (2010) in the United Kingdom market, is going to be made:
𝐴𝑅𝑖= 𝛼𝑖 + 𝛽1𝑅𝐼𝐺𝐻𝑇𝑆𝑖 + 𝛽2𝐶𝑂𝑆𝐼𝑍𝐸𝑖+ 𝛽3𝐴𝑁𝑁𝑆𝐼𝑍𝐸𝑖+ 𝜀𝑖 (3)
where 𝐴𝑅𝑖 is the abnormal return for company i on the announcement day, calculated from equations 1 and 2. 𝑅𝐼𝐺𝐻𝑇𝑆𝑖 is a dummy variable that is 1 for rights and 0 for non-rights. 𝐶𝑂𝑆𝐼𝑍𝐸𝑖 is the size of company i based on the market capitalization. 𝐴𝑁𝑁𝑆𝐼𝑍𝐸𝑖 is the
announcement size of the seasoned equity offer, since announcement size and company size are correlated, I divide announcement size by market value at the announcement day, creating a relative announcement size offer. 𝜀𝑖 is a random error term.
Based on economic theory and the theoretical framework of Myers and Majluf (1984) this regression should find no significant negative abnormal return for rights offers, in contrast to Burton, et all. (2010). For the announcement size however, economic theory suggests that the relative announcement size should give a significant negative abnormal return for the issuing company, this is in line with Myers and Majluf (1984); Sant and Ferris (1994) and Eckbo and Masulis (1992).
Besides this basic regression, the same regression including control variables is going to be made. This regression is the same as formula 3 but including the following control variables:
VAR: This is the return variance of firm i, and is equal to the stock variance on return over the period 200 to 51 days before the announcement. This causes a price decline for equity offerings, because a higher variance on the return of firm i increases the compensation which requires risk averse investors to hold more stock of the firm (Merton, 1987 and Loderer, et all., 1991). The price fall after announcing an equity offer is related to the
17 required increase in the expected return of the stock, therefore I expect this variable to have a negative sign.
OFFER: This variable is the offer price as a fraction of firm i's stock price. It is calculated by dividing the offer price of an equity announcement by the closing price one trading day before the announcement. A low OFFER signals a low value for the firm, this suggests a negative sign for this
variable.
DTA: This is a dummy variable that captures the debt to assets ratio of the firm. The dummy variable is set 0 if the debt to assets ratio is smaller than 1 and is set 1 if the debt to assets ratio is larger than 1. Economic theory suggests that in general firms are healthy if the debt to assets ratio is smaller than 1 and the chance of default is small. However, this is dependent on the specific industry the firm operates in. In this case, since there are different companies in different industries, the general guideline is chosen. Because a higher debt to assets ratio increases the possibility of default, I expect this variable to have a negative sign.
LIQUID: This is a measure of the firm i’s stock liquidity. This variable is calculated by the average daily trading volume divides by the total number of outstanding stock. Amihud and Mendelson (1986) suggest that investors only invest in illiquid stock if they are compensated with a higher expected return. Therefore I expect this variable to have a positive sign.
These variables are included in an additional regression because economic theory suggests they have an effect on stock prices (Amihud and Medelsom, 1986; Merton, 1987; Loderer, et all., 1991; Tsangarakis, 1996).
18
4.
Data and Descriptive Statistics
4.1 Data
To do the previous described regression I had to find a sample of Dutch equity offerings. This sample is collected from Bloomberg. The sample starts in January 2000 and ends in December 2014, simply because Dutch rights offers are available in this time span in Bloomberg. In the Corporate Action Calendar available in Bloomberg are listed all the equity offerings, rights and non-rights. To filter for the rights offerings I added the search criteria to only equity as the asset type and rights offers in the action type in the Corporate Action Calendar.
All the non-right offerings were also found in the Corporate Action Calendar in Bloomberg, also after selecting asset type equity as a search criteria, but I had to choose action type IPO/Addl to find all the known equity offerings, including IPO’s. I had to delete all the IPO’s to have left only the SEOs. In the downloaded file information was found about the date, announcement date and the size of the offer.
After I found all the companies that did either a rights offer or a non-rights offer in the selected period I had to calculate all the abnormal returns. I started by calculating all the expected returns by the conventional market model (equation 2). But before I could calculate these expected returns I had to hand collect all the intercepts and historical beta’s. These are listed in Bloomberg after searching for every company ticker in combination with the command beta. I listed these beta’s and intercepts in table 4. in the appendix.
Once I had calculated all the expected returns I hand collected all the actual returns for the companies at the specific announcement date and for the AEX at the specific announcement date in Bloomberg. Hereafter I applied equation 1 to find all the abnormal returns.
19 To do the regression I was left with collecting the company size. The company size is hand collected by searching for the company ticker and give the command financial analyses, where after I searched for the market capitalization the day before the announcement day (t-1).
The different control variables are also collected from Bloomberg.
The variance is calculated by collecting the returns for the different companies 200 to 51 days before the announcement date.
The offer price was for a lot of companies included in the initially downloaded file. The OFFER variable is calculated by dividing the offer price by the closing price, which are hand-collected from Bloomberg, one trading day before the announcement day.
Bloomberg lists debt to asset ratios for a lot of companies. I had to search for the individual company at a certain time (before the announcement date) and collect these.
For the liquidity variable I had to collect information on average trading volume and total outstanding stock for each individual company at a certain time (before the announcement date) and collect these to calculate the LIQUID variable.
20 4.2 Descriptive Statistics
Since the announcement size and the company size are positive correlated I had to find a way to eliminate this multicollinearity. This is done by dividing the announcement size by the market capitalization (company size), as mentioned before. See the table 1. below.
Table 1. Correlation
Correlation between company size and announcement size / relative announcement size
Co. Size Ann. Size Rel. Ann. Size
Co. Size 1,0000
Ann. Size 0,3826 1,0000
Rel. Ann. Size -0,1373 0,1938 1,0000
As can be seen in the above table the correlation between company size and announcement size goes down from 0.3826 to -0.1373 after dividing the announcement size with market capitalization (company size). Indicating that it is better to use the relative announcement size than the absolute value of this variable.
Table 2. below lists some descriptive statistics about the different used variables. Panel A describes all the dependent and independent variables. Abnormal returns range between -34.26 and 12.62 percent. The maximum positive value is surprising since it is quite high and economic theory suggests a decrease in stock price after announcing an equity offer. This might be explained by the research done by Sant and Ferris (1994), who have obtained evidence for a significant relations between abnormal returns and the future growth of cash flows.
Company size ranges from 2.1 million to 130.8 billion, with a mean of 5.4 billion, because all Dutch companies that have issued equity in the sample period are enlisted.
21
Table 2. Summary statistics
Panel A - Summary statistics for dependant and indepandant variables
N Mean Std. Dev. min max
Abnormal returns 203 -2,5596 5,2711 -34,26 12,62
Rights Dummy 203 0,2167 0,4130 0 1
Co. Size 202 5409,7 12773,0 2,1 130823,5
Rel. Ann. Size 190 0,1919 0,3830 0,0036 4,2857
Debt to asset Ratio (%) 148 91,02 31,59 52,18 137,90
Offer Price Ratio (%) 102 62,68 20,71 19,69 116,80
Panel B - Summary statistics for abnormal returns divided by rights and non - rights
N Mean Std. Dev.
Rights 44 -4,0516 7,9746
Non-rights 159 -2,1467 4,1741
Panel C - Summary statistics for relative announcement size divided by rights and non - rights
N Mean Std. Dev.
Rights 35 0,4683 0,7781
Non-rights 155 0,1295 0,1583
Panel B describes the differences in abnormal return for rights offers and non-rights offers. Rights offers have a mean abnormal return (-4,0516) lower than non-rights offers (-2,1467), but the standard deviation of the rights offers (7,9746) is higher than for non-rights offers (4,1741).
Panel C describes the same as Panel B but for the relative announcement size, the differences in relative announcement size for rights offers and non-rights offers. Rights offers have a mean (0,4683) almost four times greater than non-rights offers (0,1295), indicating that rights offers are relatively larger than non-rights offers. The standard deviation varies even more, for rights offers (0,7780) it is more than five times more than non-rights offers (0,1583).
22
5. Results
The main findings that have emerged from this research of the relation between market reaction to equity announcements and issue method and announcement size is that the market reacts with the announcement size. Table 3. Shows that this is the case for both the announcement size (1) and the relative announcement size (2). In both cases there is a negative value, indicating that the higher the (relative) announcement size is the stronger the market will react negatively, this is in line with property 2 in the study of Myers and Majluf (1984). In both cases there is a significance of less than 1 percent.
However, these results present no significant relationship between market reaction and the issue method. In both cases there is a negative value indicating that an announcement of a rights offer is perceived negative by existing stockholders, but there is no significant evidence to conclude that.
Company size is significant at the five percent level in regression (1), this is probably the case because it correlates with announcement size. There is no economic theory that suggests a relation between company size and market reaction in case of a seasoned equity offer announcement. Therefore regression (2) is explains the situation better. The adjusted R² confirms this.
The main findings, the negative relation between market reaction and announcement size, is explained by economic theory. Sant and Ferris found evidence for their information content argument. This argument assumes that there is a difference in information about the value of a certain company known between managers and investors (Sant and Ferris, 1994). Therefore, investors relate the announcement size to a signal. This explains the findings of this research.
23
Table 3. Estimated Coefficients
This table represents the estimated coefficients and t-statistics (in parentheses) from regressing Abnormal Return on a constant and the variables RIGHTS, COSIZE, ANNSIZE, VAR, OFFER, DTO and LIQUID for equity issues in the Netherlands between 2000 and 2014. These results show that in all three regressions the relative announcement size is significant at the 1 percent level. The rights dummy variable has a negative sign, but is not significant in any of the cases. Therefore, the
announced size of an equity offer has a significant effect on abnormal returns on the announcement day. (1) (2) (3) Constant -1,6481*** -0,1824*** -0,2186 (-4,29) (-4,45) (-0,57) Rights dummy -1,2012 -0,6345 -0,3122 (-1.51) (-0,72) (-0,34) Co. size 0,0506** -0,0101 0,0224 (1,98) (-0,41) (0,59) Ann. size -0,0022714*** (-5,35)
Rel. Ann. Size -2,641592*** -0,2766***
(-2,94) (-3,19)
Variance -2,1167**
(-2,15)
Offer price ratio -0,0746
(0,71)
Debt to assets ratio -0,0751
(-1,73) Liquidity 17,2629 (1,07) R²adj. 0,1477 0,2117 0,2622 N 190 190 102
Notes: *** (**) Indicates signifcance at the 1% (5%) level on the basis of two-tailed testing.
24 Also, these findings are line with Myers and Majluf (1984), who state that an equity issue signals information about the value of a certain company that is known by the managers. Therefore, when the market value is underpriced managers will not issue equity, whereas, when the market value is overpriced managers tend to issue equity. The announcement size signals how much a company is overvalued in case of an equity issue.
However, Burton, et all. (2010) did not find a significant announcement size effect on market reaction in the United Kingdom. Their results showed a negative relation between announcement size of an equity issue and market reaction, but no significant one.
These results did not find a significant relation between issue method and market reaction. The results present however a negative sign, which is in line with the evidence found by Eckbo and Masulis (1992), who studied the announcement period market reaction for different equity issue methods. They found negative abnormal returns for all equity issue methods, but not all methods show significant negative market reactions. The underwritten equity offers show the strongest significant negative relation with abnormal returns. Whereas, rights offers have an insignificant negative relation with abnormal returns.
In contrast to Burton, et all. (2010). They found that rights offers cause a significant negative market reaction in the announcement period. There is no literature that is compatible with these findings and it is in contrast to the evidence of the information content hypotheses by Myers and Majluf (1984) and Sant and Ferris (1994).
In the third regression there are added some control variables that have economic relevance for the return on a certain stock (Amihud and Medelsom, 1986; Merton, 1987; Loderer, et all., 1991; Tsangarakis, 1996). The relative announcement size shows also a 1 percent significance in this third regression. The same explanations apply as in the first two regressions.
The control variable variance has a negative value and shows a significance of 5 percent. This suggests that the higher the variance of a certain return on a stock is the lower the
25 abnormal return is in case of an equity offer, which is in line with the studies done by Merton (1987) and Loderer, et all. (1991). The three other control variables, offer price ratio; debt to asset ratio; and liquidity, show no significant effect on abnormal returns in case of an equity offer.
From this research can be concluded that in this sample the announcement size of an equity issue is an important factor in the price fall of stock prices. Managers that are aware of this should think in the interest of their investors and may not want to issue too much equity if the risk exists that the stock prices fall under the market value.
26
6. Conclusion
This study researched seasoned equity offerings. There has been a lot of discussion about the seasoned equity offerings and their implications for investors. Several studies (Asquith and Mullins, 1986; Schipper and Smith, 1986; Masulis and Korwar, 1986; Eckbo and Masulis, 1992; Slovin et al., 1994) found evidence that stock prices fall by 3 percent after companies announce a seasoned equity offering.
Although these findings are very consistent, attempts (Masulis and Korwar, 1986; Mikklesen and Partch, 1986; Sant and Ferris, 1994) to find factors that explain these significant market reactions after announcing a seasoned equity offer are far less successful.
Therefore, I tried to answer the following research question ‘ Are the issue method and the issue size explaining the negative market reaction when announcing a seasoned equity offering? ’. Based on prior research and the theoretical framework of Myers and Majluf (1984), the first hypotheses, derived in the introduction, is that the announcement of a rights offers do not cause market to react negatively, since managers are assumed to act in the interest of their investors. The second hypotheses, also derived in the introduction, is that the announced issue size causes the market to react negatively, since there is asymmetric information and it signals that the more managers want to issue, the more a company is overvalued.
For this study I hand collected a sample of Dutch based companies that announced a seasoned equity offering in the period between January 2000 and December 2014. The reason that this time frame is chosen is because the used database (Bloomberg) collected the first Dutch seasoned equity offering in January 2000 and at the time of this research there were no seasoned equity offerings announced after December 2014.
The main findings that have emerged from this research is that the market reacts with the announcement size. The higher the (relative) announcement size is the stronger the market
27 will react negatively, this is in line with property 2 in the theoretical framework of Myers and Majluf (1984) and the information content argument of Sant and Ferris (1994). I found a significance less than 1 percent.
However, these results present no significant relationship between market reaction and the issue method. There is a negative value for the rights dummy, indicating that an announcement of a rights offer is perceived negative by existing stockholders, but there is no significant evidence to conclude that, which is in line with the evidence found by Eckbo and Masulis (1992). The negative sign is inconsistent with Myers and Majluf (1984) and Sant and Ferris (1994). Burton, et all. (2010) found a significant negative abnormal return for rights offers, but is incompatible with the information content hypotheses and prior research in the area.
From this research can be concluded that in this sample the announcement size of an equity issue is an important factor in the price fall of stock prices. Managers that are aware of this should think in the interest of their investors and may not want to issue too much equity if the risk exists that the stock prices fall under the market value.
There are some limitations in this research. First of all, I included only three independent variables, like the model of Burton, et all. (2010), to find whether it is rights issues that cause market to react negatively and if announcement size explains why markets react negatively. There may be some other independent variables that contain information or explain why markets react negatively on a seasoned equity offer announcement.
Second, this research has been done in the Netherlands with Dutch data, whereas most research about the negative market reaction after announcing a seasoned equity offering are in the United States. Therefore, it is important to see whether the same conclusions can be drawn from United Stated data.
28
7. References
Amihud, Y. and H. Mendelson, 1986, Asset Pricing and the Bid-Ask Spread, Journal of Financial Economics (December), 223-249.
Asquith, P. and Mullins Jr., D.W. (1986) Equity issues and offering dilution,
Journal of Financial Economics, 15, 61–89.
Beck, M. (2014) associate in the Capital Markets Group of Morrison & Foerster LLP. Berk, J. B., & DeMarzo, P. M. (2007). Corporate finance. Pearson Education.
Bhagat, S. (1983). The effect of pre-emptive right amendments on shareholder wealth. Journal
of Financial Economics, 12(3), 289-310.
Brealey, R.A. and S.C. Myers, 1991, Principles of Corporate Finance, New York, NY, McGraw-Hill. Burton, B. M., Alasdair Lonie, A., & Power, D. M. (1999). Does the issue method influence the market reaction to seasoned equity offer announcements?. Applied Economics Letters,
6(7), 459-462.
Cronqvist, H., & Nilsson, M. (2005). The choice between rights offerings and private equity placements. Journal of Financial Economics, 78(2), 375-407.
Hess, A.C. and P.A. Frost, 1982, Tests for Price Effects of New Issues of Seasoned Securities, Journal of Finance (March), 11-25.
Eckbo, B.E. and Masulis, R.W. (1992) Adverse selection and the rights offer paradox, Journal of Financial Economics, 32, 293–332.
Kim, E. H., & Purnanandam, A. (2014). Seasoned equity offerings, corporate governance, and investments. Review of Finance, 18(3), 1023-1057.
29 Loderer, C. and H. Zimmermann, 1988, Stock Offerings in a Different Institutional Setting: The Swiss Case, Journal of Banking and Finance (September), 353-378.
Loughran, T., & Ritter, J. R. (1995). The new issues puzzle. The Journal of finance, 50(1), 23-51. Masulis, R. and Korwar, A. (1986) Seasoned equity offerings: an empirical investigation, Journal of Financial Economics, 15, 91–118.
Merton, R. C. (1987). A simple model of capital market equilibrium with incomplete information. The journal of finance, 42(3), 483-510.
Mikkleson, W.H. and Partch, M.M. (1986) Valuation effects of security offerings and the issuance process, Journal of Financial Economics, 15, 21–60
Myers, S.C. and Majluf, N.S. (1984) Corporate financing and investment decisions when have information that investors do not have, Journal of Financial
Economics, 13, 187–222.
Sant, R. and Ferris, S.P. (1994) Seasoned equity offerings: the case of all-equity Ž firms, Journal of Business Finance and Accounting, 21, 429–44.
Armitage, S. (1998). Seasoned equity offers and rights issues: a review of the evidence. The
European Journal of Finance, 4(1), 29-59.
Schipper, K. and Smith, A. (1986) A comparison of equity carveouts and seasoned equity offerings: share price effects and corporate restructuring, Journal of
Financial Economics, 15, 153–86.
Spiess, D. K., & Affleck-Graves, J. (1995). Underperformance in long-run stock returns following seasoned equity offerings. Journal of Financial Economics, 38(3), 243-267.
Slovin, M.B., Sushka, M.E. and Bendeck, Y.M. (1994) Seasoned common stock issuance following an IPO, Journal of Banking and Finance, 18, 207–26.
30 Tsangarakis, N. V. (1996). Shareholder wealth effects of equity issues in emerging markets: Evidence from rights offerings in Greece. Financial Management, 21-32.
Wertheimer, E. (2006) Underwriting: The poetics of Insurance in America, Stanford University
Press, 1722-1872.
White, H. (1980) A heteroskedasticity-consistent covariance matrix estimator and a direct test for heteroskedasticity, Econometrica , 48, 817–38.
31
8. Appendix
Table 4. Beta's and intercept
Hereunder are lined up the beta's and intercepts (alpha's) that are used to calculate the expected return.
Company Raw Beta Adjusted Beta Alpha (intercept)
Aalberts Industries NV 1,030 1,020 0,197 Accell Group 0,736 0,824 -0,017 Aegon NV 1,080 1,054 0,110 Air France-KLM 0,889 0,926 0,063 Akzo Nobel NV 1,073 1,049 0,034 AM BV 0,471 0,648 0,037
AMG Advanced Metallurgical Group NV 0,811 0,874 0,080
Amsterdam Molecular Therapeutics 0,709 0,806 -1,470
Arcadis NV 0,795 0,864 0,091 ArcelorMittal 1,213 1,142 -0,413 ASML Holding NV 0,936 0,958 0,304 Ballast Nedam NV 0,826 0,884 -1,100 BinckBank NV 0,766 0,844 -0,167 Boskalis Westminster NV 0,923 0,949 0,093 Corio NV 0,672 0,781 0,202 Corporate Express NV 1,131 1,087 -0,021 Crucell NV 0,727 0,818 0,375
Danone Baby and Medical Nutrition BV 0,695 0,797 0,219
DE Master Blenders 1753 NV -0,041 0,306 0,103 Delta Lloyd NV 0,635 0,756 0,086 DNC de Nederlanden Co NV 0,293 0,529 0,523 DPA Group NV 0,220 0,480 0,185 Draka Holding BV 0,939 0,960 0,093 Eriks NV 0,476 0,651 0,117 Eurocommercial Properties NV 0,579 0,719 0,206 EVC International NV 0,571 0,714 -0,071 Exact Holding NV 0,048 0,365 0,693 Fugro NV 0,636 0,757 -0,369 Getronics NV 1,110 1,073 0,154
32
Company Raw Beta Adjusted Beta Alpha (intercept)
Grontmij 0,567 0,711 -0,034 Hagemeyer NV 0,919 0,946 0,007 Heijmans NV 1,179 1,119 0,257 Heineken NV 0,853 0,902 -0,049 ING Groep NV 1,491 1,327 0,371 Innoconcepts NV 0,635 0,757 -0,077 Kardan NV 1,599 1,399 -0,707 KAS Bank NV 0,367 0,578 0,185 Kendrion NV 0,669 0,779 0,157 Koninklijke Ahold NV 0,920 0,947 0,049
Koninklijke BAM Groep NV 1,671 1,447 -0,020
Koninklijke KPN NV 0,727 0,818 0,549
Koninklijke Porceleyne Fles NV 0,289 0,526 0,531
Koninklijke Ten Cate NV 0,645 0,763 -0,060
Koninklijke Vopak NV 0,693 0,796 -0,102
Lavide Holding NV -0,197 0,202 -0,424
LBi International NV 0,753 0,835 0,488
Macintosh Retail Group NV 0,649 0,766 -1,192
Nielsen Co BV/The 0,881 0,920 0,166 NSI NV 0,600 0,733 -0,350 Nutreco NV 0,477 0,651 0,197 OCI NV 0,530 0,687 0,340 OctoPlus NV 0,635 0,757 0,299 Ordina NV 1,019 1,013 0,141 Pharming Group NV 0,262 0,508 1,736 PinkRoccade NV 0,890 0,927 0,180 PostNL NV 0,945 0,963 0,669 RBS Holdings NV 1,072 1,048 0,106 Reed Elsevier NV 0,848 0,899 0,310 Rodamco Asia NV 0,228 0,485 0,203
Rodamco Europe NV/Netherlands 0,270 0,514 0,178
Roodmicrotec NV 1,051 1,034 0,237
Royal Dutch Petroleum Co 0,672 0,781 0,077
33
Company Raw Beta Adjusted Beta Alpha (intercept)
Royal P&O Nedlloyd NV 0,683 0,789 0,117
Samas NV 0,453 0,636 -0,476
SBM Offshore NV 1,241 1,161 -0,386
Scala Business Solutions NV 1,035 1,023 0,152
Simac Techniek NV 0,543 0,695 0,579
SnowWorld NV 0,141 0,428 0,314
SNS REAAL NV 1,532 1,355 -0,561
Stern Groep NV 0,097 0,398 -0,069
Super de Boer NV 0,466 0,644 -0,093
Tele Atlas Survey BV 0,249 0,499 0,436
Tele2 Netherlands Holding NV 0,938 0,959 0,037
TKH Group NV 0,835 0,890 0,204 TNT Express NV 0,978 0,986 -0,288 TomTom NV 0,995 0,997 0,539 Unit4 NV 0,576 0,717 0,692 Univar NV 0,408 0,606 0,720 USG People NV 1,498 1,332 0,215
Van der Moolen Holding NV 0,966 0,977 -0,141
Van Lanschot NV 0,646 0,764 0,110
Vastned Offices/Industrial NV 0,458 0,638 -0,045
