國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
8
III. Hypotheses A. Market Timing Considerations for SEOs
In this section we lay out our studies by market timing hypothesis, which argues that managers exploit windows of opportunity to sell overvalued stock (Myers and Majluf, 1984; Clarke, Dunbar, and Kahle's 2004). To test the relative impact on the SEO decision of market timing, we examine the abnormal returns by the degree of industry wave. We argue that, firms will mimic their industry peers only in the market prospect well. The market timing implies the industry condition is permit (Loughran and Ritter, 1995; Baker and Wurgler, 2000). Firms will observe their counterparts to conduct SEO announcements and consider it is during SEO industry wave and make a decision to mimic in the following six months. If firms make SEO announcements by learning within the industry competitors, they will catch the timing of industry wave and experience lower negative SEO announcement returns.
We develop it as follows,
H1：The returns of SEO announcements will be less negative during the relatively
high industry SEO wave.
B. Measures of product market concentration and strategic interaction
As previous paper supports, concentration has an impact on industry strategic interaction. Ali et al. (2012) reflect that firms in more concentrated industry prefer
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
9
private placements, which have minimal SEC-mandated disclosure requirements, over seasoned equity offerings. Gomes and Philips (2007) mention that firms with greater information asymmetry are more likely to sell new shares via a private placement rather than through seasoned equity offerings. Bamber and Cheon (1998) argue that proprietary costs of disclosure are higher in more concentrated industries. Harris (1998) argues that firms in more concentrated industries are less competitive and firms in such industries disclose less to protect their abnormal profits. That‟s, we presume that firms in less concentrated industry have more motivation to learn from their industry peers, firms in less concentrated industries are more likely to conduct a SEO since industry is competitive. We conjecture that a less pronounced negative share price reaction in less concentrated industries. Our proxy measure for the degree of industry concentration is based on the Herfindahl Index. It is measured as the sum of the squares of market shares of all the firms in the particular industry for a fiscal year. We therefore propose the following hypothesis.
H2：The returns of SEO announcements should be more correlated among less
concentrated industry than among more concentrated industry.
According to Ali et al. (2012), firms in less concentrated industries have more incentive to conduct seasoned equity offerings. That‟s, we further presume that firms in less concentrated industry will conduct SEO during the industry wave to catch
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
10
market timing. Erwin and Miller (1998) show that rival firms experience significant negative stock price reactions on the day of a firm‟s repurchase announcements if the industry is concentrated.
This implies：
H3：The SEO industry wave is stronger when strategic interaction exists.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
11
IV. Data and Methodology
A. Data selection
The time period for this study ranges from 1990 to 2010. The Securities Data Corporation (SDC) Global New Issues Database dedicates the announcement of seasoned equity offerings.
For they must be the sample criteria：
(a) The issuer's stock must be listed on the New York Stock Exchange (NYSE), American Stock Exchange (Amex), or Nasdaq, and be presented on the University of Chicago Center for Research in Security Prices (CRSP) database for stock price and Compustat for accounting related data.
(b) In line with Loughran and Ritter (1995), only the first filing date of the same fiscal year is retained in the sample. In other words, if a firm issues multiple SEOs within the fiscal year, we keep only the first filing date in our sample. Subsequent issues are dropped. This is consistent with the regulation of U.S. Securities and Exchange Commission as well. There are a few safe harbors that offer some level of comfort by assuring non-integration of certain exempt offerings separated by at least six months from another offering.1
1 Securities Act Regulation D [17 C.F.R.230.502]; Securities Act Rule 147 [17 C.F.R.]
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
12
(c) SEO firms should be a non-financial, non-utility firm. Financials and utilities are removed since these firms‟ incentive to issue equity (for example,
banks may issue equity to meet regulatory capital requirements) as well as the applicability of multiple based valuations to these firms are not comparable to other firms. American Depository Receipts (ADRs), private placements, rights offers, unit offers, closed-end funds, Real Estate Investment Trusts (REITs) are excluded as well.
The original raw data contains 13,247 samples from SDC. We exclude 148 offers with lack of filing date and 1,709 samples with minor exchanges. 3,604 financial services firms and utilities samples are excluded as well. Lastly, we drop 2,152 offers that occur within a year of the same firm‟s seasoned issue, which results in a final sample of 5,634 seasoned equity offers.
Table 1 outlines our final sample of SEOs by year. The sample is observed in 1996 with highest frequency of 490 firms and the lowest in 1990, which has 89 observations.
B. Variables selections
To start with, we construct the INDSEOWAVE term to measure the industry intensity of SEO activity. The INDSEOWAVE is computed from SDC data, and it is
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
13
the sum of the number of SEOs announcements, excluding those of the firm in question, that occur in the same three-digit SIC code industry for all the other firms in the previous six months. The wave term is commonly documented in M&A and equity activity ( See Massa et al., 2007; Bradley and Yuan , 2013).
All of SEO samples are ranked in terms of concentration, they are classified in high (low) concentration industries are identified by sorting all the SEO samples within industries each year.
We capture the Conc/SEOs interaction term to investigate the strategic interaction between industry wave and concentration. The SEOs term is brought by the INDSEOWAVE as mentioned before. This Conc/SEOs interaction is a proxy to measure the effect on the probability of a SEO announcement resulting from an increase in both the concentration of the industry and the intensity of the SEOs intra-industry activity.
The CRSP-Compustat Database is dedicated for all the accounting variables that have been used as controls. We choose the Herfindahl-Hirschman Index as our measure of concentration, which is the sum of the squared fraction of industry sales by all firms in the 3-digit SIC code for the year prior to the announcement. The Herfindahl index scores rank is expected to be negative. This indicates that firms in less concentrated industries outperform than those in more concentrated industries.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
14
Following Clarke and Kahle (2004), RUNUP defines the issuer‟s buy-and-hold return over the year prior to the issue. ROA is earnings divided by total assets (AT) in the year before the SEO filing date. lnTA is the logarithm of total assets (AT) in the year before the SEO filing date. lnMKT is the natural logarithm of the issuer‟s market capitalization (shares outstanding multiplied by price per share) on the announcement
date, which measures a firm‟s competitive position in the market place. We get it from a firm‟s total sales in the prior year. We expect this variable to be positively related to firm‟s cumulative abnormal returns should be larger for firms with a greater market
presence. Price is the SEO offer price. Tobin‟s Q is calculated as the sum of market capitalization and total assets minus total common equity, scaled by total assets.
lnTOBIN is the natural log of Tobin's Q, We expect the natural log of Tobin‟s Q is
positively related to SEO returns.
Table 2 provides summary statistics of all SEO firms. Panel A compares high wave and low wave firms. Panel B compares high concentration and low concentration firms.
C. Market reactions
We examine the market reaction to the announcement of an SEO by event study method. We compute abnormal returns using the market model,
𝑅𝑖𝑡 = 𝛼𝑖+ 𝛽𝑖𝑅𝑚𝑡+ 𝑒𝑖 (1)
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
15
where 𝑅𝑖𝑡 is the return of firm i at time t, 𝛼𝑖𝑡 is the average abnormal return of firm i, and 𝑅𝑚𝑡 is the value-weighted market return index. Abnormal returns 𝐴𝑅𝑖𝑡 are
calculated as,
𝐴𝑅𝑖𝑡 = 𝑅𝑖𝑡 − (𝛼̂𝑖+ 𝛽̂𝑖𝑅𝑚𝑡) (2)
Where 𝛼̂𝑖 and 𝛽̂𝑖 are estimated from Eq. (1). The cumulative abnormal return (CAR) from firm i is calculated as follows：
𝐶𝐴𝑅𝑖 = ∑𝑇𝑖=1𝐴𝑅𝑖𝑡 (3)
The average cumulative abnormal returns for the time period (-1,+1) window relative to the announcement event for which the abnormal return is being measured.
Our estimation period is -250 to -11 trading days before the SEO event.
The regression is presented as follows,
CAR(−1, +1) = α + 𝛽1𝑅𝑈𝑁𝑈𝑃 + 𝛽2𝑅𝑂𝐴 + 𝛽3𝑙𝑛TA + 𝛽4𝑙𝑛MKT + 𝛽5𝐻𝐻𝐼 + 𝛽6𝑃𝑟𝑖𝑐𝑒 + 𝛽7𝑙𝑛TOBIN + 𝛽8𝐼𝑁𝐷𝑆𝐸𝑂𝑊𝐴𝑉𝐸 + 𝛽9𝑇𝑂𝑇𝐴𝐿𝑆𝐸𝑂 + 𝛽10 𝐶𝑜𝑛𝑐/𝑆𝐸𝑂𝑠
‧
Table 1 Number of seasoned equity offerings (SEOs) by year
This table provides the distribution of seasoned equity offers (SEOs) reported by SDC Database from 1990 to 2010. We report the mean (median) values of industry SEO wave (INDSEOWAVE) in each fiscal year.
Year Number of
‧
Table 2 Univariate tests for analyzing the sample using changes in operating performance measures
This table provides summary statistics and degree of wave (concentration). Panel A compares firms in high industry SEO wave and low industry SEO wave.
Panel B compares firms in more concentrated and low concentrated industries. If the industry SEO wave in the particular industry wave is greater (small) than the median, the industry is classified as high-wave (low-wave).
Panel A：Univariate tests for analyzing of SEOs firms in high and low wave
Variable High Wave Low Wave P-value
Panel B：Univariate tests for analyzing of SEOs firms in high and low concentration
Variable High Concentration Low Concentration P-value
(differences in means)
‧
Table 3 Market reaction to SEO announcement
This table presents the market reaction to SEO announcements. Panel A reports average cumulative abnormal returns across three different event windows of (-1,+1), (-3,+3) and (-9,+9) days relative to the date of the SEO announcement (0) by the degree of SEO industry wave and industry concentration. The market model with returns from trading day -250 to trading day -11 are used to estimate CARs. Panel B reports the strategic interaction between industry wave and industry concentration. Difference in test is presented in last low (column). t-statistics are provided in parentheses. ***, **, and * indicate statistical
significance at 1%, 5%, and 10% levels, respectively.
Panel A：Univarite tests for market reaction
(-1,+1) (-3,+3) (-9,+9)
Panel B： Interaction between Wave and Concentration to SEO announcement
(1) High wave (2) Low wave
Observations CAR(-1,+1) Observations CAR(-1,+1) (1)-(2)
(3) High concentration 878 -2.78%**
(-2.02)
1817 -3.18%**
(-1.99)
0.40%**
(-2.03)
(4) Low concentration 2181 -2.15%
(-0.61)
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
19
V. Empirical results
A. Results of Multiple Regressions
We start by report univariate results to test whether firms observe the industry wave and conduct SEO announcements in the following six months. The estimation period is -250 to -11 trading days before the SEO event, average cumulative abnormal returns (CARs) across event windows from day (-1,+1), (-3,+3) and (-9,+9) around announcements of seasoned equity offerings. We use the CRSP value-weighted market return as our market index. The results reports comparative average cumulative abnormal returns (CARs) for SEO firms in High wave/ Low wave and High concentration/ Low concentration for the separate event windows following the SEO announcements as shown in Table 3, we find that firms experience lower negative returns during high industry wave. On average, 0.39% negative returns in one day event windows disappear during the high industry wave. Firms experience lower SEO negative day returns during higher industry wave. Firms in less concentrated industries also have lower negative announcements than those in more concentrated industries. Our results are consistent with the view that there exists mimicking wave in SEO announcements.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
20
Panel B reports the interaction effect between SEO industry wave and concentration. Since corporate policy is more observed in less concentrated industries, we argue that, the mimicking behavior is more easily observed in less concentrated industries. The Conc/SEOs interaction term is more positively significant to enhance the SEO industry wave in less concentrated industries. That‟s, firms in less concentrated firms is more likely to follow to conduct SEOs when they observe their industry peers negative abnormal return is lower than average. They expect to positively benefit to get lower day negative returns as well by getting the industry SEO wave.
We further test whether this findings hold in multivariate setting by analyzing the one day abnormal returns event window on a number of control variables considers previously. Our results are consistent with the view that SEO announcements content about industry prospects. This confirms the intuition of market timing hypothesis.
B. Robustness tests
Although our evidence in Table 5 lends supports for the view that SEO announcements have important industry-specific mimicking behaviors. We further awaken interests by industry characteristics. Different to the aforementioned section, we try to find firms conduct SEO exists in some particular industries. We argue that
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
21
mimicking behavior do exist in some particular industries. SEO announcements exist frequently in particular industries as showed in Table 4.
To check the robustness of our results, we try to classify our samples into industries based on four-digit SIC codes as in Fama and French (1997) instead.
TOTALSEO is classified by four-digit SIC code defined as the number of firms within
the given industry in the fiscal year. INDSEOWAVE is defined as the sum of the number of SEO announcements, excluding those of the firm in question, that occur in the same four-digit SIC code industry for all the other firms in the previous six months.
Our analysis in Table 6 suggests the results remains consistent with our earlier studies. The findings support the market timing hypothesis and provide strong evidence the strategic interaction do exists in different industry classification. Firms are likely issuing equity to catch the SEO industry wave and strategic interaction enhances the industry SEO wave.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
22
Table 4 Distribution of Offering by Industry
Firms are classified into industries based on four-digit SIC codes as in Fama and French (1997).
Industry Number of Firms Industry Number of Firms
Business Services 655 Machinery 116
Pharmaceutical products 358 Restaurants, Hotel, Motel 110
Electric Equipment 298 Transportation 110
Protroleum and Natural Gas 231 Chemicals 71
Retail 229 Measuring and Control Equip 66
Telecommunications 180 Steel works, Etc. 53
Medical Equipment 169 Construction Materials 50
Computers 162 Entertainment 50
Healthcare 136 Construction 49
Wholesale 119 Other 497
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
23
Table 5 Results of Multiple Regressions in the same three-digit SIC code industry
This table presents the firm‟s decision to conduct SEOs using multiple regressions. The dependent variable is defined as SEO abnormal returns across event window form day -1 to day +1 around announcements on seasoned equity offerings. Model 1 and Model 2 present INDSEOWAVE and TOTALSEO respectively that industry wave do have an impact on firm‟s SEO decision. In Model 3 and Model 4, we take the Herfindahl index into considered to examine industry concentration measure do enhances the industry wave. Model 5 investigates cumulative abnormal returns during the SEO industry wave. In Model 6 we further examine the strategic interaction between the industry wave and concentration. t-statistics are provided in parentheses. ***, **, and * indicate statistical significance at 1%, 5%, and 10% levels.
‧
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
25
Table 6 Robust tests in Fama and French industry classification
This table presents the firm‟s decision to conduct SEOs using multiple regressions. The dependent variable is defined as SEO abnormal returns across event window form day -1 to day +1 around announcements on seasoned equity offerings. Model 1 and Model 2 present INDSEOWAVE and TOTALSEO respectively that industry wave do has an impact on firm‟s SEO decision. In Model 3 and Model 4, we take the Herfindahl index into considered to examine industry concentration measure do enhances the industry wave. Model 5
investigates cumulative abnormal returns during the SEO industry wave. In Model 6 we further examine the strategic interaction between the industry wave and concentration. t-statistics are provided in parentheses. ***, **, and * indicate statistical significance at 1%, 5%, and 10% levels.
‧
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
27
VI. Conclusion
In this study, we examine how the degree of product market competition affects the firm‟s decision to SEO announcements. According to previous literature, SEO announcements send a negative signal about poor investment opportunities prospect. We argue that, in the case of strategic interaction between firms, SEO announcement acquires a mimicking dimension.
Our results provide two important findings. We find that SEOs negative day returns will lower during the industry wave exists. On average, 0.39% negative returns in one day event windows disappear during the high industry wave.
Consistent with market timing hypothesis, firms follow the industry wave will perform better on their SEO abnormal returns. The other results also support industry concentration brings strategic interaction, which enhances the firm‟s mimicking behavior. These results are also robust in different industry classification. We confirm this intuition by showing that SEO firms in less concentrated industries outperform the market.
Our results blends corporate finance decision and industrial organization.
The study gives a few crumbs of information about mimicking SEO behavior.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
28
We hope provide some new ways regarding why firms tend to cluster their SEOs and why they observe SEOs happening in waves.
‧
Ali, Ashiq, Klasa, Sandy, Yenug, Eric, 2012, Industry concentration and corporate disclosure policy, SSRN elibrary.
Asquith, and David W. Mullins, 1986, Equity issues and offering dilution, Journal of Financial Economics 15, 61-89.
Bamber, L., Cheon, Y., 1998, Discretionary management earnings forecast disclosures:
antecedants and outcomes associated with forecast venue and forecast specificity, Journal of Accounting Research 36, 167-190.
Baker, M., Wurgler, J., 2002, The equity share in new issues and aggregate stock returns, Journal of Finance 55, 2219-2258.
Benveniste, Lawrence M., Walid Y Busaba, and William Wilhelm, 2002, Information externalities and the role of underwriters in primary equity markets, Journal of Financial Intermediation 11, 61-86.
Boone, A.L.,Ivanov,V., 2012, Bankruptcy spillover effects on strategic alliance partners, 2012, Journal of Financial Economics 103, 551-569.
Bradley, D., Yuan, X., Information spillovers around seasoned equity offerings, 2013, Journal of Corporate Finance 21, 106-118.
Brewer III, Elijah, William E. Jackson III, Julapa A. Jagtiani, and Thong Nguyen, 2000, The price of bank mergers in the 1990s, Economic Perspectives 24, 2-23.
Burns, Natasha, and Ivonne Liebenberg, 2011, U.S. Takeovers in foreign markets: Do they impact emerging and developed markets differently?, Journal of Corporate Finance 17, 1028-1046.
Clarke, Jonathan, Craig Dunbar, and Kathleen Kahle, 2004, The long-run performance of secondary equity issues: A test of the windows of opportunity hypothesis, Journal of Business 77, 575-603.
Dasgupta, P. and Stiglitz, J., 1980, Uncertainty, Industrial Structure and the Speed of R&D, Bell Journal of Economics 11, 1-28.
DeAngelo, Harry, Linda DeAngelo, and René Stulz, 2010, Seasoned equity offerings, market timing, and the corporate life cycle, Journal of Financial Economics 95, 275-295.
Demsetz, Harold, 1973, Industry structure, market rivalry and public policy, Journal of Law and Economics 16, 1-9.
Erwin, Gayle R., and James M. Miller, 1998, The intra-industry effects of
open-market share repurchases: contagion or competitive? The Journal of Financial Research 21, 389-406.
Funke, Christian, Timo Gebken, Lutz Johanning, and Gaston Michel, 2008,
Information signaling and competitive effects of M&A: long-term performance of rival companies, SSRN eLibrary.
‧
Gomes, A., Philips, G.M., 2007, Private and public security issuance by public firms:
The role of asymmetric information. Working Paper, University of Maryland.
Harris, M. H., 1998, The association between competition and managers‟ business segment reporting decisions. Journal of Accounting research 36, 111-128.
Hertzel, M. G., Li, Z., Officer, M. S., Rodgers, K.J., 2006, Inter-firm linkages and the wealth effects of financial distress along the supply chain. Journal of Financial Economics 87, 374-387.
Ibbotson, Roger G., 1975,“Hot Issue” Markets, Journal of Finance 30, 1027-1042.
Jorion, Philippe, and Gaiyan Zhang, 2007, Good and bad credit contagion: Evidence from credit default swaps, Journal of Financial Economics 84, 860–883.
Lang, Larry H.P. and Rene M. Stulz, 1992, Contagion and Competitive Intra-industry Effects of Bankruptcy Announcements：An Empirical Analysis. Journal of
Financial Economics 32, 45-60.
Loughran, Tim, and Jay Ritter, 1995, The new issues puzzle, The Journal of Finance 50, 23-52.
Loughran, Tim, and Jay Ritter, 1997, The operating performance of firms conducting seasoned equity offerings, Journal of Finance 52, 1823-1850.
Massa, Massimo, Zahid Rehman, and Theo Vermaelon, 2007, Mimicking repurchases, Journal of Financial Economics 84, 624-666.
Masulis, and Ashok N. Korwar, 1986, Seasoned equity offerings: An empirical investigation, Journal of Financial Economics 15, 91-118.
Myers, Stuart, and Nicholas Majluf, 1984, Corporate financing and investment decisions when firms have information that investors do not have, Journal of Financial Economics 13, 187-221.
Ritter, J., 1984, The „hot issue‟ market of 1980, Journal of Business 57, 215-240.
Rajan, Raghuram G, 1994, Why Bank Credit Policies Fluctuate: A Theory and Some Evidence. The Quarterly Journal of Economics 109, 399-441.
Rajan, Raghuram G, 1994, Why Bank Credit Policies Fluctuate: A Theory and Some Evidence. The Quarterly Journal of Economics 109, 399-441.