The Lifecycle of Takeover Defenses
William C. Johnson
Sawyer Business School
Suffolk University
Boston, MA 02108
wcjohnson@suffolk.edu
Jonathan M. Karpoff
Foster School of Business
University of Washington
Seattle, WA 98195
karpoff@uw.edu
Sangho Yi
Sogang Business School
Sogang University
Seoul, South Korea
yisangho@sogang.ac.kr
This version: December 8, 2015
Preliminary – Please do not circulate without author consent.
Abstract:
Previous studies on the economic consequences of takeover defenses report inconsistent results. For
instance, Gompers, Ishii, and Metrick (2003) and Bebchuk, Cohen, and Ferrell (2009) report that takeover
defenses destroy firm value while Smith (2013) and Johnson, Karpoff, and Yi (2015) report exactly the
opposite results. Furthermore, previous studies fail to explain why takeover defenses are a prevalent
corporate phenomenon if they are value-destroying. We develop a lifecycle view of takeover defenses and
use this new paradigm to resolve the inherent conflict within the extant literature. Our lifecycle view of
takeover defenses suggests that at the early stage of a firm’s life the bonding benefit of takeover defenses
as suggested in Johnson, Karpoff, and Yi (2015) dominate the costs. But as the firm matures, the
managerial moral hazard costs increase and the average takeover defense becomes value-decreasing. Our
empirical results support a life cycle view of takeover defenses, suggesting that contradicting results
within the prior literature are the result of researchers not considering the life cycle effect of takeover
defenses. Our life cycle view of takeover defenses also explains why a value-maximizing firm may adopt
takeover defenses early in its life while takeover defenses are value destroying for large mature firms.
JEL classification: G34, K22, L14
Keywords: Anti-takeover provisions, takeover defenses, IPO, bonding
1
I. Introduction
Takeover defenses are a prevalent corporate phenomenon and the economic consequences of
takeover defenses have been thoroughly examined in the extant literature. However, in spite of
numerous previous studies which are focused on the economic consequences of takeover
defenses and ample evidence showing the relationship between takeover defenses and firm value
and various mechanisms intermediating between them, the current state of the literature is
problematic for two reasons.
First, previous studies report inconsistent results. For instance, Gompers, Ishii and Metrick
(2003), Masulis, Wang, and Xie (2009) and Bebchuk, Cohen and Ferrell (2009) report that
takeover defenses are negatively associated with firm value. These results support the classical
agency view of takeover defenses which suggests that takeover defenses entrench managers and
allow them to seek their private benefits at the expense of firm value (Easterbrook and Fischel,
1991). On the other hand, contrary to the prediction of the classical agency view of takeover
defenses, Field and Karpoff (2002) and Johnson, Karpoff and Yi (2015) show that IPO firms in
which managerial moral hazard problem is negligible due to highly level of managerial
ownership adopt a significant number of takeover defenses. Furthermore, Field and Karpoff
(2002) show that IPO firms with a larger number of takeover defenses tend to show better ex
post operating performance. In addition, Johnson, Karpoff and Yi (2015) and Smith (2013) show
that the number of takeover defenses is positively associated with the firm value at the IPO stage.
Thus, it is clear that there is a contradiction in the literature regarding whether takeover defenses
increase or decrease firm value.
Second, the extant literature fails to answer the very basic question of why firms adopt
takeover defenses. Gompers, Ishii and Metrick (2003), Masulis, Wang, and Xie (2009) and
Bebchuk, Cohen and Ferrell (2009) suggests that takeover defenses aggravate managerial moral
hazard problem and destroy firm value. If takeover defenses are value destroying, why would
value-maximizing firms adopt takeover defenses which entrench managers thus destroying firm
value?1
1
Takeover defenses may provide benefits such as alleviating managerial myopia problem (Stein, 1988; Harris, 1990)
and improving the bargaining position of target firm managers (Grossman and Hart, 1980; Linn and McConnell,
1983; DeAngelo and Rice, 1983). These studies explain why value maximizing firms may want to adopt takeover
defenses but cannot explain why takeover defenses are negatively associated with firm value.
2
In this paper, we aim at reconciling these conflicting empirical results by developing and
investigating a life cycle view of takeover defenses. For the first time in the literature, we
examine the evolution of takeover defenses over a firm’s life cycle and the life cycle dynamics of
the benefits and costs of takeover defenses. The previous literature examining the economic
consequences of corporate takeover defenses employs a cross-sectional approach which
eliminates the time dimension from empirical analyses.2 We argue that the evolution of takeover
defenses over a firm’s life cycle underlies the seeming inconsistencies documented in the extant
literature. Our life cycle view of takeover defenses is as follows: The costs of corporate takeover
defenses start at a negligible level at the early stage of a firm’s life cycle and increase as a firm
grows. On the other hand, the benefits of corporate takeover defenses start large at the early
stages of a firm’s life cycle and decline as the firm grows. Thus, the net benefit of takeover
defenses declines over a firm’s life. Our life cycle view of takeover defenses predicts that young
firms have a positive relationship between firm value and takeover defenses but the sign of this
relationship reverses and there is a negative relationship between firm value and takeover
defenses for mature firms, which we call a value reversal. In our view, the conflicting results
reported by previous studies is not a puzzle but only highlights the importance of incorporating
the dynamics of takeover defenses over a firm’s life cycle into the empirical analysis.
Our life cycle view of takeover defenses is hinted at in some previous studies. The classical
agency view suggests that costs of corporate takeover defenses are smaller at the earlier stage of
a firm’s life but grow steadily over a firm’s life cycle as managerial ownership diffuses and
founders are likely to leave the firm. In addition, as firms get larger, managers have greater
incentives to extract higher rents by their opportunistic behavior. On the other hand, Johnson,
Karpoff and Yi (2015) note that the benefit of corporate takeover defenses is biggest at the early
stage of a firm’s life cycle and declines as a firm grows. At the early stage of a firm’s life, the
firm usually needs stakeholder investment in highly specific assets, which makes the firm’s
opportunistic behavior to extract quasi-rents more likely. Furthermore, a young firm does not
2
In addition, the prior literature (Gompers, Ishii, and Metrick (2003) and Bebchuk, Cohen, and Ferrell (2009)) have
almost exclusively utilized data concerning large public companies from the ISS dataset of large public firms. In
contrast, other papers such as Smith (2015) uses data from Voting Analytics where Johnson, Karpoff, and Yi (2015)
utilizes data collected from the SEC filings of the firms. The former dataset is dominated by large companies that
are largely in the S&P 1500 where the latter datasets will likely include a large number of firms early in their life
cycle. This may help to explain the difference in findings between these two sets of papers..
3
have a long history over which to build a reputation to honor stakeholders’ implicit claims. This
increases the value of corporate takeover defenses as a commitment device to honor these
stakeholders’ implicit claims. Furthermore, a young firm typically lacks the resources to
construct diffuse distribution channels and instead relies heavily on a few customers and
suppliers, which gives large customers and suppliers significant bargaining power over a young
firm.3
Aside from the implications of the previous studies, our investigation of a life cycle view of
takeover defenses is also motivated by anecdotal evidence which suggests that when a firm
adopts takeover defenses it considers its current lifecycle stage. For instance, the prospectus4 of
Advantage Learning System states that:
“The provisions of the Company's Articles of Incorporation and By-Laws
and the WBCL described in this section may delay or make more difficult
acquisitions or changes of control of the Company not approved by the
Company's Board of Directors. Such provisions have been implemented
to enable the Company, particularly (but not exclusively) in the initial
years of its existence as a publicly-traded company, to develop its
business in a manner which will foster its long-term growth without
disruption caused by the threat of a takeover not deemed by its Board of
Directors to be in the best interests of the Company and its shareholders.”
If takeover defenses enhance firm value early in the life cycle of the firm and destroy value
later, then a value-maximizing firm would adopt more takeover defenses at the earlier stage of its
life when the net benefit of takeover defenses is greater and remove them drastically at the later
stage of its life when net benefit of takeover defenses becomes negative. But, this is not the case.
On the contrary, a firm tends to adopt a menu of takeover defenses at the IPO stage and seldom
revise them as it matures (Coates, 2001). We confirm this stagnation of corporate takeover
3
Consistent with this argument, Johnson, Kang, and Yi (2006) find that 60% of IPO firms disclose a large customer
at the time of the IPO of a firm while Cen, Dasgupta, and Sen (2011) find that 41% of firms have a large customer
as mature firms. The aforementioned results indicate that a firm’s reliance on large customers decreases and its
customer-base diversifies over its life cycle.
4
http://www.sec.gov/Archives/edgar/data/1030484/0000950131-97-001443.txt
4
defenses throughout our empirical results. There are many potential reasons why corporate
takeover defenses are insensitive to the changes in various factors which affect the benefits and
costs of takeover defenses. First, managers may prohibit firms from removing takeover defenses
to protect themselves from disciplinary outside takeover attempts. Second, as Coates (2001)
suggests, the power of managers who try to retain takeover defenses to entrench themselves is
exactly offset by the power of institutional shareholders which try to remove value-destroying
takeover defenses. Third, according to Johnson, Karpoff and Yi (2015), corporate takeover
defenses are used as a contractual commitment device to persuade stakeholders to make valuable
relationship-specific investments. The irreversible nature of such a comment device may lead to
a difficulty in modifying corporate takeover defenses at a later stage of a firm’s life. This
suggests that rather than takeover defenses being dynamically modified to maximize firm value
at all times, the level of takeover defenses may stagnate, resulting in a value-destroying level of
takeover defenses being in force as the firm matures.
The declining net benefit of corporate takeover defenses over a firm’s life cycle and the
stagnation of these defenses over a firm’s life cycle are two main themes of our paper. These
yield a rich set of testable implications which have not been examined in the extant literature.
First, the declining net benefit of corporate takeover defenses over a firm’s life cycle implies that
there is a distinct life cycle effect for the value impact of takeover defenses. Young firms have a
positive relationship between firm value and takeover defenses but the sign of this relationship
reverses and there is a negative relationship between firm value and takeover defenses for mature
firms. Second, a firm’s needs for making a commitment to honor stakeholders’ claims has a first
order effect on the valuation effect of takeover defenses at the earlier stage of its life and
managerial moral hazard problem has the first order effect on the valuation effect of takeover
defenses when the firm matures. Third, takeover defenses which have a commitment nature will
especially enhance the life cycle effect of takeover defenses regarding firm value creation.
Takeover defenses used as an irreversible commitment device may lead to ex ante value creation
but ex post value destruction and thus cause a value reversal (Stout (2002)). We devise a
takeover defense index which we call c-index (commitment index) which is defined as the
number of takeover defenses whose sole purpose is to make it hard to remove takeover defenses.
This c-index includes supermajority requirement to amend the charter and supermajority to
5
amend the firm bylaws.5 Some features of c-index are noteworthy. First, the results reported in
the Appendix table 6 show that c-index is significantly negatively related to the likelihood of
removal of other takeover defenses, which validate our use of c-index to measure the value of
commitment in takeover defenses. Second, our use of c-index to measure the value of
commitment in takeover defenses is justified by another observation. Takeover defenses which
are counted in c-index are adopted at the IPO stage and remain stagnant over a firm’s life, which
is consistent with the nature of commitment. On the other hand, other takeover defenses are more
flexibly revised at firm discretion over a firm’s life. Thus, our c-index is a useful measure of the
degree of stagnation of takeover defenses over time.
Our life cycle of takeover defenses hypothesis explains why previous studies which utilize a
cross-sectional approach find inconsistent results concerning the valuation effects of takeover
defenses. These studies simply examine the valuation effects of takeover defenses at different
stage of a firm’s life. Furthermore, the life cycle of takeover defenses explains the puzzle of why
many firms adopt takeover defenses even when such an adoption appears to be value-destroying.
When takeover defenses are adopted at the earlier stage of a firm’s life, the benefit of takeover
defenses exceeds the cost of takeover defenses. Later, the valuation effects of takeover defenses
are reversed as a firm matures but takeover defenses are difficult to remove since takeover
defenses are stagnant over time.
Our empirical findings are consistent with our life cycle view of takeover defenses
hypothesis. We find that there are two categories of takeover defenses. There are takeover
defenses which are stable over time and takeover defenses which are flexibly adopted and
removed depending on firm discretion of takeover defense adoption over a firm’s life cycle. The
first one corresponds to the bonding hypothesis of takeover defenses suggested by Johnson,
Karpoff and Yi (2015) which emphasizes the role of takeover defenses as a commitment device
at the earlier stage of a firm’s life. The latter corresponds to the classical agency view of takeover
defenses which argues that managers try to entrench themselves by adopting takeover defenses.
Thus, we find that certain takeover defenses are adopted at different stages in a firm’s life for
different purposes. Previous studies which focus on changes in aggregate score takeover defense
index (Gompers, Ishii, and Metrick (2003)) cannot tease out the importance of individual
5
In addition, we examine the interactions between the c-index and other takeover defenses in our Appendix.
6
takeover defenses. Furthermore, we find that this distinction between two categories of takeover
defenses plays a significant role in the life cycle effect of takeover defenses and value reversal.
We introduce the commitment index (c-index) which includes firm characteristics making it hard
to revise firm takeover defenses. Our empirical analyses confirms that it is the c-index that
causes a reversal in the sign of the relationship between firm value and takeover defenses. This is
consistent with Stout (2002) who argues that takeover defenses used as a commitment to honor
stakeholder claims will create ex ante efficiency and ex post inefficiency. Also, consistent with
the bonding hypothesis of takeover defenses (Johnson, Kaporff and Yi (2015), we find that
takeover defenses not in the c-index (non-c-index) do not show value reversal in the relationship
between firm value and takeover defenses but only decreases firm value.
In addition, we find evidence of a life cycle effect of takeover defenses on firm value. In the
year of an IPO and in the year after the IPO, we find that the industry adjusted Tobin’s Q of
firms which have above the median e-index is greater than the industry adjusted Tobin’s Q of
firms which have below the median e-index. However, starting 3-4 after the IPO event, we find
that firms with more than the sample median number of takeover defenses have a significantly
lower industry adjusted Tobin’s Q compared with the firms with fewer than the sample median
number of takeover defenses. This indicates that there is a reversal in the valuation effects of
takeover defenses over a firm’s life cycle. We find a similar pattern of reversal of the valuation
effects of takeover defenses when we use classified board and c-index instead of e-index.
Consistent with the aforementioned univariate results, in multivariate analysis using the exact
control variables utilized by Gompers, Ishii and Metrick (2003), we find that the interaction
between e-index and firms being more than four years from their IPO is significantly negatively
related to industry adjusted Tobin’s Q at the 1% level. This result still holds when we utilize
staggered board or c-index as the measure for takeover defenses.
We also find evidence that takeover defenses which are utilized as a commitment device and
are hard to revise are the driving force which leads to the value reversal we document. We find
that the valuation effect of c-index goes from value-increasing to value-decreasing over a firm’s
life cycle. On the contrary, other takeover defenses which can be easily removed by the board
without shareholder approval does not show a strong value reversal. We find that the valuation
effect of e-index goes from value-increasing to value-decreasing over a firm’s life cycle even
7
when we utilize firm age and sales growth rate of the firm or the firm’s industry as alternative
proxies of IPO firm maturity.
Importantly, we find that a firm’s needs to promote stakeholder relationships using takeover
defenses as a commitment device make takeover defenses value-increasing but these takeover
defenses become value-destroying since takeover defenses are insensitive to changes in the
economic environment which affect the valuation effect of takeover defenses. We use two
variables to capture a firm’s needs to promote stakeholder relationships using takeover defenses
as a commitment device. One is an indicator variable for having a large customer and the other is
and indicator for having strategic alliances. The fact that a firm relies on a large customer
indicates that the firm fails to diversify its stakeholder-base possibly due to undeveloped
distribution channels and a small number of large customers will hold a significant bargaining
power over the firm (Johnson, Kang, Masulis and Yi, 2015). Furthermore, previous studies
suggest that sustaining strategic alliances require underlying relationship-specific investments by
strategic alliance partners. Thus, these two variables can effectively capture a firm’s needs to
promote stakeholder relationships using takeover defenses as a commitment device. We find that
when the firm is older than the median firm the interaction between e-index, and the large
customer indicator variable is significantly negatively related to industry-adjusted Tobin’s Q. We
find similar results when we use strategic alliance measure capture a firm’s needs to promote
stakeholder relationships. As the firm matures, the importance of the firm’s relationships with
outside parties such as customers and alliance partners declines, making the commitment value
of takeover defenses lower as time passes.
In addition, we examine the life cycle of insider shareholdings and find that the costs of
takeover defenses increase over time. Specifically, we find that as the managers reduce their
shareholdings in the firm, the value increasing benefits of takeover defenses declines. This result
is important since it shows that as the insider ownership becomes more diffuse, the alignment of
shareholder – manager interests begins to diverge and takeover defenses become more
problematic for the firm. Finally, we examine the empire-building of the managers and find that
firms with more takeover defenses are more likely to conduct diversifying acquisitions when the
firm has more takeover defenses. This result implies that there are real costs to adopting takeover
8
defenses at the IPO stage, but these costs are only felt by the firm when the benefits of takeover
defenses decline as the firm ages.
Finally, we control for the endogeneity of takeover defense adoption using law firm
indicators and an indicator for a firm being located in the state of California. The life cycle effect
of takeover defenses still remain even after we control for the endogeneity associated with the
adoption of takeover defenses.
Our paper makes contributions to the extant literature in several important ways. First, our
paper is the first study to examine a life cycle pattern in valuation effects of takeover defenses in
a panel data setting. For the first time in the literature, we report a rich set of stylized facts and
empirical test results concerning the evolution of takeover defenses over a firm’s life cycle and
corresponding changes in the valuation effect of takeover defenses over a firm’s life cycle. We
first find that there are two categories of takeover defenses which show different dynamic
patterns in the evolutions over a firm’s life cycle. We also find that this difference is very
important in explaining the valuation effect of takeover defenses. Our finding of value reversal
of takeover defenses over a firm’s life cycle is new to the literature as well. We document the
near invariability of takeover defenses over time and provide evidence that this invariability is
the key determining the negative relationship between takeover defenses and firm value as
documented in the previous studies including Gompers, Ishii and Metrick (2003) and Masulis,
Wang, and Xie (2007). Second, utilizing our panel-data based life cycle approach, we reconcile
seemingly inconsistent results documented in the previous literature and explain why takeover
defenses are a prevalent corporate phenomenon even when they are value destroying for mature
firms. Third, we introduce a c-index which is designed to capture the value of commitment in
takeover defenses. Compared with g-index or e-index which are aggregate scores based on the
number of a firm’s takeover defenses, our c-index can reflect the fact that individual takeover
defenses are adopted at different stage of a firm’s life for different purposes. Using our c-index
and panel data of firm value and takeover defenses, we provide evidence supporting the bonding
hypothesis which suggests that takeover defenses are used as a commitment device to honor
stakeholder claims (Johnson, Karpoff and Yi (2015)).
9
Our paper is organized as follows. In Section 2 we discuss our data selection and sample. In
Section 3 we present summary statistics and empirical results. Section 4 reports various
robustness check results. Section 5 concludes the paper.
2. Data and Sample
Our sample of firms starts with the universe of IPOs going public between 1997 and 2011. We begin
our sample in 1997 to ensure that the firms have annual reports, proxies, and prospectus filings available
in the SEC’s EDGAR database. We eliminate all REITS, ADRs, funds, firms without CRSP and
COMPUSTAT coverage, firms incorporated outside the US, and firms with a dual share class structure.
In addition, we merge in data from Jay Ritter’s web site on firm founding dates.6 This yields a sample of
2,285 IPO firms in our sample with sufficient data concerning stock prices in CRSP, accounting data in
COMPUSTAT, and takeover defense data in the SEC filings of the firm.
For each firm in our sample, we then create a panel dataset, examining how long the firm remains in
COMPUSTAT. For each firm-year observation, we hand collect the holdings of insiders disclosed in the
SEC filings of the firm (predominantly the proxy statements). In addition, we use the Thompson Reuters
13f filing database to collect the shareholdings of institutional shareholders in the firm after its IPO.
We then use the SEC’s EDGAR database to collect the takeover defenses of the firm in the IPO year
and each subsequent year that the firm remains in the COMPUSTAT database. We collect the takeover
defenses determined by Bebchuk, Cohen, and Ferrell (2009) to be particular important to shareholders
and firm valuation. Specifically, we collect information about board classification, poison pills,
supermajority requirements to change firm bylaws, supermajority requirements to change the firm charter,
golden parachutes, and supermajority requirements to approve mergers. We then follow Bebchuk, Cohen,
and Ferrell (2009) in coding each of these takeover defenses using an indicator variable and creating an eindex (entrenchment index) as well as examining each takeover defense individually.
To collect the takeover defenses, we begin in the IPO year by examining the IPO firm prospectus
with its attached bylaws and charter. This allows us to collect the full suite of takeover defenses adopted
at the IPO stage, similar to Johnson, Karpoff, and Yi (2015).7 We then proceed through the annual proxy
statements, annual reports, and (where necessary) press releases of the firm for all the years the firm
remains in COMPUSTAT to track all changes to the takeover defenses adopted at the IPO stage. For
further details of the takeover defense collection, please see Data Appendix I. This process allows us to
6
We thank Jay Ritter for generously providing this data at https://site.warrington.ufl.edu/ritter/ipo-data/.
Our sample is almost twice the size of Johnson, Karpoff, and Yi (2015) (2,285 versus 1,219), including many more
firms after their sample ends.
7
10
create a panel of 15,740 firm-year observations where we have full COMPUSTAT data as well as the
firm takeover defenses.
2. Data and Sample
Our sample of firms starts with the universe of IPOs going public between 1997 and 2011. We begin
our sample in 1997 to ensure that the firms have annual reports, proxies, and prospectus filings available
in the SEC’s EDGAR database. We eliminate all REITS, ADRs, funds, firms without CRSP and
COMPUSTAT coverage, firms incorporated outside the US, and firms with a dual share class structure.
In addition, we merge in data from Jay Ritter’s web site on firm founding dates.8 This yields a sample of
2,285 IPO firms in our sample with sufficient data concerning stock prices in CRSP, accounting data in
COMPUSTAT, and takeover defense data in the SEC filings of the firm.
For each firm in our sample, we then create a panel dataset, examining how long the firm remains in
COMPUSTAT. For each firm-year observation, we hand collect the holdings of insiders disclosed in the
SEC filings of the firm (predominantly the proxy statements). In addition, we use the Thompson Reuters
13f filing database to collect the shareholdings of institutional shareholders in the firm after its IPO.
We then use the SEC’s EDGAR database to collect the takeover defenses of the firm in the IPO year
and each subsequent year that the firm remains in the COMPUSTAT database. We collect the takeover
defenses determined by Bebchuk, Cohen, and Ferrell (2009) to be particular important to shareholders
and firm valuation. Specifically, we collect information about board classification, poison pills,
supermajority requirements to change firm bylaws, supermajority requirements to change the firm charter,
golden parachutes, and supermajority requirements to approve mergers. We then follow Bebchuk, Cohen,
and Ferrell (2009) in coding each of these takeover defenses using an indicator variable and creating an eindex (entrenchment index) as well as examining each takeover defense individually.
To collect the takeover defenses, we begin in the IPO year by examining the IPO firm prospectus
with its attached bylaws and charter. This allows us to collect the full suite of takeover defenses adopted
at the IPO stage, similar to Johnson, Karpoff, and Yi (2015).9 We then proceed through the annual proxy
statements, annual reports, and (where necessary) press releases of the firm for all the years the firm
remains in COMPUSTAT to track all changes to the takeover defenses adopted at the IPO stage. For
further details of the takeover defense collection, please see Data Appendix I. This process allows us to
8
We thank Jay Ritter for generously providing this data at https://site.warrington.ufl.edu/ritter/ipo-data/.
Our sample is almost twice the size of Johnson, Karpoff, and Yi (2015) (2,285 versus 1,219), including many more
firms after their sample ends.
9
11
create a panel of 15,740 firm-year observations where we have full COMPUSTAT data as well as the
firm takeover defenses.
3.
Summary Statistics
Our sample firms begin in 1997 to allow us to collect data from the SEC’s EDGAR web site which
reliably provides firm data for small firms starting in this year. Examining the distribution of firms by
IPO year, we report in Table 1 which our sample includes firms both before and after the Internet boom.
This allows us to ensure that our results are not driven by the Internet boom period. As can be seen from
our sample, there were many more IPO firms per year prior to 2001 compared to after 2001, consistent
with the findings of Gao, Ritter, and Zhu (2013). We find that the year with the greatest number of IPOs
is 1999 with 410 and the year with the fewest is 2008 with 16.
Since we are most interested in the way a firm changes its performance, valuation, accounting
information, and takeover defenses over time, in Table 2, we examine the IPO firm characteristics at
several points in time. We find that at the IPO stage, the firm total assets are $867 million with a post IPO
market capitalization of $825 million. Moving across columns to year 5, 10, and 14, we find that the
average firm size measured both in terms of assets and market capitalization increases substantially. For
instance, the average assets of a firm at year 14 is $6.6 billion and the average market capitalization is
$2.9 billion. The average firm age in our sample is 15 years at the IPO and (not surprisingly) 30 years 14
years later, suggesting a slight survivorship bias for older firms. There is no firm in the S&P 500 at the
IPO stage, but by year 14, 7% of our sample is in the S&P 500. At IPO, 79% of IPO firms are
incorporated in Delaware but this percentage declines to 71% at year 14. Since firms are unlikely to
reincorporate outside of Delaware, this result is most likely the result of Delaware firms
disproportionately leaving our sample through either acquisition or delisting. Likewise, we find that
insider holdings at the IPO stage hold 42% of the shares of the firm, declining to 14% by year 14. Finally,
we find at the IPO stage, 8% of the shares are held by large institutional shareholders, increasing to 54%
of shares at year 14.
It is important to note, however, that this preliminary examination over time may result in what
appear to be changes in firm characteristics caused not by individual firm changes, but rather by
survivorship bias. We will address this issue more formally in the Appendix. It is important to note that
where the initial sample at the IPO stage includes 2,285 firms, the sample only has 235 IPO firms that
have observations 14 years later. This is caused both by firms leaving the sample, and by the end of our
sample period in 2014.
12
3.1
Changes in Corporate Takeover Defenses over a Firm’s Life
In Table 3.A we examine the prominence of takeover defenses over time for the IPO firm. We find,
for instance, at the IPO stage, 65% of IPO firms adopt a classified board, consistent with Johnson,
Karpoff, and Yi (2015). However, consistent with the general trend of firms removing takeover defenses
(Cohen and Wang (2013)), we find that by 15 years after the IPO, only 47% of firms still have a classified
board. The fact that IPO firms reduce their use of classified boards after their initial public offering at
lower rates than large firms is not surprising since the prior literature has found that small firms are more
likely to retain a classified board than large firms.
We find that only 5% of IPOs have a poison pill at the time of their IPO. But IPO firms adopt
substantial numbers of poison pills in the years after the IPO. For instance, by year 5, 21% of firms in our
sample have a poison pill. The percent of firms with a poison pill peaks at 28% in year 10 and then
declines to only 17% by year 15. A substantial portion of the removal of poison pills are automatically
triggered by the passage of time. Nearly all the firms in our sample pass a poison pill with a ten year
sunset clause.
Supermajority requirements to amend the bylaws and charter of the firm are relatively stable from
the IPO stage to later stages. For instance, at the IPO stage, only 34% of firms require a supermajority to
amend the firm bylaws and 30% of firms require a supermajority to amend the firm charter. By year 15,
these percentages are nearly the same at 32% and 31%, respectively. In contrast, there appear to be larger
changes in the supermajority requirement to approve mergers. This figure starts in the IPO year at 41%
and by year 15 has increased to 50%. Finally, we find that at the IPO stage, 64% of firms have a golden
parachute for their CEO. This figure increases sharply to 97% at year 15. This particular increase,
however, seems to be driven by the very high levels of golden parachutes being adopted in the last 5
years.10 Our results suggest that there are two categories of takeover defenses: those adopted chiefly at the
IPO and those which are flexibly changed by the board over time. The former corresponds to commitment
benefit of takeover defenses suggested by Johnson, Karpoff and Yi (2015) and the latter corresponds to
manager-entrenchment effect suggested by the agency view (Easterbrook and Fischel, 2001).
We follow Bebchuk, Cohen, and Ferrell (2009) by creating an entrenchment index or e-index in each
year after the IPO. This index is created by coding all six of the previously described takeover defenses as
10
As can be seen by the data in Table 3, there is a substantial change in the percent of firms with golden parachutes
starting in the 9th year after the IPOs. This substantial increase corresponds to observations after 2007 when IPO
firms tend to have golden parachutes for the CEOs and other officers at rates above 90%. To ensure that this is not a
coding problem or other difficulty in our data, we randomly select 101 IPO firms after 2007 that are coded as having
a golden parachute and examine their annual proxy statements to determine if they do in fact have golden
parachutes. We find only one case where the firm was disclosed as having a golden parachute, but in fact did not.
This error rate (1/101 = 1%) is well below the expected rate in the sample and implies that our data in fact accurately
represent the trend in golden parachute use.
13
indicator variables and adding these six indicators to create the e-index. The e-index is based on six
takeover defenses which are consistently voted against by institutional investors, implying that these six
takeover defenses disproportionately destroy firm value. We find in the IPO year that the average firm’s
e-index begins at 2.40 at the IPO, increases to 3.05 in year 10, and finally declines to 2.73 in year 15.
Much of the increase in e-index during the first years of an IPO firm is caused by poison pill adoption and
golden parachute adoption. In contrast, the subsequent decline is largely caused by board declassification
and automatic poison pill expirations. Thus, it appears that individual takeover defenses are used for
different purposes at different points of time in the IPO firm’s lifecycle. An aggregate takeover defense
index based on the number of takeover defenses adopted by a firm cannot capture this effect. This
observation motivates us to devise takeover defense indices based on qualitative differences of individual
takeover defenses.
Since our intent is to study the lifecycle of firms, we wish to avoid as far as possible the impact of
macro events and changes in industry practice in the examination of our data. For instance, it is well
known that nearly half of all S&P 500 firms who had classified boards in 2000 no longer have them
(Cohen and Wang (2013)). Likewise, the trends in golden parachutes or poison pills could also be driven
by the fact that most of our observations above year +10 occur after 2000, a time when there was a
definite push to eliminate certain takeover defenses.
To help to mitigate some of these concerns, we split our sample in half. We choose to split the
sample at March 2000 as this is the end of the Internet boom and approximately splits our sample into two
equal halves. In Table 3.B we examine the Internet Boom IPOs, finding similar trends in classified board,
poison pill, and golden parachute. Likewise, there is a definite trend in the increase in supermajority
requirements to approve mergers. In addition, we find a low e-index at the beginning of the firm’s
lifecycle (2.05) which peaks at year 10 (3.01) and then declines thereafter (2.77). In Table 3.C we find
that the overall trends for post-Internet Boom IPOs are similar with only one exception. Where the earlier
classified board trend is downward, the later IPOs tend to have a slight increase in the presence of
classified boards. Otherwise, the overall data shows similar trends to earlier IPO firms. In untabulated
tests, we repeat this tabulation of IPO cohorts by year, once again finding similar results. This implies that
our results are mostly lifecycle driven as opposed to driven by a macro-change in the acceptance of
certain takeover defenses.
One difficulty in interpreting the results in Table 3 is in determining if the change in sample
characteristics is driven by firms adopting and removing certain takeover defenses or the disappearance of
certain firms from our sample who have or do not have certain takeover defenses. For instance, if all IPO
14
firms without a poison pill are acquired within a certain number of years, the remaining firms will all
have poison pills, leading the data to show an increase in poison pills over time.
To assist in interpreting the trends in Table 3, we enumerate the individual firm adoptions (Table
4.A) and removals (Table 4.B) of takeover defenses. In Table 4.A we see that from year 0 to 1, no IPO
firm adopted a classified board. However, in the following year, 4 firms adopted a classified board. In the
subsequent four years, we find that one firm in our sample adopted a classified board each year. These
eight observations account for all board classifications within our sample. We find that a substantial
number of firms are adopting poison pills for the first ten years after an IPO. However, the rate of poison
pill adoption seems to level off thereafter. The table also shows few adoptions of supermajority
requirements to amend the bylaws or charter of the firm. But we find substantial numbers of firms
incorporating supermajority requirements to approve mergers in the first ten years of the IPO. Likewise,
many firms are adopting poison pills in the early life of the firm.
In Table 4.B we examine the removal of takeover defenses. Where there are very few adoptions of
classified boards in our sample, we find many more firms removing classified boards. Strikingly, there are
fewer poison pill removals (75) compared to classified board removals (107). The removal of poison pills
tend to be clustered around 10-12 years after the IPO, consistent with the majority of these poison pills
being eliminated through the sunset clause, rather than by a direct vote of the directors.11 There are more
firms removing than adopting supermajority requirements to amend bylaws (34) and charters (27)
although they do not seem to be clustered at any particular point in time. The supermajority requirements
to amend the bylaws and the charter are by far the most stable takeover defenses, with very few changes
in these takeover defenses. Supermajority requirements to approve mergers may be clustered early in the
firm lifecycle with 12 adoptions in year 1. Finally, golden parachutes have many adoptions (380), but
only a few total removals (19) with most occurring within a few years after the IPO.12
3.2
Changes in Firm valuation over a Firm’s Life Cycle
11
More than 95% of the poison pills are allowed to expire through the sunset clause. A few firms reduce the life of
the original ten year subset clause to (for instance) three years and then allow the poison pill to expire. For instance,
the board of Biomarin Pharmaceutical authorized a stockholder rights plan in September 2002 and “accelerated the
final expiration date of the Company’s preferred share purchase rights under the Rights Agreement.” After
accelerating the expiration of these rights, the board eventually allows the rights to expire. See BioMarin’s 2013
Proxy Statement for further details.
http://www.sec.gov/Archives/edgar/data/1048477/000119312513077031/d455570d10k.htm
12
These results do not examine how some firm characteristics included in the e-index as “takeover defenses” affect
the likelihood of changing other takeover defenses adopted by a firm. For instance, when a firm adopts a
supermajority requirement for charter and bylaws amendments, other takeover defenses adopted by the firm are less
likely to change over a firm’s life cycle. See the Appendix table 6.
15
To better understand the relationship between firm valuation and firm lifecycle, we must first
examine the way an IPO firm valuation changes over time. In Table 5 we report mean and median
Tobin’s Q for the first 15 years of our sample of firms. Note that all firm values for Tobin’s Q are
winsorized at the 99th percentile.13 We find that at the IPO stage, the mean (median) Tobin’s Q is 4.17
(2.48). This figure is quite high and suggests that IPO firms are considered likely to have quite high
growth rates. Subsequent to the IPO, we find that Tobin’s Q declines substantially even by year 1, with a
mean (median) value of 2.50 (1.68). While this may still seem high relative to expectations, industry
adjusting the Tobin’s Q suggests that this is not the case.
In the final two columns of Table 5, we report the industry median adjusted Tobin’s Q. We follow
Gompers, Ishii, and Metrick (2003) by taking each firm Tobin’s Q and subtracting the industry median
Tobin’s Q. We then report the mean and median of this value in the table. Our results are similar to the
unadjusted Tobin’s Q values. We find a substantially higher value at the IPO stage compared to all
subsequent years after the IPO. This implies that the markets expect a substantially higher growth rate at
the IPO stage relative to in subsequent years (or that expected stock returns are low or that market
valuation is irrationally high just after the IPO). It does appear, however, that this initially higher Tobin’s
Q adjusts by the year after the IPO to be quite similar to the rest of the industry for each firm, as seen by
the near zero industry adjusted Tobin’s Q medians in the last column of Table 5.
3.3
The Relationship between Valuation and e-index over a Firm’s Life Cycle: Value Reversal
3.3a Univariate results
In Tables 6 we report the relationship between industry adjusted Tobin’s Q and e-index over the
lifecycle of the firm. Where some of the prior literature (Johnson, Karpoff, and Yi (2015)) shows a
positive relationship between firm valuation and takeover defenses for IPO firms, other literature
(Gompers, Ishii, and Metrick (2003) and Bebchuk, Cohen, and Ferrell (2009)) shows a negative
relationship for mature firms. To reconcile these inconsistencies in the literature, we examine the Tobin’s
Q – e-index relationship in all years from the IPO until firm maturity.
We find in Table 6.A, for instance, that the industry adjusted Tobin’s Q for firms with above the
median e-index (2.00) is 2.51 where the Tobin’s Q for firms with below the medina e-index is 2.09. This
difference between Tobin’s Q for firms above versus below the median e-index is statistically significant
at the 95% level with a t-stat of 1.97. Likewise, in the year after the IPO, we find that the Tobin’s Q for
firms with above the median e-index value is 0.92 compared to firms with below the median e-index
13
Our results are not appreciably changed by winsorizing at the 95 th percentile or by not winsorizing at all.
However, unwinsorized values lead to anomalous reported means due to the influence of significant outliers. We
choose to winsorize at the 99th percentile to be consistent with Gompers, Ishii, and Metrick (2003).
16
value at 0.59. Once again, this difference is statistically significant at the 1% level with a t-stat of 3.18.
Two years after the IPO we find that the IPO firms with above the median number of takeover defenses
have a higher Tobin’s Q, but the difference is no longer significant. Starting in years 3-4, we find that the
firms with more than the median number of takeover defenses no longer have a higher valuation, but in
fact have a lower valuation, albeit insignificantly lower. For years 5-6, 7-9, and ≥10 we find a
significantly lower firm valuation for firms with a higher level of takeover defenses as measured by eindex.
To ensure that our results are not driven by the construction of the e-index measure, we repeat our
major results using only the classified board measure. This measure shows remarkably similar results
with a positive and significant relationship between Tobin’s Q and classified board in the early years of
the firm and a negative and significant relationship in the later years of the firm. As with the e-index
measure, the relationship between Tobin’s Q and classified board turn negative and significant starting
five years after the IPO.
Finally, we construct two new measures of takeover defenses which we call the commitment index
or c-index and the voting index or v-index. The c-index is important for firms wishing to commit to
strong relationships with customers and alliance partners since it only includes bylaw and charter
characteristics that make “overall” takeover defenses adopted by a firm stable. The c-index is composed
of the supermajority requirements to amend the charter, or firm bylaws since these two characteristics
make firm takeover defenses substantially more stable. The voting index measure is based on the takeover
defenses that can only be changed by the firm with the vote of the firm’s shareholders. This index,
composed of classified board, and supermajority requirements to change the firm bylaws, change the firm
charter, or approve a merger, is important because the board cannot make changes to this particular index
without the shareholders voting to approve the change. V-index may measure the degree of difficulty in
revising takeover defenses. Alternatively, v-index may measure how closely adoption of takeover
defenses are attached to share value maximization since the adoption of takeover defenses included in vindex require a shareholder vote.
As we discussed before, an aggregate takeover defense index which is based on the number of
takeover defenses cannot capture the fact that various takeover defenses are adopted or removed at
particular times in a firm’s life cycle and serve different purposes. Furthermore, traditional takeover
defense measures will suffer from statistical problems if the addition or removal of some takeover
defenses are systematically related to other important firm characteristics. Our c-index is designed to
capture the value of commitment in takeover defenses where our v-index is meant to capture the difficulty
with which managers may change firm takeover defenses.
17
As with the classified board measure, we find that the firms in their early years after the IPO have a
positive and significant relationship with c-index but starting in year 5, have a negative and significant
relationship between firm value and c-index. Table 6.D reports these results. The table consistently shows
that there is a positive relationship between takeover defenses and firm value in the early years of a public
firm. We then repeat our analyses utilizing the takeover defenses not in the c-index and call this the Nonc-index. These takeover defenses include classified board, golden parachute, poison pill, and
supermajority to approve mergers. Interestingly, when we examine the relationship between industry
adjusted Tobin’s Q and non-c-index over the lifecycle of the firm, we cannot find a strong pattern of
value reversal in the sign of the relationship over a firm’s life cycle. We do find that firms with a low
level of Non-c-index at the IPO stage have a lower firm valuation, but not significantly so. It is not until
years 5-6 that the valuation of the low Non-c-index firms have a significantly higher value relative to high
Non-c-index firms. These results are reported in Table 6.D. The aforementioned results indicate that
reversal in the sign of the relationship over a firm’s life cycle is closely related to the bonding role of
takeover defenses which make a commitment at the early stage of a firm’s life and the takeover defenses
which are flexibly revised over a firm’s life cycle should not contribute to the occurrence of value
reversal.
In examining the v-index we find results qualitatively similar to the c-index and the e-index. In year
0 and year 1 the firm has a significantly higher valuation if the firm has above the median number of
takeover defenses. But as the firm matures, the sign of the relationship changes. Firms with below the
median number of v-index takeover defenses have a significantly higher valuation starting in years 5-6.
These results are reported in Table 6.E. We then repeat our analyses using the Non-v-index composed of
takeover defenses which can be adopted or removed by the board without shareholder approval (golden
parachute and poison pill). Our results show in Table 6.F that there is almost no reliable relationship
between the Non-v-index and firm valuation except more than 9 years after the IPO. In this case, firms
have a negative and significant relationship between having more takeover defenses and firm value. The
aforementioned finding is consistent with our previous finding that c-index is related to the occurrence of
value reversal and non-c index is not. Also, this finding implies that it is not managerial discretion that
causes value reversal.
Overall, our results very powerfully reconcile the inconsistencies in the prior literature.
3.3b Multivariate results
To ensure that our results are not driven by other variables known to be related to Tobin’s Q, we
conduct a series of tests using a multivariate regression setting. Specifically, we conduct our analyses
18
using the exact control variables utilized by Gompers, Ishii, and Metrick (2003) plus we include two other
important controls.14 The Gompers, Ishii, and Metrick (2003) variables we include as controls are an
indicator variable taking a value of one if the firm is incorporated in Delaware, the firm age in years, log
(firm assets), and an indicator taking a value of one if the firm is in the S&P 500. In addition, we include
the percent of shares held by directors as reported in the firm’s SEC filings and the total percent of shares
held by institutional shareholders as reported in the firm’s 13f filings. In addition, since we cannot obtain
the insider holdings in all years, we replace missing values with a zero and add an indicator variable
taking a value of one if the insider shareholdings are not available.15 We feel that these two controls are
also necessary since the insider shareholdings and institutional shareholdings could have a significant
impact on firm value.
In Table 7.A we report our multivariate regression results using industry adjusted Tobin’s Q as the
dependent variable. These results are similar to Gompers, Ishii, and Metrick (2003) in that they pool each
firm-year observation into one large regression. We find in model 1 that there is a negative but
insignificant relationship between firm value and e-index. In model 2 we include an indicator variable for
observations where it has been more than four years since the firm’s IPO. This variable is significantly
negative, consistent with the results in Table 5 showing that Tobin’s Q tends to decline over time. Finally,
in model 3, we examine the interaction between e-index and firms being more than four years from their
IPO and find that this interaction is statistically significant and negative at the 1% level. In addition, once
we control for firm age and this interaction, we find that the relationship between firm value and e-index
becomes positive and significant. This suggest that the prior literature suggesting an unequivocal negative
relationship between firm value and takeover defenses only tells part of the story.
In Table 7.B we repeat our analyses using sub-samples of the IPO firms based on how far the firms
are from their IPO. Our firm regression in model 1 uses only IPO firms at the state of their IPO. This
particular regression is most similar to Johnson, Karpoff, and Yi (2015) which only considers IPO
valuation at the IPO stage. But as we move to model 2-7, we go cross-sectionally from the firms having
most recently gone through their IPO (model 1) to IPO firms who are one year old (model 2) and so forth.
This allows us to simply examine when and if there is a change in the Tobin’s Q – takeover defense
relationship as the firm matures.
We find in model 1 that there is a positive and statistically significant relationship between Q and eindex. The coefficient of 0.162 implies that at the mean e-index value of 2.40, adding one standard
14
In addition, when we repeat our analyses using only the control variables from Gompers, Ishii, and Metrick (2003)
without controlling for institutional ownership or insider share holdings, we find almost identical results.
15
Our results are nearly identical if we eliminate the N=974 observations with missing insider holdings. This
accounts for 6.19% of our firm-year observations.
19
deviation in takeover defenses (1.60) results in an increase in industry adjusted Tobin’s Q of 0.25 or
10.4% of firm value. This result is statistically significant at the 1% level. In model 2, we find that there is
also a positive and statistically significant relationship between Tobin’s Q and the level of takeover
defenses adopted by the firm. Thus, it is clear that the IPO firm takeover defenses enhance firm value not
just at the IPO stage, but also one year later. In model 3 we find that the relationship between Tobin’s Q
and the level of takeover defenses is also positive, but is no longer statistically significant. Likewise, in
year 3-4 in model 4, we find a positive but statistically insignificant relationship between firm value and
takeover defenses.
When we move on through time to years 5-6 after the IPO in model 5, we find a negative and
statistically significant coefficient of -0.071, significant at the 5% level. This result implies that starting
five years after the IPO of a firm, the takeover defenses will cause a decrease in the value of the firm.
Likewise, as we move to years 7-9 (model 6) and year ≥10 (model 7) we find a more negative and more
statistically significant relationship between the IPO firm value and e-index. This result is consistent with
the negative and robust relationship found by Bebchuk, Cohen, and Ferrell (2009) between firm value and
e-index. The sample of firms examined in model 7, IPO firms at least ten years old is the closest to the
Bebchuk, Cohen, and Ferrell (2009) sample.
In addition, we plot the year-by-year coefficients for the first 15 years the IPO firm is public in
Figure 1. This figure also plots the standard errors of the regression. It is important to note that as the
number of observations decreases, particularly after 10 years from the IPO, the standard error increases
substantially. The figure clearly shows that for the first two years, the relationship between takeover
defenses and firm value is positive and significant. Although the relationship is not statistically
significant, the point estimate of the valuation is positive through year three after the IPO. At year 4, the
relationship between takeover defenses and firm value becomes negative. However, not until year 6 is the
relationship negative and statistically significant. Subsequently, the relationship remains negative and
statistically significant for all future years through year 15.
In Table 7.C we repeat our analyses using classified board as the only takeover defense considered.
Most importantly, we find our results are quite similar to the results already discussed in Table 7.A. In
particular, in model 1 we find that a classified board is negatively (albeit insignificantly) related to firm
value. In model 2 we find that as the firm ages, the value of the firm declines as seen by the negative and
significant coefficient on the indicator variable for firms more than four years after their IPO. Finally in
model 3 we find that the interaction term between the classified board indicator and the age > 4 years
variable is negative and highly statistically significant at the 1% level. Although the level of statistical
20
significance is not identical to Table 7.A using the e-index measure, the results are strikingly similar in
Table 7.C.
We then repeat our year-by-year regressions in Table 7.D using classified board as the independent
variable of interest. As in Table 7.B using e-index, we find that in the early years of the firm (year=0) the
presence of a classified board is positively and significantly related to firm value. As time progresses
from year=1 (model 2) to years 5-6 (models 3-5) the coefficients go from being positive to negative and
eventually, significant. Moving past years 5-6, we find that the coefficient on a classified board is reliably
statistically significant for years 7-9 (model 6) and years ≥10 (model 7). This result is analogous to the
result tabulated for e-index and shows a strong decline in the valuation impact of classified boards as the
firm ages.
One important point of our research is that the commitment value of takeover defenses are strongest
if they are difficult to add or remove. We therefore examine the commitment index (or c-index). In Table
7.E and 7.F we find strikingly similar results to those tabulated for e-index and classified boards. As the
firm ages, the presence of the c-index goes from being value-increasing to being value-decreasing. In
particular, in Table 7.F we see that the takeover defenses in the c-index measure are associated with
higher firm values for years 0, 1, 2, and 3-4 (models 1-4), albeit insignificantly higher in the last two
models. However, starting in years 5-6 (model 5) and thereafter, the presence of takeover defenses is
generally associated with declining firm values. Note, however, that the negative coefficients are not
statistically significant for the c-index in any of the models.
To further confirm our previous results which suggest that value reversal is caused by those takeover
defenses which have a commitment nature, we then examine the takeover defenses which can be adopted
by the board without shareholder approval, the so-called Non-c-index which is composed of golden
parachutes and poison pills. These takeover defenses can be adopted or removed at the discretion of the
board and so they are not considered permanent fixtures in the governance of the firm. We find in Tables
7.G and 7.H that the firm value does not appear to be enhanced by these two particular takeover defenses.
However, we also find evidence that these particular takeover defenses may serve to reduce firm value as
the firm ages. For instance, in Table 7.G model 3 we find an insignificant coefficient on the interaction
term between the Non-c-index measure and the indicator for firms more than four years after their IPO. In
Table 7.H we do find some evidence that after years 7-9 and beyond (models 6 and 7) poison pills and
golden parachutes are associated with lower firm value. Once again, the expected takeover defense
lifecycle effect with value enhancement early in the life of the firm and value destruction later in the life
of the firm is present.
21
We then examine the v-index of takeover defenses which can be changed by a shareholder vote in
Table 7.I and 7.J. Once again, we find strong support for the v-index takeover defenses being associated
with value enhancement early in the life of the firm, but value-destruction later in the life of the firm. In
Table 7.K and 7.L we examine the Non-v-index of takeover defenses which can be altered by
management without a vote of shareholders. We find that early in the life of the firm, these takeover
defenses have no value enhancing possibility. However, by year 7, there is a negative and statistically
significant relationship between firm value and the Non-v-index. This result implies that takeover
defenses which the board can adopt or remove without shareholder voting may not enhance firm value,
but could still result in decreases in firm value.
In Table 8 we repeat our tests using various methods to confirm the robustness of our tests. We first
repeat our analyses in Table 7 ignoring the fact that the firm can change its takeover defenses. This test
allows us to determine if the lifecycle effect we have documented is caused by changes in the takeover
defenses of the firm or caused by changes in the firm. If our results using IPO date takeover defenses for
all analyses generate results similar to results using the actual level of takeover defenses at the firm, then
our results are driven by changes in the firm and not changes in the takeover defenses of the firm.
However, if using IPO date takeover defenses yields weaker or substantially different results, it would
imply that the changes in takeover defenses are driving the lifecycle. Table 8.A and 8.B repeat the
analyses from Table 7.A and 7.B using the IPO date takeover defenses throughout. We find that our
results are almost identical to the results generated using the actual level of takeover defenses throughout
the lifecycle of the firm. This implies that the lifecycle effect is more driven by changes in the firm than
changes in the level of takeover defenses.
The prior literature has long recognized the fact that a correlation between Tobin’s Q and takeover
defenses need not be causal (Gompers, Ishii, and Metrick (2003), Johnson, Karpoff, and Yi (2015)). The
prior literature has controlled for endogeneity either by using a regression discontinuity technique (Smith
(2015)) or by utilizing an instrumental variables technique (Karpoff, Schonlau, and Wehrly (2015)). We
choose the latter approach, using law firm indicator variables and an indicator variable if the firm is
located in California, as first suggested by Johnson, Karpoff, and Yi (2015). While we would also like to
control for the merger market as well in our instrumental variables, IVs using the merger market would
surely be correlated with the residual in our valuation regression, violating the exclusion restriction.
Therefore, we focus on law firm and location as instrumental variables.
In Table 8.C we report the second stage regressions from a 2SLS using law firm indicators and a
California indicator as our instruments. We find in model 1that the e-index is insignificantly correlated
with firm value in the overall sample of firm-year observations. When we add in an indicator variable for
22
firms more than four years from their IPO in model 2, the coefficient on e-index becomes positive and
marginally significant at 0.075. In model 3, we include both the instrumented e-index measure plus the
interaction between this measure and firms more than four years from their IPO. Importantly, the
coefficient on the instrumented interaction term is -0.456 and is significant at the 1% level. This result
confirms that our prior OLS results from Table 7 are not driven by endogeneity or omitted variables. Our
results suggest that there is a causal relationship with e-index causing higher firm valuations early in the
firm lifecycle but causing lower firm valuations later in the firm lifecycle.
In Table 8.D we repeat our analyses using 2SLS year-by-year regressions using the same
instrumental variables as in Table 8.C. We find qualitatively similar results with higher firm valuations
early in the life cycle of the firm and lower valuations later. For instance, in model 1 we examine the IPO
year results and we find that controlling for endogeneity, the coefficient on the instrumented e-index is
0.294, significant at the 10% level. The reduction in power due to the reduced sample size largely
accounts for the significance only at the 10% level. Moving through time to year 1 in model 2, we find the
coefficient is 0.114, but is no longer statistically significant, moving to later years, the coefficient
generally moves from positive to negative. In model 6 (years 7-9) we find a negative coefficient of 0.110, significant at the 10% level. We find that in years beyond this, the coefficient is -0.120, but is not
statistically significant. These results show that the apparent value reversal associated with the life cycle
theory is not driven by endogeneity.
To further analyze the robustness of our results, we run a first-difference regression in Table 8.E and
Table 8.F. While we find qualitatively similar results, the coefficients on the change in E-index and the
interaction between change in E-index and an indicator for the firm being over 4 years from its IPO is
insignificant. This may be due to the relatively low number of changes in the E-index measure. In
untabulated tests, we also follow Cremers, Litov, and Seppe (2014) by regressing firm Q onto changes in
E-index. We find with this specification, results are considerably weaker, with some evidence of a
positive relationship between changes in e-index and firm value in the early years after the IPO, but no
relationship in later years.
3.3c Alternative measures of firm maturity (perhaps this belongs in an appendix)
We now move on to measures of firm maturity other than the number of years since the IPO. One
problem with splitting our sample by years after the IPO is the arbitrary nature of this measure of IPO
firm maturity.
We begin by using the age of the IPO firm – simply the number of years from the time of IPO
founding to the present time. While firm age is linked to years since IPO, there are many firms that are
23
quite old at the time of the IPO. Therefore, firm age and years since the IPO only have a correlation of
0.18. Loughran and Ritter (2002) provide the founding dates of IPO firms and use this measure as a proxy
for the uncertainty of the IPO firms. Note that in this regression format we can only observe the firm after
it is public. This implies that if a firm is founded many years before our sample, we can only observe such
a firm as a mature firm. Nevertheless, we feel that a firm that is founded and goes through an IPO
relatively quickly (within 0-2 years) is certainly less mature than a firm that was private for many years
and subsequently went public.
In Table 9.A we tabulate our regression results using IPO firm age in years in separating firms by
their level of maturity. We include all the control variables utilized in prior regressions, but do not
tabulate them for brevity. Note that because of the discontinuous nature of firm age, the number of
observations is not perfectly consistent across all the models in Table 9.A. We find that in the early life of
the firm the relationship between e-index and firm value is positive and significant at the 1% and 10%
level in models 1 and 2. But starting as an IPO firm age of 8 years and above, the relationship between
firm value and takeover defenses becomes negative and marginally significant at the 10% level. Finally
for firm ages 12-14 (model 4), ages 15-18 (model 5), and ages 19-23 (model 6) we find a consistently
negative and statistically significant relationship between firm value and takeover defenses. We do find in
model 7 that when a firm is older than 24 years, the relationship between firm value and takeover
defenses is no longer statistically significant, but the coefficient remains negative. Other than model 7’s
insignificant coefficient, our results are entirely consistent with our prior assertion that there is a lifecycle
relationship in the takeover defense relationship with firm value.
We then move on to industry-specific measures of firm maturity. An industry where the average firm
is quite young is likely to be an immature industry where an industry with the mean firm age being old
will likely be more mature. We therefore repeat our analyses using average industry firm age to break our
sample into early, middle, and mature lifecycle industries. We first calculate the Fama and French (1997)
48 firm industry classification age. We define age as the first time the firm has a non-zero assets value on
COMPUSTAT.16 We report our results in Table 8.B. We find that firm age and industry mean age have a
correlation of 0.36. We find in model 1 that firms in an industry where the mean industry age is 0-7 years
has a positive and statistically significant relationship between the e-index and the firm valuation. As we
move to industries where the average age goes up, the coefficient on e-index becomes negative, albeit
insignificant. Starting at year 15-17 (model 6) we find a negative and statistically significant relationship
between e-index and firm value. Finally, in model 7 we find that firms in an industry where the mean firm
16
We recognize that utilizing the actual firm ages would be preferable to using ages based on when firms show up in
the in COMPUSTAT database. However, the prior literature has found that using the COMPUSTAT listing as the
founding date of firm existence is a reasonable proxy for firm age.
24
age is above 17 years are a negative and statistically significant relationship. These results show that both
firm-specific and industry-specific measures of firm maturity show consistent results to our prior patterns.
Firms early in their lifecycle show a positive relationship between firm value and e-index. Firms later in
their lifecycle show a negative relationship between firm value and e-index.
We then look at the sales growth rate of the firms in our sample. As with firm age, we calculate the
sales growth rate for both the individual firm and the Fama and French (1997) 48-industry group of firms.
For firms in an immature industry, the sales growth rate should be quite high (Klepper and Grady (1990)).
However, as the firm and the industry matures, the growth rate of sales show slow down. This allows us
to use sales growth rate as a measure of firm lifecycle for both the IPO firm and its industry as a whole.
We separate sales growth into quintiles and regress industry adjusted Tobin’s Q onto e-index plus the
control variables we have been using.17 We find in Table 9.C that for high growth firms (model 1) there is
a positive and statistically significant relationship between firm value and e-index. However, as sales
growth declines from quintile 5 to 4, the coefficient turns negative. Moving across quintiles to 3, 2, and 1,
the coefficient always becomes more negative and starting in the second quintile, becomes significantly
negative. Finally, in quintile 1, the lowest sales growth for the IPO firms, we find that the relationship
between firm value and e-index is negative and statistically significant at the 1% level. This result implies
that as a firm matures as measured by its growth rate, the relationship between firm value and e-index
becomes more negative.
We repeat our results in Table 9.D using sales growth rates by industry and find qualitatively similar
results. Once again, at the highest growth rates (quintile 5) we find a positive and statistically significant
relationship between firm value and e-index. Moving from quintile 5 to lower quintiles of industry sales
growth rates, we find qualitatively similar results as for firm sales growth rates. As the growth rate
declines, implying the firm matures, the relationship between firm value and e-index first becomes
negative and then statistically significant. Both quintile 3 and quintile 4 have a negative and statistically
significant relationship between firm value and e-index. While the lowest industry sales growth rate does
not show a statistically significant relationship between firm value and e-index, the coefficient remains
reliably negative. On the whole, these results imply a strong lifecycle impact for firms on the valuetakeover defense relationship.
3.4 Mechanism of a Life Cycle Effect of Takeover Defenses: Commitment Benefit of Takeover
Defenses
17
For the IPO firms, we use firm growth rate as (salet+1/salet -1 ) since sales in the year before the IPO (t-1) are not
available in COMPUSTAT. We find qualitatively similar results using (sale t/salet-1 -1 ) but have fewer observations
in our regressions using this measure.
25
3.4a Univariate results
To understand the mechanism of firm value enhancement, we need to get a better handle on what
happens to the firm needs over the lifecycle of the firm. Our life cycle hypothesis of takeover defenses
argues that a commitment benefit of takeover defenses is greater at the IPO stage and diminishes over a
firm’s life cycle while the agency cost of takeover defenses is small at the IPO stage and increases over a
firm’s life cycle. In this section, we empirically examine whether and to what degree the commitment
benefit of takeover defenses evolves over a firm’s life cycle and how this evolution affects the firm value.
A firm’s need for takeover defenses as a commitment mechanism toward stakeholders is greatest at the
IPO stage because of an IPO firm’s lack of reputation in the production input markets and limited
resources to build wide distribution channels. Empirical evidence support this. Johnson, Kang, and Yi
(2010) and Cen, Dasgupta, and Sen (2011) provide some hint for some of these changes. Where Johnson,
Kang, and Yi (2006) find that 60% of IPO firms disclose a large customer at the time of the IPO of a firm,
Cen, Dasgupta, and Sen (2011) find that 41% of firms have a large customer as mature firms. This result
implies that there may be a changing reliance on large customers as the firm matures. Specifically, it
appears that at the IPO stage, large customers are quite common and important, but as the firm matures,
these relationships become significantly less common or less important.18 Likewise, Johnson, Karpoff,
and Yi (2015) suggest that strategic alliances are another important value-enhancing relationship for IPO
firms.19
To document this significant change in the importance of large customers and strategic alliances, we
track the percent of IPO firms with large customers from the IPO stage to the end of our sample period.
At the IPO stage, we find that 45% of IPO firms have a large customer at the IPO stage of the firm.
However, at the later lifecycle stages, we find that the percent of IPO firms with large customers drops to
10% after ten years. Note that the beginning and ending levels are somewhat lower than those
documented in the prior literature. The difference is due to the change in the kinds of firms going public
in the Internet boom years (through 2000) compared to the kinds of firms going public after 2005. We
find results similar to Johnson, Karpoff, and Yi (2015) if we restrict our sample to pre-2005 IPO firms.
It is important to note, however, that takeover defenses are generally associated with these large
customers, as documented in the prior literature (Johnson, Karpoff, and Yi (2015)). In Table 10.A we
show that at the IPO stage, 51% of firms with above the median e-index value have a large customer
18
Johnson, Kang, and Yi (2010) also document that the presence of a large customer significantly increases the
value of the IPO firms when they go public.
19
In addition, Johnson, Karpoff, and Yi (2015) examine the presence of a dependent supplier in their analyses. We
omit this particular kind of relationship with IPO firms for two reasons. First, suppliers dependent upon IPO firms
are not all that common. Second, it is ambiguous as to whether an IPO firm would have more dependent suppliers as
it ages or fewer.
26
where only 41% of firms with below the median e-index value do. This difference is statistically
significant at the 1% level (t=4.85). Likewise, we find that in year 1, there are significantly more large
customers with firms having a high e-index level (46% versus 40%). In year 2, the results are not
statistically significant, but the pattern of high e-index firms having more large customers remains.
However, by years 5-6, the percent of firms with large customers is no longer statistically significant.
This result implies that by years 5-6 the benefits of takeover defenses in forming relationships with
trading partners is smaller than for the initial years of the IPO firm lifecycle. In Table 10.B we repeat our
analyses using classified board as our measure of takeover defenses: we find similar results. Early in the
life of the firm, having a classified board increases the likelihood of having a takeover defense. As the
firm matures, however, there is no longer a statistically significant difference in the presence of large
customers between firms with versus without a classified board. In Table 10.C we tabulate our results
using our c-index measure. For firms with above the median measure of c-index, we find that these firms
are more likely to have a large customer for the first two years after the initial public offering.
Subsequently, however, the value of the c-index takeover defenses seems to decline as generally, firms
with a higher level of c-index do not have more large customers, as shown for years 3-10. Finally, in
Table 10.D we repeat the analyses examining the percent of firms with large customers by firms with
above versus below the median v-index value. We continue to find a higher percentage of firms with large
customers when the firm has a higher v-index compared to a firm with a lower v-index.
Another important value-increasing relationship documented by Johnson, Karpoff, and Yi (2015) as
being more common with more takeover defenses is a strategic alliance. Instituting and maintaining a
strategic alliance generates a firm’s need for takeover defenses as a commitment mechanism to honor
stakeholders’ claims since strategic alliance partners’ relationship-specific investments often underlie a
strategic alliance.
We find that at the IPO stage, 19% of the IPO firms announce a strategic alliance in the first year of
the firm. We find that this percentage declines nearly monotonically with only 11% of IPO firms
announcing strategic alliances when it has been more than ten years after their IPO. This result is
generally consistent with the idea that there is a lifecycle effect for strategic alliances as well as large
customers.
In addition, we are interested in showing that firms with more takeover defenses are more likely to
announce these strategic alliances. In Table 10.E we show that generally speaking, IPO firms with more
takeover defenses are more likely to announce a strategic alliance. We find that this particular result is
only significant for certain years after the IPO. However, as a general rule, firms with more takeover
defenses are more likely to announce strategic alliances.
27
In Table 10.F we report that firms with classified boards are significantly more likely to announce a
strategic alliance. These results are significant for all years after the IPO until ten years. Our results are
not novel in that the prior literature (Johnson, Kang, and Yi (2010)), and Johnson, Karpoff, and Yi
(2015)) has documented that important value-increasing relationships are more common for firms with
certain takeover defenses. However, our results do show for the first time that the commonness of these
relationships declines significantly in the years after the IPO. Additionally, in Table 10.G we show that
having above the median level of takeover defenses as measured by c-index generally results in a firm
being more likely to have a strategic alliance. This result holds for all periods out to ten years after the
IPO. However, past ten years, it appears that firms with a high c-index measure are no longer more likely
to have a strategic alliance. In Table 10.H we repeat our analyses using the v-index measure of takeover
defenses. As with prior measures, we find qualitatively similar results. This suggests a strong role for
takeover defenses in enhancing relationship formation early in the lifecycle of the firm, but not later.
3.4b Firm value and important relationships
Firms early in their lifecycle may more easily form value-increasing relationships by adopting higher
levels of takeover defenses. However, as the firm matures, it may be difficult to remove these takeover
defenses and consistent with Stout (2002) and Bebchuk, Cohen, and Ferrell (2009) these defenses may
cause a decline in firm value. To document this effect, we must use our panel dataset of firm values,
relationships, and takeover defenses. This allows us to document the changing nature of takeover defenses
and their impact on firm value as the firm ages.
We begin with the baseline regression from Table 7.A model 1 which controls for the known
determinants of firm value, including the lifecycle impact of takeover defenses. In Table 11.A model 1 we
simply include an indicator variable for having a large customer and an interaction term between having a
large customer and e-index. Consistent with the prior literature (Johnson, Kang, and Yi (2010)), we find
that firms with more takeover defenses have a particularly higher valuation when the firm also has
important relationships such as a large customer. We do find that the coefficient on having a large
customer is negative and statistically significant, but this is due to the panel data nature of our dataset. To
ensure that our results are consistent with the prior literature, we repeat our regression at the IPO stage
and find that the customer coefficient is 1.02 and significant at the 1% level (t-stat = 5.01). Thus, we can
see that the negative coefficient for customer is largely driven as the firm ages.
In Table 11.A model 2 we consider what might happen when a firm’s customer base becomes diffuse
and no longer has a large customer accounting for more than 10% of its sales. We therefore create an
indicator variable taking a value of one if the firm is no longer reliant upon its large customer. All
28
subsequent observations in the panel are coded as one unless the firm develops a new large customer. We
find are particularly interested in what happens when the firm has a high level of takeover defenses, but
no longer needs them for bonding to its large customers. In this case, we find that the interaction between
no longer being reliant on the large customer and the e-index is negative and statistically significant at the
1% level with a value of -0.143. This result implies that when the firm no longer needs the bonding
purpose of the takeover defense, it will cause a decline in firm value. This result also implies that as a
firm’s customer-base diffuse over a firm’s life cycle, the benefit of takeover defenses will decline. This is
exactly what our life cycle hypothesis of takeover defenses argues.
In Table 11.A model 3 we repeat our analyses using both the large customer interaction with e-index
and the large customer reliance gone interaction with e-index and find that our results are unchanged.
After a firm loses its dependence upon its large customer, the takeover defenses it adopts are no longer
value increasing and in fact become value decreasing. In Table 11.B we repeat our analyses replacing eindex as our measure of takeover defenses with c-index. We find qualitatively similar results throughout,
with c-index resulting in a lower firm value after the relationship with customers is no longer present in
the data. Finally, in Table 11.C we use v-index as our measure of takeover defenses and find qualitatively
similar results.
In Table 11.D we repeat our analyses using our strategic alliance measure. In model 1 we find that
the value of the firm is always higher when the firm has a strategic alliance and adopts more takeover
defenses, consistent with Johnson, Karpoff, and Yi (2015). We then examine the impact of a year with no
strategic alliance and find in model 2 that this results in a lower firm value when the firm has more
takeover defenses. This result is consistent with the results in Table 11.A model 2. We find a coefficient
of -0.086, significant at the 1% level for the interaction term between the strategic alliance gone indicator
and the e-index. Finally, in Table 11.D model 3 we find that including all our indicators for having a
strategic alliance and for strategic alliances being gone with their interactions with e-index, we see that
the interactions between the strategic alliance being gone and the e-index is -0.057 and significant at the
10% level. We also repeat our analyses using our c-index measure for takeover defenses in Table 11.E
models 1-3. We find similar results with higher levels of c-index being associated with lower firm value
after the joint venture relationships are no longer present. Likewise, in Table 11.F we utilize the v-index
measure and find qualitatively similar results. As with our analyses concerning large customers, it appears
that takeover defenses are value increasing in the presence of strategic alliances, but the positive impact
of these takeover defenses fades as the firm matures.
3.5 Mechanism of a Life Cycle Effect of Takeover Defenses: Agency Cost of Takeover Defenses
29
3.5a Univariate results regarding insider ownership
Thus far, we have shown that takeover defenses can create value when they are adopted in the
presence of important relationships such as large customers and strategic alliances. But this raises an
important question about why takeover defenses destroy value for mature firms. To better understand this
effect, we examine the impact of insider ownership on the value impact of takeover defenses. Insider
ownership may serve as a mechanism to align managerial incentives to those of shareholders, ensuring
that even for a mature firm, having takeover defenses does not destroy firm value (Morck, Shleifer, and
Vishny (1988)). Our life cycle of takeover defense hypothesis argues that as the firm matures, the
shareholdings of the firm will become more diffuse, aggravating managerial agency costs. We therefore
begin by examine the insider ownership of our firms to look for regularities regarding ownership that may
lead us to clues regarding the source of value destructions regarding takeover defenses.
We begin by examining the insider holdings as the firm matures for firms with above versus below
the median level of takeover defenses. We find in Table 12.A that there is a clear decline in the insider
holdings from the time of the IPO until ten years later. For instance, firms with below the median number
of takeover defenses go from 42% insider holdings at the IPO stage to 16% ten years later. Of particular
note is the large decline after the first year, from 42% to 34%. This is likely due to the end of the lock-up
period which is associated with large shareholder sales by insiders. The pattern of declining insider
ownership is the same for firms with above the median number of takeover defenses which starts at 41%
ownership and declines to 11% ownership by year 10. However, it is interesting to note that by year 2, the
ownership for firms with above the median number of takeover defenses have a lower percent of
ownership compared to firms with above the median number of takeover defenses. This difference is
statistically significant beginning in year 2 and continues through the balance of our observations.
Likewise, in Table 12.B we find that using a classified board as our measure of takeover defenses, firms
with a classified board tend to own fewer shares in the firm. This difference is statistically significant
every year subsequent to the year the firm goes public. In Table 12.C we repeat our analyses using the cindex as our takeover defense measure. We find consistent with prior results that after the IPO year and
the next two years, firms with above the median c-index have a lower level of insider ownership. Finally,
in Table 12.D we repeat our analyses using v-index and find qualitatively similar results. The first two
years after the IPO the firm has similar levels of insider ownership whether it has high or low v-index.
However, in subsequent years, the insiders of firms with a lower v-index have a significantly higher level
of stock ownership. This pattern is consistent with the idea that insiders at firms with high takeover
defenses are more likely to sell their shares to allow them to consume more perquisites from the
shareholders of the firm without bearing the consequences.
30
3.5b Multivariate results regarding insider ownership
To better understand the value impact of insider share holdings, we repeat our analyses using the
industry adjusted Tobin’s Q as the dependent variable in a multivariate regression. We then create an
indicator variable taking a value of one if the IPO firm has above the median percent of insider holdings
and zero otherwise. This allows us to determine the value impact of insider holdings in the IPO firm.
In Table 13.A model 1 we repeat our base case regression, including e-index, an indicator for when
insiders have below the median amount of shares, and the interaction between the two. We find in this
case that the coefficient on the interaction is negative and statistically significant, implying that when
insiders have below median ownership, the presence of takeover defenses reduces firm value. In model 2
we examine changes in ownership since as a firm matures, the insider ownership of the firm will decline,
resulting in exacerbating managerial moral hazard. We create an indicator variable taking a value of one
when the insiders of the firm sell more than the median amount in shares in the firm. We interact this
indicator with e-index and find that the coefficient on the interaction is negative and statistically
significant at the 1% level. This implies that as the firm matures, if managers sell their positions in the
firm, the valuation of firms with high levels of e-index are more negative. Finally, in model 3 we include
both sets of indicators (below median holdings and above median sales) along with their interactions with
e-index. We continue to find a negative and statistically significant relationship with the firm value for the
interaction between large insider sales and e-index.
In Table 13.B we repeat our analyses using c-index instead of e-index and find qualitatively similar
results with the interaction between c-index and high insider sales resulting in a negative and statistically
significant coefficient. Finally, in Table 13.C we repeat our analyses using v-index as our measure of
takeover defenses. Once again, we find qualitatively similar results, with the interaction between having a
large decline in insider shares and v-index being negative and statistically significant. This result is
consistent with the sales of insider stock holdings being associated with a decline in firm value when the
firm has a high level of takeover defenses.20
4. Robustness Checks
To confirm the robustness of our general results, we repeat our major tests using various measures of
Tobin’s Q. For instance, we repeat all our lifecycle tests using raw (unadjusted) Tobin’s Q instead of the
industry adjusted Tobin’s Q reported in the paper. We find qualitatively similar results. In addition, we
20
We find that omitting the continuous variable, insider ownership from our regressions in Table 13 does not
appreciably impact our results.
31
repeat our analyses using Tobin’s Q adjusted by industry median results that are not winsorized. The
major results tabulated in the paper all utilize industry median adjusted Tobin’s Q winsorized at the 99th
percentile. When we repeat our analyses using industry adjusted Tobin’s Q without winsorizing, we find
remarkably consistent results. Likewise, when we repeat our analyses using winsorized values at the 95th
or the 90th percentile, we find similar results. In all cases, we find a striking pattern of value increase with
takeover defenses early in the lifecycle of the firm and value decrease with takeover defenses later in the
lifecycle of the firm.
We also examine the industry clustering of the lifecycle effect to determine if our results are
disproportionately caused by financial firms, technology firms, or other kinds of firms. We find that the
lifecycle effect persists even in financial firms (sic code 6000-6999), technology firms (as defined by
Loughran and Ritter (2004) Appendix 4), and many other industry definitions. Likewise, our results are
robust to the elimination of all utilities IPO firms. We repeat our analyses for each of the Fama and
French (1997) 10 industry definitions and find that in all cases but one, when there are more than 20
observations, the coefficient on e-index goes from positive at time=0 and remains positive for 1-2 years
before turning negative in the later years of the firm lifecycle. Most of these coefficients are not
significant, however, due to the limited sample size in these regressions. These results suggest that our
results are not driven by any industry per se, but are driven by a general trend in the data across all
industries.
In addition, we repeat our analyses on subsamples of IPO firms from before and after the Internet
boom period of 2000. Repeating our analyses on only IPOs from before or only after the Internet boom
period results in qualitatively similar results. In the early years after the IPO, there is a strong and
significant relationship between the level of takeover defenses and firm value as measured by Tobin’s Q.
However, in later years, there is a strong negative and significant relationship between takeover defenses
and Tobin’s Q. Thus, it appears that our major results are not driven by any particular time period for the
IPOs in our sample.
Likewise, we repeat our analyses examining the e-index measure without golden parachute. Because
golden parachute goes from being adopted at the IPO stage 64% of the time to nearly 100% of the time,
perhaps our results are driven by the change in this particular takeover defense. We therefore repeat our
analyses using an e-index without the golden parachute measure. When we do so, we find that there is
still a strong and significant lifecycle effect. In the early years after an IPO, the firms still have a positive
and significant relationship between takeover defenses and Tobin’s Q but in later years, the relationship
becomes negative and significant.
32
We also repeat our analyses on the subset of IPO firms with missing insider holdings. Even on this
subset of firms, we find a positive and significant relationship between takeover defenses and firm value
early in the life of the firm. Late in the life of these firms, we find a negative relationship between
takeover defenses and firm value, although the relationship is not statistically significant due to the small
sample size of the firms. In addition, we repeat our analyses for the subset of firms with above (below)
the median level of insider holdings and find a significant value reversal for these firms. Likewise, when
we repeat our analyses for firms with only above (below) the median level of institutional shareholders,
we find a significant value reversal.
Additional tests and robustness checks are contained in the appendix.
5. Conclusions
In this paper, we propose and empirically examine a life cycle hypothesis of takeover defenses which
suggests that the bonding benefits and the agency costs of takeover defenses affect the valuation effect of
takeover defenses differently at different stages of a firm’s life cycle. The life cycle hypothesis of
takeover defenses predicts that there is a reversal in the sign of a relationship between takeover defenses
and firm value as a firm matures from a positive one to a negative one. Consistent with the life cycle
hypothesis of takeover defenses, we find that there is a reversal in the relationship between takeover
defenses and firm value over a firm’s life cycle. We also find that the bonding benefit which is especially
valuable at the IPO stage and the agency costs of takeover defenses which aggravate managerial moral
hazard as manager ownership declines are two significant determinants of the valuation effect of takeover
defenses over a firm’s life cycle. We also find that there are two categories of takeover defenses which
exhibit different dynamic patterns of adoption and removal, which suggests that individual takeover
defenses are adopted and removed at different stage in a firm’s life for different purposes. Since aggregate
score takeover defense indices including G-index and e-index cannot capture this systematic differences
in the nature of individual takeover defenses, we devise a c-index and find that it is hard-to-revise
takeover defenses which have a commitment nature that causes a reversal in the relationship between
takeover defenses and firm value over a firm’s life cycle. Furthermore, we find that a firm’s need for
promoting important customer-supplier relationships and strategic alliances is greater at the IPO stage and
takeover defenses enhance firm value at the IPO stage by satisfying the bonding needs of a firm. Also, we
find that the agency cost of takeover defenses increase over a firm’s life and agency cost of takeover
defenses make takeover defenses value-destroying at a later stage of a firm’s life. All the aforementioned
findings support the life cycle hypothesis of takeover defenses. Our empirical investigation of the life
cycle hypothesis of takeover defenses reconciles inconsistencies in the empirical results reported in the
33
previous literature and sheds new light on some aspect of the corporate governance implications of
takeover defenses which have attracted less academic attention in the previous literature. Furthermore,
our paper explains how it is possible that a value-maximizing firm may adopt takeover defenses while
takeover defenses are value destroying.
34
References
Akerlof, George A., 1970, The market for lemons: Quality uncertainty and the market mechanism,
Quarterly Journal of Economics 84, 488–500.
Agrawal, A, and C. Knoeber, 1998, Managerial compensation and the threat of takeover, Journal of
Financial Economics 47, 219-239.
Arlen, Jennifer, 2006, Regulating post-bid embedded defenses: Lessons from Oracle and PeopleSoft,
Harvard Negotiation Law Review.
Balakrishnan, S. and I. Fox, 1993, Asset specificity, firm heterogeneity and capital structure, Strategic
Management Journal 14, 3-16.
Bebchuk, Lucian A., 2013, The myth that insulating boards serves long-term value, Columbia Law Review
113, 1637-1694.
Bebchuk, Lucian A., John C. Coates and Guhan Subramanian, 2002, The powerful antitakeover force of
staggered boards: further findings and a reply to symposium participants, Stanford Law Review 55,
885-917.
Bebchuk, Lucian A., Alma Cohen, and Allen Ferrell, 2009, What matters in corporate governance? Review
of Financial Studies 22, 738-827.
Boone, A., L. Field, J. Karpoff, and C. Raheja, 2007, The determinants of corporate board size and
composition: An empirical analysis, Journal of Financial Economics 85, 67-101.
Cen, Ling, S. Dasgupta, and R. Sen, 2013, Discipline or disruption? Stakeholder relationships and the effect
of takeover threat, University of Toronto working paper available at http://ssrn.com/abstract=1540346.
Chan, S. H., J. Kensinger, A. Keown, and J. Martin, 1997, Do strategic alliances create value?, Journal of
Financial Economics 46, 199-221.
Chemmanur, T.J. and Y. Jiao, 2012, Dual class IPOs: A theoretical analysis, Journal of Banking and
Finance 36, 305-319.
Chemmanur, T.J. and I. Paeglis, and K. Simonyan, 2011, Management quality and anti-takeover provisions,
Journal of Law and Economics 54, 651-692.
Chemmanur, T.J. and E. Loutskina, 2006, The role of venture capital backing in initial public offerings:
Certification, screening, or market power, Boston College working paper.
Coase, R. H., 1960, The problem of social cost, Journal of Law and Economics 3, 1-44.
Coase, R. H., 2000, The acquisition of Fisher Body by General Motors, Journal of Law and Economics 43,
15-32.
35
Coates, J., 2001, Explaining variation in takeover defenses: Blame the lawyers, California Law Review 89,
1301-1422.
Coates, J., and Kraakman, R., 2011, The link between the acquisitions market and the market for CEOs,
Harvard Law School working paper.
Cremers, K., J. M., V. B. Nair, and U. Peyer, 2006, Weak shareholder rights: a product market rationale,
Yale Working paper.
Cremers, K., J. M., V. B. Nair, and U. Peyer, 2008, Takeover defenses and competition, Journal of
Empirical Legal Studies 5, 791-818.
Cremers, Martijn and L. P. Litov, and S.M. Sepe, 2014, Staggered boards and firm value, revisited.
University of Notre Dame working paper, http://ssrn.com/abstract=2364165.
Daines, Robert and Michael Klausner, 2001, Do IPO charters maximize firm value? Antitakeover
protection in IPOs, Journal of Law, Economics, and Organization 17, 83-120.
DeAngelo, Harry and Edward M. Rice, 1983, Antitakeover charter amendments and shareholders wealth,
Journal of Financial Economics 11, 329-60.
Denis, David J., D. K. Denis, and A. Sarin, 1997, Agency problems, equity ownership, and corporate
diversification, Journal of Finance 52, 135-160.
Easterbrook, Frank H., and Daniel R. Fischel, 1991, The economic structure of corporate law, Harvard
University Press (Cambridge, MA).
Fee, C. E., C. Hadlock, and S. Thomas, 2006, Corporate equity ownership and the governance of product
market relationships, Journal of Finance 61, 1217-1251.
Fee, C. E., S. Thomas, 2004, Sources of gains in horizontal takeovers: Evidence from customer, supplier,
and rival firms, Journal of Financial Economics 74, 423-460.
Field, L.S. and G. Hanka, 2001, The expiration of IPO share lockups, Journal of Finance 56, 101-117.
Field, L.S. and J. M. Karpoff, 2002, Takeover defenses of IPO firms, Journal of Finance 57, 1857-1889.
Gillan, S. L., Hartzell, J. C., and R. Parrino, 2009, Explicit vs. implicit contracts: Evidence from CEO
employment agreements, Journal of Finance 64, 1629-1655.
Giroud, X., and Holger Mueller, 2011, Corporate governance, product market competition, and equity
prices, Journal of Finance 66, 563-600.
Gompers, Paul, Joy Ishii, and Andrew Metrick, 2003, Corporate governance and equity prices, Quarterly
Journal of Economics 118, 107-155.
Grossman, S. J., and O. D. Hart, 1980, Takeover Bids, the Free-Rider Problem, and the Theory of the
Corporation, The Bell Journal of Economics 11, 42-64.
36
Grossman, S. J., and O. D. Hart, 1986, The cost and benefits of ownership: A theory of vertical and lateral
integration, Journal of Political Economy 4, 691-719.
Hannes, Sharon, 2006, A demand-side theory of antitakeover defenses, Journal of Legal Studies 35(2), 475524.
Harris, E., 1990, Antitakeover measures, golden parachutes, and target firm shareholder welfare, RAND
Journal of Economics 21, 614-625.
Hart, O. D., and J. Moore, 1990, Property rights and the nature of the firm, Journal of Political Economy 98,
1119-1158.
Hertzel, M., Li, Z., Officer, M. and Rodgers, K., 2008, Inter-firm linkages and the wealth effects of
financial distress along the supply chain, Journal of Financial Economics 87, 374-387.
Hsu, H-C., A.V. Reed, and J. Rocholl, 2010, The new game in town: Competitive effects of IPOs, Journal
of Finance 65(2), 495-528.
Humphery-Jenner M., 2014, Takeover defenses, innovation, and value creation: Evidence from acquisition
decisions, Strategic Management Journal 35, 668-690.
Hwang, B. and S. Kim, 2009, It pays to have friends, Journal of Financial Economics 93, 138-158.
Jain, B., and O. Kini, 1994, The post-issue operating performance of IPO firms, Journal of Finance 49,
1699-1726.
Jensen, Michael C., 1986, Agency costs of free cash flow, corporate finance, and takeovers, American
Economic Review 76, 323-329.
Johnson, W. C., J. Kang, and S.Yi, 2010, The certification role of large customers in the new issues market,
Financial Management 39, 1425-1474.
Johnson, W. C., J. Karpoff, and S.Yi, 2015, The bonding hypothesis of takeover defenses: Evidence from
IPO firms, Journal of Financial Economics 117, 307-332.
Johnson, W. C., J. Kang, Masulis, R. W., and S.Yi, 2011, Supply-chain spillover effects and the
interdependence of firm financing decisions, Working paper, University of New South Wales.
Joskow, P. L., 1987, Contract duration and relationship-specific investments: Empirical evidence from coal
markets, The American Economic Review 77, 168-185.
Kim, M., and J.R. Ritter, 1999, Valuing IPOs, Journal of Financial Economics 53, 409-37.
Kadyrzhanova, Dalida and Matthew Rhodes-Kropf, 2011, Concerntrating on governance, Journal of
Finance 66, 1649-1685.
Klausner, Michael D., 2003, Institutional shareholders, private equity, and antitakeover protection at the
IPO stage. University of Pennsylvania Law Review, Vol. 152.
37
Klein, B., Crawford, R. G., Alchian, A. A., 1978. Vertical integration, appropriable rents, and the
competitive contracting process. Journal of Law and Economics 21, 297-326.
Klein, B., 1988, Vertical integration as organizational ownership: The Fisher Body - General Motors
relationship revisited, The Journal of Law, Economics and Organization 4, 199-213.
Krackhardt, D., 1992, The strength of strong ties: The importance of philosophy in organizations, In N.
Nohria & R. G. Eccles (Eds.), Networks and organizations: Structure, form, and action: 216-239.
Boston: Harvard Business School Press.
Linn, S.C., and J.J. McConnell, 1983, An empirical examination of antitakeover amendments on common
stock prices, Journal of Financial Economics 11, 361-399.
Lipton, Martin, 2002, Pills, polls, and professors redux, The University of Chicago Law Review 69: 1037.
Masulis, R.W., C. Wang, and F. Xie, 2007, Corporate governance and acquirer returns, The Journal of
Finance 62 (4), 1851–1889.
Pontiff, J., A. Shleifer, and M. S. Weisbach, 1990, Reversions of excess pension assets after takeovers,
Rand Journal of Economics 21, 600-613.
Posner, R. A., 1975, The social costs of monopoly regulation, Journal of Political Economy 83, 807-827.
Purnanandam, A.K., and B. Swaminathan, 2004, Are IPOs really underpriced?, Review of Financial Studies
17, 811-48.
Shapiro, C., 1983, Premiums for high quality products as returns to reputations, Quarterly Journal of
Economics 98, 659-79.
Shleifer, A. and L. Summers, 1988, Breach of trust in hostile takeovers, in A.J. Auerbach (ed.),. Corporate
Takeovers: Causes and Consequences (University of Chicago Press, Chicago).
Smart, Scott B., Ramabhadran S. Thirumalai, and Chad J. Zutter, 2008, What's in a vote? The short- and
long-run impact of dual-class equity on IPO firm values, Journal of Accounting and Economics 45, 94115.
Smith, Erin, 2013, Do shareholders want less governance? New York University working paper.
Stein, J., 1988, Takeover threats and managerial myopia, Journal of Political Economy 96, 61-80.
Stout, L., 2002, Do antitakeover defenses decrease shareholder wealth? The ex post / ex ante valuation
problem, Stanford Law Review 55, 845-861.
Stulz, R., 1988, Managerial control and voting rights: Financial policies and the market for corporate
control, Journal of Financial Economics 20, 25-54.
Tian, Xuan, 2011, The causes and consequences of venture capital stage financing, Journal of Financial
Economics 101, 132-159.
38
Titman, S. and R. Wessels, 1988, The determinants of capital structure choice, Journal of Finance 43, 1-19.
Williamson, O. E., 1979, Transaction cost economics: The governance of contractual relations, Journal of
Law and Economics 22, 233-261.
Williamson, O. E., 1985, The economic institutions of capitalism (The Free Press, N.Y.).
39
Table 1.
Sample of firms
Our sample consists of 2,285 IPO firms from 1997-2011. We eliminate all REITS, ADRs, funds, firms
without CRSP and COMPUSTAT coverage, firms incorporated outside the United States, firms with a
dual class share structure, and firms not in Jay Ritter’s database of firms with a founding date. In addition,
we eliminate all IPO firms that do not have prospectus filings, annual reports, and proxy statement filings
available in the SEC’s EDGAR database.
IPO Year
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
N
384
249
410
310
64
49
59
153
129
138
153
16
36
76
59
Total
2,285
40
Table 2.
IPO firm summary statistics at the time for the IPO and in subsequent periods
Our sample consists of 2,285 IPO firms from 1997-2011. We eliminate all REITS, ADRs, funds, firms
without CRSP and COMPUSTAT coverage, firms incorporated outside the United States, firms with a
dual class share structure, and firms not in Jay Ritter’s database of firms with a founding date. In addition,
we eliminate all IPO firms that do not have prospectus filings, annual reports, and proxy statement filings
available in the SEC’s EDGAR database.
Variable
N
Total Assets ($ millions)
Market capitalization ($ millions)
Firm age
S&P 500 firm (indicator)
Delaware incorporation (indicator)
Insider holdings
Institutional shareholder holdings
Mean
At year = 0
2,285
867.22
824.92
15.00
0.00
0.79
0.42
0.08
41
Mean
At year = 5
1,179
1,948.10
914.19
21.41
0.02
0.79
0.21
0.41
Mean
At year = 10
509
5,110.10
1,573.19
25.46
0.06
0.76
0.16
0.50
Mean
At year = 14
235
6,592.31
2,933.45
29.94
0.07
0.71
0.14
0.54
Table 3.
Percent of IPO firms with takeover provisions by years from IPO
Our sample consists of 2,285 IPO firms from 1997-2011. We eliminate all REITS, ADRs, funds, firms
without CRSP and COMPUSTAT coverage, firms incorporated outside the United States, firms with a
dual class share structure, and firms not in Jay Ritter’s database of firms with a founding date. In addition,
we eliminate all IPO firms that do not have prospectus filings, annual reports, and proxy statement filings
available in the SEC’s EDGAR database.
Panel A. Individual takeover defenses and e-index for IPO firms from 1997-2011
Years
N
Classified
Poison
Supermajority
Supermajority
from IPO
board
Pill
to amend bylaw
to amend
charter
0
2,285
65.30%
5.34%
33.83%
30.07%
1
2,098
65.44%
9.10%
34.94%
31.27%
2
1,829
65.50%
12.52%
36.52%
32.64%
3
1,568
64.54%
15.94%
35.91%
32.02%
4
1,351
63.66%
18.50%
36.27%
32.20%
5
1,179
63.61%
21.20%
36.81%
32.74%
6
1,055
63.32%
22.94%
36.68%
32.04%
7
901
60.93%
24.08%
35.85%
31.30%
8
755
61.85%
26.49%
35.36%
30.99%
9
615
60.81%
26.83%
34.80%
29.59%
10
509
60.71%
27.50%
34.18%
28.88%
11
443
59.82%
25.73%
33.63%
28.89%
12
387
57.88%
23.51%
32.82%
27.65%
13
331
58.01%
23.56%
32.02%
26.89%
14
235
53.19%
18.30%
33.19%
30.21%
≥ 15
199
47.23%
16.08%
32.16%
31.16%
Total
15,740
63.23%
16.61%
35.26%
31.15%
Supermajority
to approve
mergers
40.61%
42.23%
44.51%
46.05%
47.08%
48.43%
48.82%
51.50%
53.91%
55.28%
55.99%
57.11%
55.81%
55.89%
55.32%
49.75%
Golden
Parachute
e-index
64.46%
65.63%
67.91%
70.34%
71.72%
73.79%
74.50%
79.25%
84.50%
89.59%
98.23%
97.97%
98.19%
97.58%
97.02%
96.98%
2.40
2.49
2.60
2.65
2.69
2.77
2.78
2.83
2.93
2.97
3.05
3.03
2.96
2.94
2.87
2.73
47.34%
74.85%
2.68
Golden
Parachute
e-index
53.86%
53.87%
54.04%
54.59%
56.75%
58.28%
58.48%
61.43%
73.10%
83.72%
97.75%
97.54%
97.66%
97.06%
96.97%
97.71%
2.05
2.13
2.24
2.33
2.40
2.48
2.54
2.53
2.72
2.82
3.01
2.98
2.92
2.89
2.86
2.77
66.10%
2.46
Golden
e-index
Panel B. Individual takeover defenses and e-index for IPO firms from January 1997-March 2000
Years
N
Classified
Poison
Supermajority
Supermajority
Supermajority
from IPO
board
Pill
to amend bylaw
to amend
to approve
charter
mergers
0
1,140
62.28%
5.96%
28.07%
22.54%
33.16%
1
1,021
62.29%
10.28%
29.19%
23.51%
34.28%
2
866
62.36%
13.86%
31.52%
25.40%
36.95%
3
751
62.85%
17.84%
32.36%
26.63%
39.28%
4
659
62.22%
21.24%
32.78%
27.16%
40.52%
5
580
62.24%
23.62%
33.79%
28.28%
42.07%
6
525
63.43%
25.52%
34.48%
29.14%
43.24%
7
477
61.22%
24.31%
33.75%
28.72%
43.82%
8
435
60.46%
25.29%
34.48%
29.66%
49.89%
9
387
58.14%
25.84%
33.59%
28.94%
52.20%
10
355
58.87%
25.35%
33.80%
29.30%
56.62%
11
325
57.23%
24.31%
33.23%
29.23%
56.92%
12
299
55.52%
21.74%
33.11%
28.43%
56.19%
13
272
55.15%
20.96%
32.35%
28.31%
55.51%
14
231
53.25%
18.18%
32.90%
29.87%
54.98%
≥ 15
199
50.38%
16.08%
32.06%
29.77%
51.15%
Total
8,522
60.67%
17.94%
31.95%
26.79%
42.71%
Panel C. Individual takeover defenses and e-index for IPO firms from April 2000 – December 2011
Years
N
Classified
Poison
Supermajority
Supermajority
Supermajority
42
from IPO
board
Pill
to amend bylaw
0
1
2
3
4
5
6
7
8
9
10
11
12
≥ 13
1,145
1,077
963
817
692
599
530
424
320
228
154
118
88
63
68.30%
68.43%
68.33%
66.10%
65.03%
64.94%
63.21%
60.61%
63.75%
65.35%
64.94%
66.95%
65.91%
71.19%
4.72%
7.99%
11.32%
14.20%
15.90%
18.86%
20.38%
23.82%
28.13%
28.51%
32.47%
29.66%
29.55%
34.92%
39.56%
40.39%
41.02%
39.17%
39.60%
39.73%
38.87%
38.21%
36.56%
36.84%
35.06%
34.75%
31.82%
30.51%
to amend
charter
37.55%
38.63%
39.15%
36.96%
36.99%
37.06%
34.91%
34.20%
32.81%
30.70%
27.92%
27.97%
25.00%
20.34%
Total
7,218
66.25%
15.03%
39.19%
36.30%
43
to approve
mergers
48.03%
49.77%
51.30%
52.26%
53.32%
54.59%
54.34%
60.14%
59.38%
60.53%
54.55%
57.63%
54.55%
57.63%
Parachute
74.93%
76.69%
80.27%
84.70%
85.98%
88.65%
90.38%
99.29%
99.69%
99.56%
99.35%
99.15%
100.00%
100.00%
2.73
2.82
2.91
2.93
2.97
3.04
3.02
3.16
3.20
3.21
3.14
3.16
3.07
3.15
52.80%
85.09%
2.95
Table 4.
Number of IPO firms adopting and removing takeover defenses by years from IPO
Our sample consists of 2,285 IPO firms from 1997-2011. We eliminate all REITS, ADRs, funds, firms
without CRSP and COMPUSTAT coverage, firms incorporated outside the United States, firms with a
dual class share structure, and firms not in Jay Ritter’s database of firms with a founding date. In addition,
we eliminate all IPO firms that do not have prospectus filings, annual reports, and proxy statement filings
available in the SEC’s EDGAR database.
Panel A. IPO firms adopting new takeover defenses
Years
N
Classified
Poison
Supermajority
from IPO
board
Pill
to amend bylaw
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
≥15
2,285
2,098
1,829
1,568
1,351
1,179
1,055
901
755
615
509
443
387
331
235
199
0
4
1
1
1
1
0
0
0
0
0
0
0
0
0
0
20
81
66
44
35
28
15
4
12
6
6
5
5
3
1
3
0
0
1
0
0
0
0
0
1
0
1
0
0
0
0
0
Supermajority
to amend
charter
0
0
1
0
0
0
0
0
1
0
0
0
0
0
0
0
Total
15,740
8
334
3
2
170
380
Supermajority
to amend bylaw
Supermajority
to approve
mergers
0
12
1
4
3
3
4
4
7
4
8
2
3
0
0
2
Golden
Parachute
57
19
Panel B. IPO firms removing takeover defenses
Years
N
Classified
Poison
from IPO
board
Pill
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
≥15
2,285
2,098
1,829
1,568
1,351
1,179
1,055
901
755
615
509
443
387
331
235
199
0
4
0
7
9
4
9
8
9
9
12
9
8
9
7
3
0
0
1
0
2
0
0
1
5
2
11
14
17
10
8
4
0
2
1
3
0
2
3
2
4
4
3
4
2
1
1
2
Supermajority
to amend
charter
0
2
2
4
0
2
3
0
3
2
1
4
1
1
1
1
Total
15,740
107
75
34
27
44
Supermajority
to approve
mergers
0
11
14
20
9
5
11
34
31
13
21
0
0
1
0
0
Golden
Parachute
0
39
35
47
21
17
16
59
55
38
52
0
0
0
0
0
0
0
4
3
1
2
1
2
0
2
1
0
1
1
0
1
Table 5.
IPO firm Tobin’s Q and industry adjusted Tobin’s Q by years from IPO
Our sample consists of 2,285 IPO firms from 1997-2011. We eliminate all REITS, ADRs, funds, firms
without CRSP and COMPUSTAT coverage, firms incorporated outside the United States, firms with a
dual class share structure, and firms not in Jay Ritter’s database of firms with a founding date. In addition,
we eliminate all IPO firms that do not have prospectus filings, annual reports, and proxy statement filings
available in the SEC’s EDGAR database. We report the mean and median industry adjusted Tobin’s Q for
the years after the IPO. Tobin’s Q is winsorized at the 99th percentile.
Years
from IPO
N
IPO firm Q
mean
IPO firm Q
median
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
≥15
2,285
2,098
1,829
1,568
1,351
1,179
1,055
901
755
615
509
443
387
331
235
199
4.17
2.50
2.28
2.29
2.34
2.48
2.50
2.31
2.25
2.28
2.39
2.29
2.74
2.30
3.08
2.39
2.48
1.68
1.49
1.52
1.60
1.62
1.72
1.65
1.55
1.52
1.53
1.36
1.43
1.47
1.45
1.59
Industry
adjusted Q
mean
2.25
0.73
0.57
0.59
0.62
0.73
0.64
0.49
0.55
0.58
0.77
0.71
1.05
0.46
1.35
0.52
Total
15,740
2.65
1.67
0.88
45
Industry
adjusted Q
median
0.58
0.02
-0.02
-0.01
0.00
0.00
0.01
0.02
-0.01
-0.01
0.00
-0.02
-0.01
-0.03
-0.01
0.01
0.03
Table 6.
IPO firm industry adjusted Tobin’s Q by years from IPO and takeover defenses
Our sample consists of 2,285 IPO firms from 1997-2011. We eliminate all REITS, ADRs, funds, firms
without CRSP and COMPUSTAT coverage, firms incorporated outside the United States, firms with a
dual class share structure, and firms not in Jay Ritter’s database of firms with a founding date. In addition,
we eliminate all IPO firms that do not have prospectus filings, annual reports, and proxy statement filings
available in the SEC’s EDGAR database. Tobin’s Q is winsorized at the 99th percentile. c-index is the
sum of the four takeover defenses that firms cannot change without a shareholder vote: classified board,
supermajority restrictions on amending bylaws and charters, and supermajority requirements to approve
mergers.
Panel A. Industry adjusted Tobin’s Q by classified board and years from IPO
Years from IPO
N
Industry adjusted Q
Industry adjusted Q
0
1
2
3-4
5-6
7-9
≥10
2,285
2,098
1,829
2,919
2,234
2,271
2,104
Below median e-index
Firms
2.09
0.59
0.52
0.64
0.94
0.73
1.48
Above median e-index
Firms
2.51
0.92
0.63
0.56
0.44
0.37
0.28
Total
15,740
1.99
1.30
Panel B. Industry adjusted Tobin’s Q by classified board and years from IPO
Years from IPO
N
Industry adjusted Q
Industry adjusted Q
0
1
2
3-4
5-6
7-9
≥10
2,285
2,098
1,829
2,919
2,234
2,271
2,104
No Classified Board
Firms
1.82
0.71
0.66
0.64
0.99
0.71
1.42
Classified Board
Firms
2.49
0.74
0.52
0.58
0.52
0.42
0.34
Total
15,740
2.57
1.06
Panel C. Industry adjusted Tobin’s Q by -index and years from IPO
Years from IPO
N
Industry adjusted Q
Industry adjusted Q
0
1
2
3-4
5-6
7-9
≥10
2,285
2,098
1,829
2,919
2,234
2,271
2,104
Below median c-index
2.12
0.59
0.53
0.64
0.85
0.70
1.20
Above median c-index
2.47
0.96
0.63
0.55
0.44
0.28
0.12
Total
15,740
Panel D. Industry adjusted Tobin’s Q by Non-c-index and years from IPO
Years from IPO
N
Industry adjusted Q
Industry adjusted Q
46
t-stat
-1.97**
-3.18***
-1.04
1.03
4.37***
4.14***
6.38***
2.02**
t-stat
-3.04***
-0.28
1.36
0.67
3.99***
3.17***
5.76***
4.23***
t-stat
-1.59
-3.56***
-1.03
1.16
3.68***
4.72***
5.66***
t-stat
0
1
2
3-4
5-6
7-9
≥10
2,285
2,098
1,829
2,919
2,234
2,271
2,104
Below median Non-cindex
2.10
0.67
0.59
0.61
0.96
0.72
1.49
Above median Non-cindex
2.36
0.77
0.56
0.60
0.57
0.47
0.63
Total
15,740
1.02
0.82
Panel E. Industry adjusted Tobin’s Q by v-indexv-index and years from IPO
Years from IPO
N
Industry adjusted Q
Industry adjusted Q
0
1
2
3-4
5-6
7-9
≥10
2,285
2,098
1,829
2,919
2,234
2,271
2,104
Below median v-index
2.07
0.57
0.52
0.58
0.91
0.75
1.32
Above median v-index
2.51
0.93
0.63
0.62
0.44
0.32
0.24
Total
15,740
0.98
0.78
Panel F. Industry adjusted Tobin’s Q by Non-v-index and years from IPO
Years from IPO
N
Industry adjusted Q
Industry adjusted Q
0
1
2
3-4
5-6
7-9
≥10
2,285
2,098
1,829
2,919
2,234
2,271
2,104
Below median Non-vindex
2.33
0.75
0.60
0.54
0.63
0.64
0.96
Total
15,740
1.01
47
-1.23
0.91
0.30
0.21
3.21***
2.48**
3.70***
3.81***
t-stat
-2.12**
-3.52***
-1.11
-0.57
4.15***
4.87***
5.85***
3.97***
t-stat
Above median Non-vindex
2.33
0.72
0.56
0.62
0.70
0.52
0.28
-0.53
0.29
0.33
-0.89
-0.52
0.96
3.09***
0.85
2.60***
Table 7.
Multivariate regressions of industry adjusted Q on takeover defenses by years from IPO
Our sample consists of 2,285 IPO firms from 1997-2011. We eliminate all REITS, ADRs, funds, firms
without CRSP and COMPUSTAT coverage, firms incorporated outside the United States, firms with a
dual class share structure, and firms not in Jay Ritter’s database of firms with a founding date. In addition,
we eliminate all IPO firms that do not have prospectus filings, annual reports, and proxy statement filings
available in the SEC’s EDGAR database. The dependent variable is industry median adjusted IPO firm
Tobin’s Q, winsorized at the 99th percentile. c-index is the sum of the four takeover defenses that firms
cannot change without a shareholder vote: classified board, supermajority restrictions on amending
bylaws and charters, and supermajority requirements to approve mergers. Non-c-index is defined as the
sum of the indicators for poison pill and golden parachute. Standard errors are reported below the
regression coefficients.
Panel A. Regressions of industry adjusted Tobin’s Q in all years after IPO
(1)
Delaware (indicator)
0.457***
(0.061)
Firm age (years)
-0.002*
(0.001)
Log(assets)
-0.468***
(0.016)
S&P 500 firm (indicator)
2.008***
(0.178)
Insider ownership
1.008***
(0.127)
Missing insider ownership (indicator)
0.706***
(0.109)
Institutional ownership
0.473***
(0.081)
e-index
-0.022
(0.016)
Time from IPO >4 (indicator)
(2)
0.430***
(0.061)
-0.002
(0.001)
-0.474***
(0.016)
2.062***
(0.178)
0.876***
(0.129)
0.769***
(0.109)
0.559***
(0.083)
-0.018
(0.016)
-0.268***
(0.054)
e-index x Time from IPO>4
Constant
N
Adj R2
2.574***
(0.106)
15,740
7.716
2.717***
(0.110)
15,740
7.84
Panel B. Regressions of industry adjusted Tobin’s Q by years from IPO
(1)
(2)
(3)
Year = 0
Year = 1
Year=2
Delaware (indicator)
0.817***
0.161
0.306**
(0.257)
(0.130)
(0.122)
Firm age (years)
-0.023***
-0.004
-0.002
(0.005)
(0.003)
(0.002)
Log(assets)
-0.245***
-0.154***
-0.267***
(0.080)
(0.038)
(0.033)
S&P 500 firm (indicator)
7.117
0.346
1.449***
(4.861)
(0.849)
(0.551)
Insider ownership
2.473***
0.505**
-0.150
(0.455)
(0.250)
(0.247)
Missing insider ownership
0.006
0.468*
(indicator)
(0.339)
(0.284)
Institutional ownership
-2.383***
0.861***
1.180***
(0.651)
(0.204)
(0.165)
48
(4)
Year=3, 4
0.141
(0.094)
-0.001
(0.002)
-0.403***
(0.024)
1.648***
(0.310)
-0.389**
(0.197)
0.342*
(0.188)
1.013***
(0.121)
(5)
Year = 5-6
0.193
(0.132)
0.003
(0.002)
-0.682***
(0.033)
2.447***
(0.366)
-0.679**
(0.309)
0.831***
(0.209)
1.131***
(0.166)
(3)
0.428***
(0.061)
-0.002
(0.001)
-0.474***
(0.016)
2.127***
(0.178)
0.854***
(0.129)
0.747***
(0.109)
0.556***
(0.083)
0.076***
(0.020)
0.385***
(0.101)
-0.239***
(0.031)
2.489***
(0.114)
15,740
8.19
(6)
Year = 7-9
0.269***
(0.101)
0.001
(0.002)
-0.436***
(0.026)
1.737***
(0.224)
-0.033
(0.259)
0.712***
(0.146)
0.808***
(0.126)
(7)
Year≥10
0.160
(0.195)
0.016***
(0.004)
-1.064***
(0.050)
3.765***
(0.379)
-0.576
(0.578)
0.227
(0.266)
1.686***
(0.283)
e-index
Constant
N
Adj R2
0.162***
(0.066)
1.933***
(0.471)
2,285
5.03
0.075**
(0.032)
0.857***
(0.233)
2,098
1.81
0.012
(0.030)
1.341***
(0.210)
1,829
5.24
Panel C. Regressions of industry adjusted Tobin’s Q in all years after IPO
(1)
Delaware (indicator)
0.444***
(0.060)
Firm age (years)
-0.002*
(0.001)
Log(assets)
-0.469***
(0.016)
S&P 500 firm (indicator)
1.993***
(0.178)
Insider ownership
1.019***
(0.127)
Missing insider ownership (indicator)
0.704***
(0.109)
Institutional ownership
0.467***
(0.081)
Classified board (indicator)
-0.029
(0.051)
Time from IPO > 4 (indicator)
0.004
(0.023)
2.269***
(0.158)
2,919
10.21
-0.071**
(0.033)
3.815***
(0.222)
2,234
19.12
(2)
0.420***
(0.060)
-0.002
(0.001)
-0.475***
(0.016)
2.049***
(0.178)
0.879***
(0.129)
0.766***
(0.109)
0.556***
(0.083)
-0.040
(0.051)
-0.274***
(0.054)
N
Adj R2
2.551***
(0.106)
15,740
7.70
2.710***
(0.111)
15,740
7.84
Panel D. Regressions of industry adjusted Tobin’s Q by years from IPO
(1)
(2)
(3)
Year = 0
Year = 1
Year=2
Delaware (indicator)
0.874***
0.216*
0.332***
(0.254)
(0.128)
(0.120)
Firm age (years)
-0.022***
-0.004*
-0.003
(0.005)
(0.003)
(0.002)
Log(assets)
-0.244***
-0.145***
-0.261***
(0.080)
(0.038)
(0.033)
S&P 500 firm (indicator)
6.979
0.459
1.470***
(4.859)
(0.849)
(0.549)
Insider ownership
2.452***
0.499**
-0.182
(0.454)
(0.251)
(0.248)
Missing insider ownership
-0.009
0.450
(indicator)
(0.339)
(0.284)
Institutional ownership
-2.373***
0.899***
1.193***
(0.650)
(0.203)
(0.165)
Classified board (indicator)
0.635***
0.021
-0.171*
(0.216)
(0.108)
(0.102)
Constant
1.852***
0.937***
1.447***
(0.474)
(0.235)
(0.212)
N
2,285
2,098
1,829
Adj R2
5.14
1.56
5.38
(4)
Year=3, 4
0.146
(0.092)
-0.001
(0.002)
-0.402***
(0.024)
1.650***
(0.309)
-0.395**
(0.198)
0.342*
(0.188)
1.015***
(0.121)
-0.021
(0.077)
2.287***
(0.159)
2,919
10.21
Panel E. Regressions of industry adjusted Tobin’s Q by years from IPO
(1)
(2)
c-index
-0.013
-0.010
(0.195)
(0.195)
Time from IPO > 4 (indicator)
0.225
49
-0.370***
(0.059)
6.267***
(0.375)
2,104
22.34
(3)
0.413***
(0.060)
-0.001
(0.001)
-0.475***
(0.016)
2.032***
(0.178)
0.856***
(0.129)
0.728***
(0.109)
0.548***
(0.083)
0.242***
(0.067)
0.137*
(0.084)
-0.655***
(0.102)
2.542***
(0.113)
15,740
8.08
Classified board x Time from IPO > 4
Constant
-0.106***
(0.027)
2.482***
(0.179)
2,271
15.47
(5)
Year = 5-6
0.163
(0.130)
0.003
(0.002)
-0.684***
(0.032)
2.370***
(0.365)
-0.719**
(0.309)
0.811***
(0.209)
1.121***
(0.165)
-0.307***
(0.108)
3.867***
(0.223)
2,234
19.24
(6)
Year = 7-9
0.199**
(0.100)
0.001
(0.002)
-0.445***
(0.026)
1.667***
(0.225)
0.027
(0.260)
0.715***
(0.148)
0.800***
(0.126)
-0.139*
(0.084)
2.353***
(0.179)
2,271
14.99
(3)
0.538**
(0.252)
1.102**
(7)
Year≥10
0.009
(0.195)
0.016***
(0.004)
-1.063***
(0.050)
3.577***
(0.381)
-0.344
(0.580)
0.277
(0.267)
1.689***
(0.285)
-0.683***
(0.171)
5.651***
(0.354)
2,104
21.50
(0.337)
(0.456)
-1.337***
(0.390)
Yes
15,740
4.01
c-index x Time from IPO > 4
Control variables
N
Adj R2
Yes
15,740
3.95
Yes
15,740
3.94
Panel F. Regressions of industry adjusted Tobin’s Q by years from IPO
(1)
(2)
(3)
Year = 0
Year = 1
Year=2
c-index
0.444***
0.209***
0.206**
(0.172)
(0.078)
(0.100)
Control variables
Yes
Yes
Yes
N
2,285
2,098
1,829
Adj R2
3.37
1.01
7.30
(4)
Year=3, 4
0.044
(0.206)
Yes
2,919
5.24
(5)
Year = 5-6
-0.278
(1.042)
Yes
2,234
3.32
Panel G. Regressions of industry adjusted Tobin’s Q by years from IPO
(1)
(2)
Non-c-index
-0.112
-0.126
(0.174)
(0.175)
Time from IPO > 4 (indicator)
0.256
(0.380)
Non-c-index x Time from IPO > 4
Control variables
N
Adj R2
Yes
15,740
3.95
(4)
Year=3, 4
0.002
(0.183)
Yes
2,919
5.24
(5)
Year = 5-6
-0.720
(0.926)
Yes
2,234
3.34
Panel I. Regressions of industry adjusted Tobin’s Q by years from IPO
(1)
(2)
v-index
-0.008
-0.008
(0.017)
(0.017)
Time from IPO > 4 (indicator)
-0.272***
(0.054)
v-index x Time from IPO > 4
Control variables
N
Adj R2
Yes
15,740
7.70
50
(4)
Year=3, 4
0.022
(0.026)
Yes
2,919
(6)
Year = 7-9
0.139
(0.350)
Yes
2,271
5.63
(7)
Year≥10
-1.937***
(0.712)
Yes
2,104
13.28
(3)
0.106***
(0.022)
0.242***
(0.082)
-0.291***
(0.035)
Yes
15,740
8.25
Yes
15,740
7.84
Panel J. Regressions of industry adjusted Tobin’s Q by years from IPO
(1)
(2)
(3)
Year = 0
Year = 1
Year=2
v-index
0.203***
0.106***
0.039
(0.072)
(0.035)
(0.034)
Control variables
Yes
Yes
Yes
N
2,285
2,098
1,829
(7)
Year≥10
-0.993
(0.751)
Yes
2,104
13.05
(3)
0.314
(0.225)
-1.054***
(0.801)
-1.054***
(0.340)
Yes
15,740
4.00
Yes
15,740
3.94
Panel H. Regressions of industry adjusted Tobin’s Q by years from IPO
(1)
(2)
(3)
Year = 0
Year = 1
Year=2
Non-c-index
0.396***
0.092
-0.017
(0.159)
(0.071)
(0.090)
Control variables
Yes
Yes
Yes
N
2,285
2,098
1,829
Adj R2
3.35
0.75
7.09
(6)
Year = 7-9
-0.253
(0.402)
Yes
2,271
5.64
(5)
Year = 5-6
-0.096***
(0.037)
Yes
2,234
(6)
Year = 7-9
-0.104***
(0.030)
Yes
2,271
(7)
Year≥10
-0.354***
(0.064)
Yes
2,104
Adj R2
3.40
1.08
2.43
7.53
16.32
Panel K. Regressions of industry adjusted Tobin’s Q by years from IPO
(1)
(2)
Non-v-index
-0.124
-0.089**
(0.044)
(0.017)
Time from IPO > 4 (indicator)
-0.252***
(0.055)
Non-v-index x Time from IPO > 4
Control variables
N
Adj R2
Yes
15,740
7.74
51
(4)
Year=3, 4
0.046
(0.065)
Yes
2,919
7.52
20.49
(3)
-0.074
(0.058)
-0.220**
(0.101)
-0.034
(0.088)
Yes
15,740
7.86
Yes
15,740
7.86
Panel L. Regressions of industry adjusted Tobin’s Q by years from IPO
(1)
(2)
(3)
Year = 0
Year = 1
Year=2
Non-v-index
-0.121
-0.013
-0.007
(0.199)
(0.095)
(0.087)
Control variables
Yes
Yes
Yes
N
2,285
2,098
1,829
Adj R2
3.08
0.67
2.36
13.00
(5)
Year = 5-6
0.147
(0.090)
Yes
2,234
16.17
(6)
Year = 7-9
-0.135*
(0.071)
Yes
2,271
12.68
(7)
Year≥10
-0.596***
(0.184)
Yes
2,104
19.73
Table 8.
Multivariate regressions of industry adjusted Q on takeover defenses by years from IPO
Our sample consists of 2,285 IPO firms from 1997-2011. We eliminate all REITS, ADRs, funds, firms
without CRSP and COMPUSTAT coverage, firms incorporated outside the United States, firms with a
dual class share structure, and firms not in Jay Ritter’s database of firms with a founding date. In addition,
we eliminate all IPO firms that do not have prospectus filings, annual reports, and proxy statement filings
available in the SEC’s EDGAR database. The dependent variable is industry median adjusted IPO firm
Tobin’s Q, winsorized at the 99th percentile. In Panels C and D we use law firm dummy variables and an
indicator for firms located in California as instrumental variables.
Panel A. Regressions of industry adjusted Tobin’s Q by years from IPO
(1)
(2)
IPO stage e-index
-0.004
-0.007*
(0.016)
(0.016)
Time from IPO > 4 (indicator)
-0.273***
(0.054)
IPO stage e-index x Time from IPO > 4
Control variables
N
Adj R2
Yes
15,740
7.70
(3)
0.096***
(0.020)
0.334***
(0.092)
-0.250***
(0.031)
Yes
15,740
8.22
Yes
15,740
7.84
Panel B. Regressions of industry adjusted Tobin’s Q by years from IPO
(1)
(2)
(3)
(4)
Year = 0
Year = 1
Year=2
Year=3, 4
IPO stage e-index
0.126**
0.085***
0.016
0.016
(0.066)
(0.032)
(0.030)
(0.023)
Control variables
Yes
Yes
Yes
Yes
N
2,285
2,098
1,829
2,919
Adj R2
5.03
1.87
5.24
10.23
Panel C. 2SLS regressions of industry adjusted Tobin’s Q on instrumented e-index
Second Stage
Second Stage
Regression
Regression
(1)
(2)
Instrumented e-index
0.067
0.075*
(0.042)
(0.041)
Time from IPO > 4 (indicator)
-0.287***
(0.055)
Instrumented e-index x Time from IPO > 4
Control variables
Yes
Yes
N
15,740
15,740
R2
7.52
7.64
Panel D. 2SLS regressions of industry adjusted Tobin’s Q on instrumented e-index
(1)
(2)
(3)
(4)
Year = 0
Year = 1
Year=2
Year=3, 4
Instrumented e-index
0.294*
0.114
0.000
0.033
(0.169)
(0.081)
(0.075)
(0.058)
Control variables
Yes
Yes
Yes
Yes
N
2,285
2,098
1,829
2,919
2
Adj R
4.86
1.74
5.23
10.16
Panel E. First difference regressions of industry adjusted Tobin’s Q by years from IPO
(1)
(2)
Change in e-index
-0.093
-0.078
(0.072)
(0.072)
Time from IPO > 4 (indicator)
0.463***
52
(5)
Year = 5-6
-0.073**
(0.033)
Yes
2,234
19.13
(6)
Year = 7-9
-0.095***
(0.026)
Yes
2,271
15.39
(7)
Year≥10
-0.277***
(0.055)
Yes
2,104
21.84
Second Stage
Regression
(3)
0.261***
(0.072)
0.960**
(0.399)
-0.456***
(0.144)
Yes
15,740
7.66
(5)
Year = 5-6
-0.027
(0.082)
Yes
2,234
19.05
(6)
Year = 7-9
-0.110*
(0.026)
Yes
2,271
15.47
(3)
-0.141
(0.116)
0.459***
(7)
Year≥10
-0.120
(0.109)
Yes
2,104
21.68
(0.049)
(0.049)
0.104
(0.148)
Yes
13,455
10.52
Change in e-index x Time from IPO > 4
Control variables
Yes
Yes
N
13,455
13,455
Adj R2
9.93
10.52
Panel F. First difference regressions of industry adjusted Tobin’s Q by years from IPO
(1)
(2)
(3)
(4)
Year = 1
Year = 2
Year=3, 4
Year = 56
Change in e-index
-0.269
-0.042
-0.117
-0.033
(0.362)
(0.168)
(0.107)
(0.134)
Control variables
Yes
Yes
Yes
Yes
N
2,098
1,829
2,919
2,234
Adj R2
11.16
5.29
5.84
24.78
53
(5)
Year = 7-9
(6)
Year≥10
-0.100
(0.071)
Yes
2,271
17.46
0.060
(0.216)
Yes
2,104
7.73
Table 9.
Multivariate regressions of industry adjusted Q on takeover defenses by alternative measures of
firm and industry maturity
Our sample consists of 2,285 IPO firms from 1997-2011. We eliminate all REITS, ADRs, funds, firms
without CRSP and COMPUSTAT coverage, firms incorporated outside the United States, firms with a
dual class share structure, and firms not in Jay Ritter’s database of firms with a founding date. In addition,
we eliminate all IPO firms that do not have prospectus filings, annual reports, and proxy statement filings
available in the SEC’s EDGAR database. The dependent variable is industry median adjusted IPO firm
Tobin’s Q, winsorized at the 99th percentile. Sales growth rate is calculated from the current year sales
value divided by the following year’s sales value minus one. Standard errors are reported below the
regression coefficients.
Panel A. Regressions of industry adjusted Tobin’s Q by firm lifecycle measured by firm age in years
(1)
(2)
(3)
(4)
(5)
Age = 0-4
Age = 5-7
Age=8-11
Age=12-14
Age=15-18
e-index
0.185***
0.076*
-0.059*
-0.126***
-0.128***
(0.092)
(0.042)
(0.0023)
(0.042)
(0.050)
Control variables
Yes
Yes
Yes
Yes
Yes
N
1,419
2,105
3,110
2,020
2,001
Adj R2
5.02
5.00
10.59
16.76
14.63
(6)
Age=19-23
-0.063**
(0.029)
Yes
1,518
3.12
Panel B. Regressions of industry adjusted Tobin’s Q by firm lifecycle measured by mean industry age in years
(1)
(2)
(3)
(4)
(5)
(6)
Average
Average
Average
Average
Average
Average
industry
industry
industry
industry
industry
industry
age=0-7
age=8-9
age=10-11
age=12-13
age=14-15
age=15-17
e-index
0.173**
-0.027
-0.045
-0.024
-0.004
-0.060*
(0.075)
(0.035)
(0.042)
(0.035)
(0.036)
(0.033)
Control variables
Yes
Yes
Yes
Yes
Yes
Yes
N
2,036
2,384
2,415
2,727
1,985
2,031
Adj R2
4.94
6.71
8.79
16.23
11.65
11.87
(7)
Age≥24
-0.001
(0.023)
Yes
3,567
6.84
(7)
Average
industry
age>17
-0.053**
(0.024)
Yes
2,162
14.99
Panel C. Regressions of industry adjusted Tobin’s Q by firm lifecycle measured by firm sales growth rate quintiles
(1)
(2)
(3)
(4)
(5)
Firm sales growth
Firm sales growth
Firm sales growth
Firm sales growth
Firm sales growth
Quintile 5
Quintile 4
Quintile 3
Quintile 2
Quintile 1
High sales growth
Low sales growth
e-index
0.072*
-0.002
-0.010
-0.038*
-0.081***
(0.044)
(0.032)
(0.024)
(0.023)
(0.027)
Control variables
Yes
Yes
Yes
Yes
Yes
N
3,674
2,730
2,786
2,731
1,920
2
Adj R
6.67
9.17
7.06
8.21
10.94
Panel D. Regressions of industry adjusted Tobin’s Q by firm lifecycle measured by industry sales growth rate quintiles
(1)
(2)
(3)
(4)
(5)
Industry sales
Industry sales
Industry sales
Industry sales
Industry sales
growth Quintile 5
growth Quintile 4
growth Quintile 3
growth Quintile 2
growth Quintile 1
High sales growth
Low sales growth
e-index
0.093**
-0.015
-0.060*
-0.078**
-0.012
(0.047)
(0.033)
(0.034)
(0.028)
(0.028)
Control variables
Yes
Yes
Yes
Yes
Yes
N
3,185
3,261
3,018
3,128
3,148
2
Adj R
7.59
6.05
14.02
9.97
13.35
54
Table 10.
Univariate description of value-increasing relationships over time
Our sample consists of 2,285 IPO firms from 1997-2011. We eliminate all REITS, ADRs, funds, firms
without CRSP and COMPUSTAT coverage, firms incorporated outside the United States, firms with a
dual class share structure, and firms not in Jay Ritter’s database of firms with a founding date. In addition,
we eliminate all IPO firms that do not have prospectus filings, annual reports, and proxy statement filings
available in the SEC’s EDGAR database. The dependent variable is industry median adjusted IPO firm
Tobin’s Q, winsorized at the 99th percentile.
Panel A. Percent of firms with a large customer by e-index
Years from IPO
N
Percent with large
customer
Percent with large
customer
Above median e-index
Firms
50.99%
46.38%
44.00%
44.68%
36.26%
34.80%
11.04%
-4.85***
-2.87***
0.73
1.76*
-0.94
0.02
1.42
37.38%
1.33
Percent with large
customer
Percent with large
customer
t-stat
Classified Board
Firms
46.72%
43.36%
43.74%
43.86%
39.07%
35.61%
10.75%
-2.51**
-0.74
-0.78
-1.26
-2.41**
-1.04
-0.94
38.34%
5.10***
Percent with large
customer
Percent with large
customer
t-stat
Above median c-index
Firms
50.35%
44.60%
42.95%
43.36%
34.70%
34.12%
11.25%
-4.11***
-1.35
0.09
-0.33
2.02**
0.54
-1.20
37.72%
-1.81*
0
1
2
3-4
5-6
7-9
≥10
2,285
2,098
1,829
2,919
2,234
2,271
2,104
Below median e-index
Firms
40.72%
40.12%
42.31%
41.45%
38.17%
34.77%
9.14%
Total
15,740
36.35%
t-stat
Panel B. Percent of firms with a large customer by classified board
Years from IPO
N
0
1
2
3-4
5-6
7-9
≥10
2,285
2,098
1,829
2,919
2,234
2,271
2,104
No Classified Board
Firms
41.24%
41.66%
41.83%
41.45%
33.95%
33.48%
9.50%
Total
15,740
34.28%
Panel C. Percent of firms with a large customer by c-index
Years from IPO
N
0
1
2
3-4
5-6
7-9
≥10
2,285
2,098
1,829
2,919
2,234
2,271
2,104
Below median c-index
Firms
41.53%
41.61%
43.17%
42.75%
38.90%
35.22%
9.60%
Total
15,740
36.29%
Panel D. Percent of firms with a large customer by v-index
55
Years from IPO
N
Percent with large
customer
Percent with large
customer
Above median v-index
Firms
50.51%
45.41%
43.29%
43.63%
34.86%
33.77%
11.38%
-4.75***
-2.18**
0.17
.66
2.41**
1.04
-1.71*
37.44%
1.46
Percent with strategic
alliance
Percent with strategic
alliance
t-stat
Above median e-index
Firms
21.98%
24.10%
19.30%
17.39%
18.43%
17.24%
11.71%
-3.24***
-1.91*
-3.01***
-2.76***
-2.74***
-3.07***
-1.55
18.16%
-5.83***
Percent with strategic
alliance
Percent with strategic
alliance
t-stat
Classified Board
Firms
19.24%
23.38%
18.36%
16.35%
17.63%
15.54%
9.84%
-3.47***
-3.82***
-3.69***
-3.74***
-3.54***
-1.79*
0.58
17.27%
8.01***
Percent with strategic
alliance
Percent with strategic
alliance
t-stat
Above median c-index
Firms
20.82%
22.86%
18.33%
16.64%
18.01%
17.11%
12.28%
-1.97**
-0.69
-1.76*
-1.44
-1.77*
-2.06**
-1.69*
17.95%
4.23***
0
1
2
3-4
5-6
7-9
≥10
2,285
2,098
1,829
2,919
2,234
2,271
2,104
Below median v-index
Firms
40.57%
40.66%
42.90%
42.42%
39.44%
35.85%
9.12%
Total
15,740
36.32%
t-stat
Panel E. Percent of firms with a strategic alliance by e-index
Years from IPO
N
0
1
2
3-4
5-6
7-9
≥10
2,285
2,098
1,829
2,919
2,234
2,271
2,104
Below median e-index
Firms
16.58%
20.59%
14.07%
13.69%
14.16%
12.59%
9.58%
Total
15,740
14.73%
Panel F. Percent of firms with a strategic alliance by classified board
Years from IPO
N
0
1
2
3-4
5-6
7-9
≥10
2,285
2,098
1,829
2,919
2,234
2,271
2,104
No Classified Board
Firms
13.49%
16.28%
11.73%
11.27%
12.01%
12.83%
10.61%
Total
15,740
12.49%
Panel G. Percent of firms with a strategic alliance by c-index
Years from IPO
N
0
1
2
3-4
5-6
7-9
≥10
2,285
2,098
1,829
2,919
2,234
2,271
2,104
Below median c-index
Firms
17.49%
21.58%
15.21%
14.67%
15.20%
13.94%
9.90%
Total
15,740
15.40%
Panel H. Percent of firms with a strategic alliance by v-index
56
Years from IPO
N
Percent with strategic
alliance
Percent with strategic
alliance
Above median v-index
Firms
21.51%
23.68%
18.47%
16.86%
18.02%
16.41%
11.68%
-2.95***
-1.57
-2.17**
-2.00*
-2.11**
-1.64*
-1.27
17.90%
4.82
0
1
2
3-4
5-6
7-9
≥10
2,285
2,098
1,829
2,919
2,234
2,271
2,104
Below median v-index
Firms
16.65%
20.80%
14.71%
14.18%
14.72%
13.93%
9.95%
Total
15,740
15.05%
57
t-stat
Table 11.
Multivariate regressions of industry adjusted Q on takeover defenses and value-increasing
relationships
Our sample consists of 2,285 IPO firms from 1997-2011. We eliminate all REITS, ADRs, funds, firms
without CRSP and COMPUSTAT coverage, firms incorporated outside the United States, firms with a
dual class share structure, and firms not in Jay Ritter’s database of firms with a founding date. In addition,
we eliminate all IPO firms that do not have prospectus filings, annual reports, and proxy statement filings
available in the SEC’s EDGAR database. The dependent variable is industry median adjusted IPO firm
Tobin’s Q, winsorized at the 99th percentile. Standard errors are reported below the regression
coefficients.
Panel A. Regressions of industry adjusted Tobin’s Q in all years after IPO
(1)
Delaware (indicator)
0.459***
(0.061)
Firm age (years)
-0.002*
(0.001)
Log(assets)
-0.472***
(0.016)
S&P 500 firm (indicator)
2.005***
(0.178)
Insider ownership
1.012***
(0.127)
Missing insider ownership (indicator)
0.682***
(0.109)
Institutional ownership
0.470***
(0.081)
e-index
-0.068***
(0.020)
Large customer (indicator)
-0.453***
(0.097)
e-index x Large customer (indicator)
0.120***
(0.031)
Customer reliance gone (indicator)
e-index x Customer reliance gone (indicator
Constant
2.765***
(0.114)
N
15,740
Adj R2
7.82
Panel B. Regressions of industry adjusted Tobin’s Q in all years after IPO
(1)
Delaware (indicator)
0.442***
(0.060)
Firm age (years)
-0.002*
(0.001)
Log(assets)
-0.473***
(0.016)
S&P 500 firm (indicator)
1.994***
(0.178)
Insider ownership
1.029***
(0.126)
Missing insider ownership (indicator)
0.684***
(0.109)
Institutional ownership
0.462***
(0.081)
c-index
-0.072**
(0.035)
Large customer (indicator)
-0.281***
58
(2)
0.452***
(0.061)
-0.002*
(0.001)
-0.469***
(0.016)
2.005***
(0.178)
0.986***
(0.127)
0.694***
(0.109)
0.492***
(0.081)
0.021***
(0.019)
0.354***
(0.103)
-0.143***
(0.033)
2.478***
(0.112)
15,740
7.81
(2)
0.430***
(0.060)
-0.002*
(0.001)
-0.476***
(0.016)
2.048***
(0.178)
0.962***
(0.127)
0.713***
(0.109)
0.505***
(0.082)
0.067***
(0.033)
(3)
0.455***
(0.061)
-0.002*
(0.001)
-0.477***
(0.016)
2.019***
(0.178)
0.972***
(0.127)
0.673***
(0.109)
0.490***
(0.081)
-0.015
(0.027)
-0.384***
(0.114)
0.067*
(0.036)
0.143
(0.121)
-0.106***
(0.038)
2.731***
(0.131)
15,740
7.89
(3)
0.435***
(0.060)
-0.002*
(0.001)
-0.483***
(0.016)
2.055***
(0.178)
0.955***
(0.127)
0.686***
(0.109)
0.501***
(0.082)
-0.008
(0.042)
-0.307***
c-index x Large customer (indicator)
(0.064)
0.221***
(0.057)
Customer reliance gone (indicator)
c-index x Customer reliance gone (indicator
Constant
2.651***
(0.107)
N
15,740
Adj R2
7.81
Panel C. Regressions of industry adjusted Tobin’s Q in all years after IPO
(1)
Delaware (indicator)
0.445***
(0.061)
Firm age (years)
-0.002*
(0.001)
Log(assets)
-0.472***
(0.016)
S&P 500 firm (indicator)
1.993***
(0.178)
Insider ownership
1.027***
(0.126)
Missing insider ownership (indicator)
0.681***
(0.109)
Institutional ownership
0.467***
(0.081)
v-index
-0.068***
(0.022)
Large customer (indicator)
-0.427***
(0.081)
v-index x Large customer (indicator)
0.166***
(0.035)
Customer reliance gone (indicator)
v-index x Customer reliance gone (indicator
Constant
2.716***
(0.109)
N
15,740
Adj R2
7.86
Panel D. Regressions of industry adjusted Tobin’s Q in all years after IPO
(1)
Delaware (indicator)
0.414***
(0.060)
Firm age (years)
0.000
(0.001)
Log(assets)
-0.475***
(0.016)
S&P 500 firm (indicator)
1.690***
(0.178)
Insider ownership
0.980***
(0.126)
Missing insider ownership (indicator)
0.759***
(0.108)
Institutional ownership
0.444***
(0.080)
e-index
-0.040**
(0.017)
Strategic alliance (indicator)
0.775***
(0.131)
e-index x Strategic alliance (indicator)
0.085**
(0.040)
Strategic alliance gone (indicator)
e-index x Strategic alliance gone (indicator)
Constant
2.513***
59
-0.055
(0.070)
-0.220***
(0.062)
2.589***
(0.106)
15,740
7.83
(2)
0.439***
(0.061)
-0.002*
(0.001)
-0.469***
(0.016)
2.000***
(0.178)
1.000***
(0.127)
0.690***
(0.109)
0.487***
(0.081)
0.046**
(0.021)
0.282***
(0.083)
-0.181***
(0.037)
2.465***
(0.108)
15,740
7.83
(2)
0.434***
(0.061)
-0.002*
(0.001)
-0.482***
(0.016)
2.032***
(0.177)
0.936***
(0.127)
0.711***
(0.108)
0.551***
(0.081)
0.009
(0.019)
-0.231**
(0.103)
-0.086***
(0.032)
2.724***
(0.066)
0.176***
(0.059)
-0.133*
(0.072)
-0.174***
(0.065)
2.754***
(0.113)
15,740
7.95
(3)
0.441***
(0.061)
-0.002*
(0.001)
-0.477***
(0.016)
2.011***
(0.178)
0.986***
(0.127)
0.671***
(0.109)
0.486***
(0.081)
-0.008
(0.030)
-0.398***
(0.094)
0.106***
(0.041)
0.070
(0.097)
-0.126***
(0.043)
2.720***
(0.123)
15,740
7.93
(3)
0.406***
(0.060)
0.000
(0.001)
-0.482***
(0.016)
1.738***
(0.178)
0.947***
(0.126)
0.756***
(0.108)
0.487***
(0.081)
-0.015
(0.021)
0.752***
(0.136)
0.058
(0.042)
-0.078
(0.106)
-0.057*
(0.034)
2.576***
(0.107)
N
15,740
Adj R2
9.08
Panel E. Regressions of industry adjusted Tobin’s Q in all years after IPO
(1)
Delaware (indicator)
0.396***
(0.060)
Firm age (years)
0.000
(0.001)
Log(assets)
-0.478***
(0.016)
S&P 500 firm (indicator)
1.659***
(0.178)
Insider ownership
0.996***
(0.125)
Missing insider ownership (indicator)
0.762***
(0.108)
Institutional ownership
0.438***
(0.080)
c-index
-0.038
(0.031)
Strategic alliance (indicator)
0.840***
(0.084)
c-index x Strategic alliance (indicator)
0.242***
(0.073)
Strategic alliance gone (indicator)
c-index x Strategic alliance gone (indicator)
Constant
2.455***
(0.103)
N
15,740
Adj R2
9.10
Panel F. Regressions of industry adjusted Tobin’s Q in all years after IPO
(1)
Delaware (indicator)
0.406***
(0.060)
Firm age (years)
0.000
(0.001)
Log(assets)
-0.476***
(0.016)
S&P 500 firm (indicator)
1.663***
(0.178)
Insider ownership
0.992***
(0.125)
Missing insider ownership (indicator)
0.760***
(0.108)
Institutional ownership
0.437***
(0.080)
v-index
-0.043**
(0.019)
Strategic alliance (indicator)
0.692***
(0.109)
v-index x Strategic alliance (indicator)
0.170***
(0.046)
Strategic alliance gone (indicator)
v-index x Strategic alliance gone (indicator)
Constant
N
Adj R2
2.489***
(0.104)
15,740
9.12
60
(0.112)
15,740
8.20
(0.115)
15,740
9.19
(2)
0.436***
(0.060)
-0.002**
(0.001)
-0.484***
(0.016)
2.126***
(0.178)
0.930***
(0.127)
0.713***
(0.108)
0.585***
(0.082)
0.054*
(0.033)
(3)
0.392***
(0.060)
-0.001
(0.001)
-0.491***
(0.016)
1.790***
(0.179)
0.906***
(0.126)
0.767***
(0.108)
0.553***
(0.082)
0.007
(0.035)
0.841***
(0.084)
0.233***
(0.073)
-0.290***
(0.069)
-0.160***
(0.061)
2.592***
(0.105)
15,740
9.42
-0.294***
(0.070)
-0.168***
(0.062)
2.674***
(0.106)
15,740
8.03
(2)
0.426***
(0.060)
-0.002*
(0.001)
-0.483***
(0.016)
2.019***
(0.177)
0.947***
(0.126)
0.709***
(0.108)
0.548***
(0.081)
0.031
(0.021)
-0.261***
(0.083)
-0.116***
(0.036)
2.705***
(0.109)
15,740
8.21
(3)
0.399***
(0.060)
0.000
(0.001)
-0.482***
(0.016)
1.708***
(0.178)
0.957***
(0.126)
0.755***
(0.108)
0.481***
(0.081)
-0.015***
(0.024)
0.646
(0.113)
0.141***
(0.048)
-0.118***
(0.086)
-0.066*
(0.038)
2.569***
(0.110)
15,740
9.23
Table 12.
Univariate description of insider ownership over time
Our sample consists of 2,285 IPO firms from 1997-2011. We eliminate all REITS, ADRs, funds, firms
without CRSP and COMPUSTAT coverage, firms incorporated outside the United States, firms with a
dual class share structure, and firms not in Jay Ritter’s database of firms with a founding date. In addition,
we eliminate all IPO firms that do not have prospectus filings, annual reports, and proxy statement filings
available in the SEC’s EDGAR database. Insider holdings are defined as the total holdings of officers and
directors as reported in the firm Def 14A filing.
Panel A. Percent insider holdings by e-index
Years from IPO
N
Percent insider holdings
Percent insider holdings
t-stat
Above median e-index
Firms
41.39%
32.97%
26.39%
21.54%
17.36%
14.12%
11.33%
0.60
1.48
3.23***
5.78***
5.14***
6.35***
5.85***
0.3078
14.15***
Percent insider holdings
Percent insider holdings
t-stat
Classified Board
Firms
41.73%
33.11%
26.81%
22.24%
17.63%
14.96%
12.05%
0.04
1.98**
3.68***
5.58***
5.77***
4.12***
3.63***
0
1
2
3-4
5-6
7-9
≥10
2,285
2,098
1,829
2,919
2,234
2,271
2,104
Below median e-index
Firms
41.97%
34.38%
29.55%
25.88%
21.46%
18.86%
15.69%
Total
15,740
0.2097
Panel B. Percent insider holdings by classified board
Years from IPO
N
0
1
2
3-4
5-6
7-9
≥10
2,285
2,098
1,829
2,919
2,234
2,271
2,104
No Classified Board
Firms
41.77%
35.07%
30.57%
26.61%
22.39%
18.17%
14.77%
Total
15,740
26.47%
24.30%
5.94***
Percent insider holdings
Percent insider holdings
t-stat
Above median c-index
Firms
41.22%
33.32%
27.18%
22.46%
18.33%
14.82%
11.48%
0.85
0.76
1.55
2.91***
2.14**
2.95***
3.59***
Panel C. Percent insider holdings by c-index
Years from IPO
N
0
1
2
3-4
5-6
7-9
≥10
2,285
2,098
1,829
2,919
2,234
2,271
2,104
Below median c-index
Firms
42.05%
34.07%
28.72%
24.70%
20.07%
17.11%
14.23%
Total
15,740
25.84%
23.92%
5.31***
Percent insider holdings
Percent insider holdings
t-stat
Below median v-index
Above median v-index
Panel D. Percent insider holdings by v-index
Years from IPO
N
61
0
1
2
3-4
5-6
7-9
≥10
2,285
2,098
1,829
2,919
2,234
2,271
2,104
Firms
41.98%
34.15%
29.02%
25.18%
20.54%
17.63%
14.55%
Firms
41.42%
33.32%
27.06%
22.29%
18.15%
14.83%
11.79%
0.59
0.87
2.01**
3.84***
2.98***
3.69***
3.72***
Total
15,740
26.50%
23.51%
8.50***
62
Table 13.
Multivariate regressions of industry adjusted Q on takeover defenses and insider ownership
Our sample consists of 2,285 IPO firms from 1997-2011. We eliminate all REITS, ADRs, funds, firms
without CRSP and COMPUSTAT coverage, firms incorporated outside the United States, firms with a
dual class share structure, and firms not in Jay Ritter’s database of firms with a founding date. In addition,
we eliminate all IPO firms that do not have prospectus filings, annual reports, and proxy statement filings
available in the SEC’s EDGAR database. Insider holdings are defined as the total holdings of officers and
directors as reported in the firm Def 14A filing. The dependent variable is industry median adjusted IPO
firm Tobin’s Q, winsorized at the 99th percentile. Standard errors are reported below the regression
coefficients.
Panel A. Regressions of industry adjusted Tobin’s Q in all years after IPO
(1)
Delaware (indicator)
0.445***
(0.061)
Firm age (years)
-0.002**
(0.001)
Log(assets)
-0.471***
(0.016)
S&P 500 firm (indicator)
2.055***
(0.178)
Insider ownership
1.557***
(0.198)
Missing insider ownership (indicator)
0.676***
(0.109)
Institutional ownership
0.458***
(0.081)
e-index
0.026
(0.022)
Below median ownership (indicator)
0.557***
(0.121)
e-index x Below median ownership (indicator)
-0.095***
(0.031)
Above median decline in ownership (indicator)
e-index x Above median decline in ownership
(indicator)
Constant
2.202***
(0.135)
N
15,740
Adj R2
7.82
Panel B. Regressions of industry adjusted Tobin’s Q in all years after IPO
(1)
Delaware (indicator)
0.437***
(0.060)
Firm age (years)
-0.002*
(0.001)
Log(assets)
-0.473***
(0.016)
S&P 500 firm (indicator)
2.028***
(0.178)
Insider ownership
1.533***
(0.197)
Missing insider ownership (indicator)
0.679***
(0.109)
Institutional ownership
0.447***
(0.081)
c-index
0.072***
(0.040)
Below median ownership (indicator)
0.372***
63
(2)
0.425***
(0.060)
-0.003**
(0.001)
-0.504***
(0.016)
1.943***
(0.176)
0.579***
(0.128)
0.826***
(0.108)
0.812***
(0.083)
0.120***
(0.030)
-0.547***
(0.106)
-0.179***
(0.035)
3.198***
(0.135)
15,740
9.38
(2)
0.413***
(0.060)
-0.003**
(0.001)
-0.505***
(0.016)
1.945***
(0.176)
0.602***
(0.128)
0.828***
(0.108)
0.799***
(0.083)
0.208***
(0.056)
(3)
0.418***
(0.060)
-0.003**
(0.001)
-0.506***
(0.016)
1.973***
(0.177)
1.056***
(0.198)
0.801***
(0.108)
0.794***
(0.083)
0.139***
(0.032)
0.411***
(0.121)
-0.055*
(0.031)
-0.570***
(0.108)
-0.167***
(0.036)
2.923***
(0.156)
15,740
9.44
(3)
0.411***
(0.060)
-0.003**
(0.001)
-0.507***
(0.016)
1.966***
(0.177)
1.038***
(0.198)
0.805***
(0.108)
0.779***
(0.083)
0.228***
(0.059)
0.293***
c-index x Below median ownership (indicator)
(0.093)
-0.125**
(0.056)
Above median decline in ownership (indicator)
c-index x Above median decline in ownership
(indicator)
Constant
2.248***
(0.127)
N
15,740
Adj R2
7.78
Panel B. Regressions of industry adjusted Tobin’s Q in all years after IPO
(1)
Delaware (indicator)
0.436***
(0.061)
Firm age (years)
-0.002*
(0.001)
Log(assets)
-0.472***
(0.016)
S&P 500 firm (indicator)
2.049***
(0.178)
Insider ownership
1.561***
(0.198)
Missing insider ownership (indicator)
0.670***
(0.109)
Institutional ownership
0.452***
(0.081)
v-index
0.054**
(0.024)
Below median ownership (indicator)
0.519***
(0.106)
v-index x Below median ownership (indicator)
-0.124***
(0.034)
Above median decline in ownership (indicator)
v-index x Above median decline in ownership
(indicator)
Constant
N
Adj R2
2.187***
(0.130)
15,740
7.83
64
-0.823***
(0.074)
-0.267***
(0.064)
3.363***
(0.121)
15,740
9.32
(2)
0.419***
(0.060)
-0.003**
(0.001)
-0.504***
(0.016)
1.945***
(0.176)
0.587***
(0.128)
0.822***
(0.108)
0.807***
(0.083)
0.165***
(0.034)
-0.608***
(0.090)
-0.229***
(0.039)
3.215***
(0.127)
15,740
9.43
(0.093)
-0.072
(0.057)
-0.832***
(0.075)
-0.245***
(0.066)
3.136***
(0.142)
15,740
9.37
(3)
0.413***
(0.060)
-0.003***
(0.001)
-0.507***
(0.016)
1.975***
(0.176)
1.049***
(0.198)
0.795***
(0.108)
0.788***
(0.083)
0.188***
(0.035)
0.380***
(0.106)
-0.070**
(0.035)
-0.629***
(0.091)
-0.211***
(0.040)
2.955***
(0.149)
15,740
9.49
Appendix
This appendix contains additional robustness tests of our overall results. We first examine the
impact of sample selection on our results. First we look to see if the takeover trends in adoption might be
caused by survivorship bias in our sample. We find some evidence of survivorship bias, but the impact is
quite small. We then confirm that our overall results are not significantly impacted by sample selection
bias. Next we examine the long-term stock performance of portfolios of IPO firms based on if this have
high or low levels of takeover defense adoption. We then repeat analyses similar to the prior literature to
show that our results are qualitatively similar to Gompers, Ishii, and Metrick (2003). We then repeat our
analyses using the firm valuation measures proposed by Dybvig and Warachka (2015). Finally, we
examine impact adopting certain important takeover defenses has on the modification of other takeover
defenses.
In Appendix Table I we show that there is some sample selection in the firms that disappear from
the total sample of firms. The table reports in panel A the takeover defenses of 1,169 IPO firms that
survive at least six years where panel B reports the takeover defenses of the 1,116 IPO firms that
disappear from the sample in less than six years. Panel C reports the statistical significance between the
takeover defenses of these two samples of firms. The table shows that firms which disappear from the
sample are significantly less likely to have poison pills, a supermajority to amend bylaws, amend the
charter, and approve mergers.
Based on the tabulated results, firms without the takeover provisions of poison pill, supermajority
to amend bylaws, supermajority to amend charters, or supermajority to approve mergers are more likely
to disappear from the sample. This implies that at least part of the trend in the takeover defenses reported
in Table 4 may be explained by sample selection. Specifically, firms without poison pills are
disproportionately likely to leave the sample so the trend in poison pills as the firm matures may be
explained in part by this sample selection. Likewise, part of the modest increase in the use of a
supermajority to approve a merger as the firm matures may be explained by the survivorship bias since
firms with this takeover defense are less likely to leave the sample. However, the strong trend in
removing classified boards and adding golden parachutes is clearly not caused by a survivorship bias and
the strongly positive trend in poison pill adoption cannot explain all the trend since even firms which
survive in the full sample display a positive trend in poison pill adoptions.
In Appendix Table 2, we examine the lifecycle effect regressions for firms which survive at least
five (Panels A and B) and ten (Panels C and D) years. The results show that for the subset of firms that
65
survive at least five years, there is still a strong and statistically significant value reversal. In the early
years after the IPO event, there is a strong and positive relationship between the takeover defenses and the
firm value. However, as the firm matures, this relationship becomes negative, and eventually statistically
significant. These results hold where takeover defenses are measured using e-index or classified board.
For firms surviving at least ten years we find qualitatively similar results – there is a definitive
lifecycle effect with a strong value reversal. These results imply that our major findings in the paper are
unlikely to be driven by survivorship bias in our sample.
In Appendix Table 3 we examine the long-term stock performance of portfolios formed based on
investing in IPO firms with high versus low levels of takeover defense adoption. We sort our IPO firms
into portfolios based on the IPO takeover defense adoption, ignoring any changes in the level of takeover
defenses after the IPO. We then form portfolios by purchasing all IPOs that have gone public in the first
full month they appear in the CRSP database. In Table 3 we examine the portfolios of stock returns. We
hold the IPO firms in their portfolios for various time periods after the IPO event. For instance, the first
column of portfolios are formed by purchasing the stocks at the time of the IPO and selling them one year
later. The second column of portfolios are formed by purchasing the stocks at the one year anniversary of
the IPO and selling them one year later. The third column of portfolios are formed by purchasing the
stocks at the two-year anniversary of the IPO and selling them one year later. Columns 4-7 each use a
three year window of stock holdings, purchasing stocks in year 1, 4, 7, and 10 and holding the stocks for
three full years or until the stocks are delisted.
Each of these portfolios is then regressed onto the Fama and French (1993) three factor model
plus the Carhardt (1997) momentum factor. The alpha for each of these portfolios is reported in Table 3.A
for value-weighted portfolios and in Table 3.B for equally-weighted portfolios. The results show that the
first year after the IPO, firms with a high level of takeover defenses tend to outperform firms with a low
level of takeover defenses, as shown by the long-short portfolio having a positive and significant alpha.
However, the majority of the alphas reported in the table show that there are very few significant alphas
for the IPO firm portfolios. This implies that the takeover defenses adopted by the IPO firms are correctly
priced in the market after the first year the firm is public. In addition, we show in Figure 2 the three-year
holding performance for portfolios of IPO firms formed based on their e-index. This figure shows that
IPO firms with higher levels of takeover defenses tend to have a higher long-term performance, although
the results are not statistically significant.
66
Next, we compare our overall results to those of the prior literature. Although our sample firms
are different from Bebchuck, Cohen, and Ferrell (2009) and Gompers, Ishii, and Metrick (2003), we
attempt to duplicate their results as closely as possible. We therefore run year-by-year regressions of
industry adjusted Tobin’s Q onto control variables plus the e-index measure or classified board measure.
We begin our sample in 2002 since this is the first year that our sample contains firms that are at least five
years old. Running regressions prior to 2002 would be akin to running firms in their first four years of
being public, for instance. We find in Appendix Table 4 that the coefficient on e-index is generally
negative and often significant, consistent with Bebchuk, Cohen, and Ferrell (2009). Likewise, when we
use the classified board indicator as our measure of takeover defenses, we find that the coefficient on this
variable is largely negative and often statistically significant. These results correspond well to the prior
research examining the relationship between firm Tobin’s Q and takeover defenses. An important
observation from this table is that as the sample size grows and the firms mature, the coefficient generally
becomes more negative and more significant, moving from 2002 to 2013.
To ensure that our results are robust to alternative ways of measuring firm value or efficiency, we
follow Dybvig and Warachka (2015) in calculating two measures of firm operating efficiency, Ry (scale
efficiency) and Rc (cost discipline). Ry (scale efficiency) is measured as (sales – cost of goods sold) / total
capital and Rc (cost discipline) is measured as (selling, general, and administrative – R&D – advertising )
/ total capital. Following Dybvig and Warachka (2015) we use Property, plant, and equipment as our
proxy for firm capital. We replace missing values of R&D and advertising with a zero.
Consistent with our prior findings regarding Q, we conduct multivariate regressions where Ry
and Rc are the dependent variables. We utilize control variables similar to the rest of our empirical tests in
the paper including a Delaware indicator variable, firm age, log (assets), and an indicator variable which
takes a value of one if the firm is in the S&P 500. Our variables of interest are the e-index, an indicator
for firms that have been public more than four years (the median value in our sample), and the interaction
term of these two variables. Dybvig and Warachka (2015) point out that “positive…coefficients would
support the hypothesis that a higher [e-index] (more entrenchment) corresponds to worse operating
efficiency.” Thus, for the tests with the Dybvig and Warachka (2015) measures to be consistent with our
prior results, we would need to see a positive coefficient for the interaction term between the e-index
measure and the indicator for firms having been public longer than four years.
We find in Appendix Table 5.A that the scale efficiency (Ry) of our firms are largely uncorrelated
with the control variables or the governance variables used in our regressions. This is inconsistent with
governance having an influence on the scale efficiency of the firms in our sample. However, in Appendix
67
Table 5.B we find that the cost efficiency is positively related to the e-index measure (model 1), the
indicator for firms which went public more than four years prior (model 2), and the interaction between
these two variables. This positive and significant relationship on the interaction term in model 3 in
particular indicates that as the IPO firm matures, its cost efficiency declines, particularly if the firm has a
high e-index. Thus, it appears that the results using the proxies for firm efficiency from Dybvig and
Warachka (2015) are consistent with our prior Tobin’s Q results.
One little-examined question of importance for firms modifying their takeover defenses is to
determine which takeover defenses might be important in allowing or preventing substantial changes in
takeover defenses. We suggest that supermajority requirements to amend bylaws and charters in particular
make it very difficult for the firm to amend their other takeover defenses since they tend to take a
supermajority of shareholders to do so. We investigate this possibility by examining the takeover defenses
in the IPO stage and then observing the post-IPO changes in takeover defenses for these firms. Our results
are tabulated in Appendix Table 6.
We examine takeover defense adoptions and removals separately since removals of takeover
defenses may prove optimal and additions may prove sub-optimal. Our tabulated results in Appendix
Table 6 show lower takeover defense removal rates for firms with bylaw restrictions and charter
restrictions, although these are only significantly lower in the later years of the IPO firm’s life. Likewise,
we find marginally higher takeover defense adoption rates for firms with bylaw and charter restrictions,
but these results are not consistently significant. On the whole, we find some support for the idea that
adopting certain takeover defenses reduces the revision of other takeover defenses.
68
Appendix Table 1.
Examination of survivorship bias
Our sample consists of 2,285 IPO firms from 1997-2011. We eliminate all REITS, ADRs, funds, firms
without CRSP and COMPUSTAT coverage, firms incorporated outside the United States, firms with a
dual class share structure, and firms not in Jay Ritter’s database of firms with a founding date. In addition,
we eliminate all IPO firms that do not have prospectus filings, annual reports, and proxy statement filings
available in the SEC’s EDGAR database.
Panel A. Individual takeover defenses and e-index for IPO firms that survive past year 5
Years from
N
Classified
Poison
Supermajority
Supermajority
IPO
board
Pill
to amend bylaw
to amend
charter
0
1,169
64.50%
6.24%
37.64%
33.62%
1
1,169
64.59%
10.44%
37.47%
33.45%
2
1,169
64.59%
14.20%
37.47%
33.36%
3
1,169
64.16%
16.68%
37.30%
33.19%
4
1,169
63.73%
18.91%
37.30%
33.19%
5
1,169
63.90%
21.30%
37.13%
33.02%
Supermajority
to approve
mergers
45.68%
46.02%
46.79%
48.16%
48.67%
48.85%
Total
7,014
64.24%
14.63%
37.38%
33.30%
47.36%
Panel B. Individual takeover defenses and e-index for IPO firms that leave the sample by year 5
Years from
N
Classified
Poison
Supermajority
Supermajority
Supermajority
IPO
board
Pill
to amend bylaw
to amend
to approve
charter
mergers
0
1,116
66.13%
4.39%
29.84%
26.34%
35.30%
1
929
66.52%
7.43%
31.75%
28.53%
37.46%
2
660
67.12%
9.55%
34.85%
31.36%
40.45%
3
399
65.66%
13.78%
31.83%
28.57%
39.85%
4
182
63.19%
15.93%
29.67%
25.82%
36.81%
5
10
30.00%
10.00%
0.00%
0.00%
0.00%
Golden
Parachute
e-index
64.07%
65.70%
67.75%
70.92%
72.37%
73.65%
2.52
2.58
2.64
2.70
2.74
2.78
69.08%
2.66
Golden
Parachute
e-index
64.78%
65.45%
68.03%
68.42%
67.58%
80.00%
2.27
2.37
2.51
2.48
2.39
1.20
Total
3,296
66.11%
8.07%
31.52%
28.13%
37.47%
66.26%
2.38
Panel C. Test of difference in Individual takeover defenses and e-index for IPO firms that leave versus remain in the sample by year 5
Years from
Classified
Poison
Supermajority
Supermajority
Supermajority
Golden
e-index
IPO
board
Pill
to amend bylaw
to amend
to approve
Parachute
charter
mergers
0 t-stat
-0.87
1.97*
3.95***
3.80***
5.07***
0.36
3.73***
1 t-stat
-0.92
2.38**
2.73***
2.42**
3.96***
0.12
2.86***
2 t-stat
-1.09
2.89**
1.11
0.88
2.62***
-0.12
1.58
3 t-stat
-0.54
1.36
1.97**
1.71*
2.88***
0.94
2.33**
4 t-stat
0.14
0.96
1.99**
1.98**
2.99***
1.33
2.68***
5 t-stat
2.22**
0.87
2.43**
2.21**
3.08***
0.45
3.04***
69
Appendix Table 2.
Multivariate regressions of industry adjusted Q on takeover defenses by years from IPO
considering survivorship bias
Our sample consists of 2,285 IPO firms from 1997-2011. We first eliminate all firms that do not have at
least ten years of data to report regressions, reducing the number of firm observations to 612. We
eliminate all REITS, ADRs, funds, firms without CRSP and COMPUSTAT coverage, firms incorporated
outside the United States, firms with a dual class share structure, and firms not in Jay Ritter’s database of
firms with a founding date. In addition, we eliminate all IPO firms that do not have prospectus filings,
annual reports, and proxy statement filings available in the SEC’s EDGAR database. The dependent
variable is industry median adjusted IPO firm Tobin’s Q, winsorized at the 99th percentile. Standard errors
are reported below the regression coefficients.
Panel A. Regressions of industry adjusted Tobin’s Q by years from IPO for firms surviving five years or more
(1)
(2)
(3)
(4)
(5)
(6)
Year = 0
Year = 1
Year=2
Year=3, 4 Year = 5-6
Year = 7-9
e-index
0.261***
0.093**
0.011
-0.004
-0.071**
-0.106***
(0.093)
(0.047)
(0.040)
(0.026)
(0.033)
(0.027)
Control variables
Yes
Yes
Yes
Yes
Yes
Yes
N
1,194
1,192
1,186
2,373
2,234
2,271
Adj R2
5.87
2.71
6.83
12.06
19.12
15.47
(7)
Year≥10
-0.370***
(0.059)
Yes
2,104
22.34
Panel B. Regressions of industry adjusted Tobin’s Q by years from IPO for firms surviving five years or more
(1)
(2)
(3)
(4)
(5)
(6)
Year = 0
Year = 1
Year=2
Year=3, 4 Year = 5-6
Year = 7-9
Classified board
0.716*
0.094
-0.176
0.007
-0.307***
-0.139*
(0.312)
(0.159)
(0.133)
(0.086)
(0.108)
(0.084)
Control variables
Yes
Yes
Yes
Yes
Yes
Yes
N
1,194
1,192
1,186
2,373
2,234
2,271
Adj R2
5.67
2.42
6.96
12.06
19.24
14.99
(7)
Year≥10
-0.683***
(0.171)
Yes
2,104
22.50
Panel C. Regressions of industry adjusted Tobin’s Q by years from IPO for firms surviving ten years or more
(1)
(2)
(3)
(4)
(5)
(6)
Year = 0
Year = 1
Year=2
Year=3, 4 Year = 5-6
Year = 7-9
e-index
0.384***
0.154**
0.014
0.027
-0.067***
-0.108***
(0.143)
(0.074)
(0.061)
(0.037)
(0.041)
(0.031)
Control variables
Yes
Yes
Yes
Yes
Yes
Yes
N
612
612
610
1,218
1,217
1,831
Adj R2
4.51
1.36
1.50
6.02
13.14
13.99
(7)
Year≥10
-0.360***
(0.059)
Yes
2,092
20.51
Panel D. Regressions of industry adjusted Tobin’s Q by years from IPO for firms surviving ten years or more
(1)
(2)
(3)
(4)
(5)
(6)
Year = 0
Year = 1
Year=2
Year=3, 4 Year = 5-6
Year = 7-9
Classified board
0.697
0.347
-0.384*
-0.081
-0.423***
-0.228**
(0.473)
(0.245)
(0.202)
(0.122)
(0.135)
(0.095)
Control variables
Yes
Yes
Yes
Yes
Yes
Yes
N
612
612
610
1,218
1,217
1,831
Adj R2
3.73
0.98
2.08
6.01
13.64
13.67
(7)
Year≥10
-0.705***
(0.170)
Yes
2,092
19.75
70
Appendix Table 3.
Alpha of Portfolios of stocks by e-index
Our sample consists of 2,285 IPO firms from 1997-2011. We eliminate all REITS, ADRs, funds, firms
without CRSP and COMPUSTAT coverage, firms incorporated outside the United States, firms with a
dual class share structure, and firms not in Jay Ritter’s database of firms with a founding date. In addition,
we eliminate all IPO firms that do not have prospectus filings, annual reports, and proxy statement filings
available in the SEC’s EDGAR database. We then generate a portfolio of stocks by purchasing each IPO
firm the first full month after the firm goes public and holding this stock as part of an equally-weighted
portfolio of firms with the same number of takeover defenses for various windows. The portfolio is
rebalanced each month and firms that are older than the window time frame are removed from the
portfolio while new IPO stocks are added. We then regress the monthly returns of these portfolios onto
the three Fama and French (1993) factors of mkt – rf, smb, and hml as well as the Carhardt (1997)
momentum factor. All factors are obtained from Ken French’s web site. t-statistics are reported below the
coefficients in parentheses.
Rt - Rf= at + b1 (rm – rf) + b2 (smb) + b3 (hml) + b4 (mom) + et
Panel A. Regression alphas for value weighted portfolios formed based on e-index for various holding windows
e-index
Year
Year
Year
Year
Year
Year
1
2
3
1-3
4-6
7-9
0 a:
-0.14
-0.34
0.58
0.38
1.08
0.38
(0.16)
(0.51)
(0.69)
(0.57)
(1.34)
(0.96)
1
-0.30
-0.16
0.30
0.23
0.46
0.01
(0.49)
(0.40)
(0.64)
(0.62)
(1.59)
(0.02)
2
1.29**
0.13
0.47
0.58
0.33
0.23
(2.23)
(0.30)
(0.78)
(1.20)
(1.11)
(0.80)
3
1.17*
-0.48
1.31*
1.00*
0.57*
0.24
(1.80)
(1.07)
(1.70)
(1.89)
(1.72)
(0.77)
4-6 b:
1.61**
-0.68
0.50
0.81*
0.63**
0.18
(2.44)
(1.56)
(1.05)
(1.94)
(2.13)
(0.49)
Long – short (b-a)
-0.16
(0.31)
-0.33
(0.53)
Panel B. Regression alphas for equally-weighted portfolios formed based on e-index for various holding windows
e-index
Year
Year
Year
Year
Year
Year
1
2
3
1-3
4-6
7-9
0 a:
-1.56**
-0.09
0.68
-0.37
0.48
0.36
(2.11)
(0.12)
(0.84)
(0.63)
(0.44)
(0.71)
1
-1.11**
0.15
0.88*
-0.02
0.77**
0.43
(2.17)
(0.32)
(1.77)
(0.07)
(2.06)
(0.78)
2
-0.32
-0.35
0.13
0.07
0.29
0.54
(0.73)
(0.73)
(0.26)
(0.19)
(0.59)
(1.52)
3
-0.17
-0.27
0.14
-0.09
0.80
-0.24
(0.35)
(0.45)
(0.28)
(0.27)
(1.34)
(0.41)
4-6 b:
0.01
-1.07**
0.78*
0.20
0.64
-0.09
(0.01)
(2.02)
(1.93)
(0.73)
(1.44)
(0.19)
Year
10-12
-0.55
(1.44)
0.17
(0.50)
0.37
(1.33)
0.11
(0.36)
-0.47
(-1.56)
Long – short (b-a)
2.05**
(2.17)
1.65**
(2.02)
-0.50
(0.66)
-1.08
(1.21)
0.50
(0.56)
0.45
(0.52)
71
0.76
(1.10)
0.67
(1.16)
-0.41
(0.51)
Year
10-12
-0.36
(0.63)
0.35
(0.71)
0.80
(1.57)
0.27
(0.46)
-0.71*
(1.93)
0.20
(0.17)
-0.40
(0.59)
0.08
(0.20)
Appendix Table 4.
Regression of industry adjusted Q on e-index and classified board by year
Our sample consists of 2,285 IPO firms from 1997-2011. We eliminate all REITS, ADRs, funds, firms
without CRSP and COMPUSTAT coverage, firms incorporated outside the United States, firms with a
dual class share structure, and firms not in Jay Ritter’s database of firms with a founding date. In addition,
we eliminate all IPO firms that do not have prospectus filings, annual reports, and proxy statement filings
available in the SEC’s EDGAR database. The dependent variable is industry median adjusted IPO firm
Tobin’s Q, winsorized at the 99th percentile. Control variables in all regressions include a Delaware
incorporation indicator variable, firm age (years), log(assets), and an indicator if the firm is in the S&P
500 index. Standard errors are reported below the regression coefficients.
Regression of Tobin’s Q onto e-index or classified board
Year
e-index
2002
-0.006
(0.043)
2003
-0.082*
(0.047)
2004
-0.069
(0.046)
2005
-0.039
(0.038)
2006
-0.015
(0.038)
2007
0.010
(0.045)
2008
-0.103**
(0.046)
2009
-0.180***
(0.068)
2010
-0.135**
(0.061)
2011
-0.132*
(0.069)
2012
-0.158*
(0.086)
2013
-0.005
(0.055)
72
Classified board
-0.270*
(0.145)
-0.032
(0.159)
-0.280*
(0.156)
-0.227*
(0.129)
-0.213*
(0.129)
-0.132
(0.137)
-0.495***
(0.141)
-0.445**
(0.208)
0.064
(0.187)
-0.153
(0.210)
0.301
(0.265)
0.072
(0.171)
Appendix Table 5.
Regression of Firm operating efficiency on e-index
Our sample consists of 2,285 IPO firms from 1997-2011. We eliminate all REITS, ADRs, funds, firms
without CRSP and COMPUSTAT coverage, firms incorporated outside the United States, firms with a
dual class share structure, and firms not in Jay Ritter’s database of firms with a founding date. In addition,
we eliminate all IPO firms that do not have prospectus filings, annual reports, and proxy statement filings
available in the SEC’s EDGAR database. The dependent variable is industry median adjusted IPO firm
Tobin’s Q, winsorized at the 99th percentile. Standard errors are reported below the regression
coefficients.
Panel A. Regressions of Scale Efficiency Ry Tobin’s Q in all years after IPO
Ry
(1)
Delaware (indicator)
-1.246
(0.803)
Firm age (years)
-0.009
(0.016)
Log(assets)
0.132
(0.204)
S&P 500 firm (indicator)
-0.942
(2.380)
e-index
0.320
(0.208)
Time from IPO >4 (indicator)
Ry
(2)
-1.182
(0.804)
-0.012
(0.016)
0.118
(0.204)
-1.218
(2.386)
0.288
(0.209)
1.096*
(0.670)
Classified board x Time from IPO>4
Constant
N
Adj R2
4.426***
(1.218)
15,459
0.00
4.145***
(1.230)
15,459
0.00
Panel B. Regressions of Cost Efficiency Rc Tobin’s Q in all years after IPO
Rc
(1)
Delaware (indicator)
5.641***
(1.654)
Firm age (years)
0.065**
(0.032)
Log(assets)
-6.948***
(0.432)
S&P 500 firm (indicator)
14.850***
(4.896)
e-index
1.234***
(0.431)
Time from IPO >4 (indicator)
Rc
(2)
6.015***
(1.655)
0.048
(0.032)
-7.041***
(0.432)
13.091***
(4.910)
1.056**
(0.432)
5.965***
(1.384)
Classified board x Time from IPO>4
Constant
N
Adj R2
35.936***
(2.563)
13,382
2.02
34.467***
(2.584)
13,382
2.15
73
Ry
(3)
-1.181
(0.804)
-0.012
(0.016)
0.118
(0.204)
-1.268
(2.390)
0.227
(0.264)
0.671
(1.317)
0.155
(0.413)
4.298***
(1.296)
15,459
0.00
Rc
(3)
5.999***
(1.655)
0.048
(0.032)
-7.034***
(0.432)
12.601***
(4.918)
0.468
(0.551)
1.975
(2.701)
1.469*
(0.854)
35.931***
(2.720)
13,382
2.16
Appendix Table 6.
Interaction between various takeover defenses and e-index stability
Our sample consists of 2,285 IPO firms from 1997-2011. We eliminate all REITS, ADRs, funds, firms
without CRSP and COMPUSTAT coverage, firms incorporated outside the United States, firms with a
dual class share structure, and firms not in Jay Ritter’s database of firms with a founding date. In addition,
we eliminate all IPO firms that do not have prospectus filings, annual reports, and proxy statement filings
available in the SEC’s EDGAR database.
Panel A. e-index decreases by firms without (with) a bylaw supermajority requirement
Years from IPO
N
Percent of firms
changing e-index
Percent of firms
changing e-index
Firm with bylaw
amendment restriction
0.41%
0.67%
1.23%
1.83%
2.98%
8.60%
-1.18
-0.79
-0.13
0.14
0.46
0.43
2.14
0.57
1
2
3-4
5-6
7-9
≥10
2,098
1,829
2,919
2,234
2,271
2,104
Firm with no bylaw
amendment restriction
0.15%
0.34%
1.18%
1.91%
3.34%
9.17%
Total
13,455
2.29
t-stat
Panel B. e-index decreases by firms without (with) a charter supermajority requirement
Years from IPO
N
Percent of firms
changing e-index
Percent of firms
changing e-index
Firm with charter
amendment restriction
0.30%
0.50%
0.85%
1.52%
0.27%
7.12%
-0.42
-0.29
1.18
0.87
0.88
1.90*
1.75%
2.75***
1
2
3-4
5-6
7-9
≥10
2,098
1,829
2,919
2,234
2,271
2,104
Firm with no charter
amendment restriction
0.21%
0.41%
1.36%
2.05%
3.43%
9.73%
Total
15,740
2.45%
t-stat
Panel B. e-index decreases by firms without (with) a charter supermajority requirement
Years from IPO
N
Percent of firms
changing e-index
Percent of firms
changing e-index
t-stat
1
2
3-4
5-6
7-9
≥10
2,098
1,829
2,919
2,234
2,271
2,104
Firm with c-index<2
0.20%
0.38%
1.29%
1.94%
3.66%
9.53%
Firm with c-index=2
0.34%
0.56%
0.96%
1.72%
1.97%
7.31%
-0.61
-0.52
0.74
0.36
2.02**
1.53
Total
15,740
2.44%
1.71%
2.75***
74
Panel D. e-index increases by firms without (with) a bylaw supermajority requirement
Years from IPO
N
Percent of firms
changing e-index
Percent of firms
changing e-index
Firm with bylaw
amendment restriction
5.59%
10.93%
19.56%
26.31%
36.44%
46.85%
-0.68
-1.06
-2.20**
-1.34
1.07
0.32
20.83%
1.00
1
2
3-4
5-6
7-9
≥10
2,098
1,829
2,919
2,234
2,271
2,104
Firm with no bylaw
amendment restriction
4.91%
9.39%
16.35%
23.78%
38.72%
47.58%
Total
13,455
20.15%
t-stat
Panel E. e-index increases by firms without (with) a charter supermajority requirement
Years from IPO
N
Percent of firms
changing e-index
Percent of firms
changing e-index
Firm with charter
amendment restriction
5.95%
11.06%
19.42%
25.97%
35.82%
46.02%
-1.11
-1.10
-1.87*
-0.95
1.37
0.76
20.46%
-0.13
1
2
3-4
5-6
7-9
≥10
2,098
1,829
2,919
2,234
2,271
2,104
Firm with no charter
amendment restriction
4.79%
9.42%
16.60%
24.11%
38.84%
47.87%
Total
15,740
20.37%
t-stat
Panel F. e-index increases by firms without (with) a charter supermajority requirement
Years from IPO
N
Percent of firms
changing e-index
Percent of firms
changing e-index
t-stat
1
2
3-4
5-6
7-9
≥10
2,098
1,829
2,919
2,234
2,271
2,104
Firm with c-index<2
4.89%
9.58%
16.52%
24.02%
38.48%
47.60%
Firm with c-index=2
5.82%
10.86%
20.00%
26.41%
36.35%
46.54%
-0.87
-0.84
-2.23**
-1.18
0.93
0.42
Total
15,740
20.33%
20.57%
0.34
75
Marginal Value of Takeover Defenses
Marginal Value of Takeover Defense
0.4
0.2
0
-0.2
0
5
10
15
-0.4
-0.6
-0.8
-1
Years from IPO
Figure 1. This figure shows marginal relationship between the firm value measured by Tobin’s Q
and the number of takeover defenses as the firm goes from the IPO date to 15 years later. Standard error
bars are also reported showing that the point estimates are only statistically different from zero in years 0,
1, and 6-15.
76
Buy and Hold Return of IPO firm Portfolio by eindex
250%
Cumulative Portfolio Return
200%
150%
100%
50%
eindex≥4 return
eindex=3 return
0%
eindex=2 return
-50%
-100%
1997
eindex=1 return
eindex=0 return
1999
2001
2003
2005
2007
2009
2011
Year
Figure 2. This figure shows the buy and hold return of a portfolio generated by purchasing each IPO
firm the first full month after the firm goes public and holding this stock as part of an equally-weighted
portfolio of firms with the same number of takeover defenses for three years. The portfolio is rebalanced
each month and firms that are older than three years are removed from the portfolio while new IPO stocks
are added. Cumulative abnormal returns are calculated starting in February 1997 and held until March,
2012.
77
Data Appendix I
The Lifecycle of Takeover Defenses
Johnson, Karpoff, and Yi
The purpose of this data appendix is to document the technique utilized in collecting the takeover
defenses to generate the e-index (Bebchuk, Cohen, and Ferrell (2009)) measure. We make use of the firm
annual reports (10-K), quarterly reports (10-Q), proxy statements (DEF 14A), and prospectus filings (S-1,
SB-2). We begin collecting data for each IPO firm based on the firm prospectus (S-1 or SB-2) filed at the
time the firm goes through its IPO. We read through the prospectus as well as the attached bylaws and
corporate charter to collect information about the six takeover defenses used to calculate firm e-index:
poison pill, classified board, supermajority requirements to approve mergers, supermajority requirements
to amend bylaws, supermajority requirements to amend the firm charter, and golden parachutes. Once we
have the six takeover defenses necessary to calculate e-index at the IPO stage, we then go through
subsequent security filings to find additional time-series data for each takeover defense. Each takeover
defense tends to be disclosed and discussed in a different manner. Details of this data collection by
takeover defense is below.
Board Declassifications / Classifications
Board classifications and declassifications require shareholder approval at the annual meeting,
making data collection much simpler than poison pill adoptions (which typically occur without
shareholder approval). This means that examining the proxy statements (DEF 14A) of the IPO firms from
the year of their IPO until 2014 is sufficient to determine if the firms take a vote on
classification/declassification at the annual meeting. When the outcome of the vote is uncertain, the proxy
statement for the subsequent year is consulted to determine if the board is classified. The actual date of
declassification/classification used is the file date of the proxy statement.
At times during the data collection, there are situations where a firm will not file a proxy
statement for more than 12 months either because they have moved their fiscal year end or they have not
had an annual meeting with an election for more than 12 months. In addition, at times firms do not file an
annual report for more than 12 months. In these cases, we follow Gompers, Ishii, and Metrick (2003) and
fill in the missing observations using the prior year’s observations for the takeover defenses. In situations
where there is a different takeover defense coding in the year before and the year after the missing data,
78
we examine the annual report and proxy statement in the missing year to ensure accurate coding. There
are no cases where there are missing filings for two consecutive years.
This yields a total of 8 classified board adoptions and 107 classified board removals
(declassifications) over the 18 year time period of our sample.
Poison Pill Adoptions and Removals
To collect the time series of poison pill adoptions and removals, we search each of 12,491 annual
report filings (10-K, 10-K405, 10-KSB, and 10-KSB40) to find the following words and phrases: poison
pill; adopt a poison pill; remove our poison pill; shareholder rights; shareholder rights plan; remove our
shareholder rights plan; remove the shareholder rights plan; stockholder rights; stockholder rights plan;
remove our stockholder rights plan; remove the stockholder rights plan; rights plan; remove our rights
plan; adopt a rights plan; adopt a stockholder rights plan; adopt a shareholder rights plan; adopt a
stockholder rights; adopt a shareholder rights. When we find at least six of these terms in the annual
report, we read through the appropriate section of the annual report to determine if the annual report said
that there was a poison pill in place, a poison pill which was adopted that year, or a poison pill removed.
We also tracked poison pills that expired after their (typical ten year) pre-specified life at adoption.
Since many firms only mention the term poison pill once in their annual reports, we also conduct
a search looking only for the term “rights plan.” This term will effectively capture all of the variants of
the terms shareholder rights plan, stockholder rights plan, and stock appreciation rights plan. For each
firm in our sample, we begin with the most recent filing of the firm starting in 2013. We then examine
each filing at a seven year interval going back to the time of the firm’s IPO, ensuring that there is no
seven year time period over which we do not check at least one filing. Because the mean and median time
a poison pill is ten years, we capture all the poison pills in our sample using this technique. When we find
a firm with a poison pill mentioned in its annual report, we look to see if the annual report explicitly
mentions the date of adoption and expiration or removal. If present, we record these dates. If missing, we
move through the annual reports for the firm year-by-year searching for the date the rights plan is first
announced and the date the rights plan is last mentioned and records these years as being the first and last
year the poison pill is in force.
79
We classify as a poison pill any shareholder rights plan or stockholder rights plan that would
reduce the likelihood of takeover by outside parties. Stock appreciation rights plans, Employee stock
appreciation plans, share appreciation rights plans, or other plans meant for manager and employee
compensation were not considered as poison pills unless the annual report explicitly stated that these
plans could reduce the likelihood of the firm being acquired by an outside party. For instance one firm in
our sample used the following language in its annual report:
“In 2002, Asset Acceptance Holdings LLC adopted a share appreciation rights plan for certain
key employees. The purpose of the plan is to further the long-term stability and financial success
of the Company, as participants in the plan have the potential to share in the appreciation of the
value of Asset Acceptance Holdings LLC. A benefit may be earned by participants if certain
financial objectives are met upon partial or complete liquidation events, as defined in the plan.”
Such a firm is not considered to have a shareholder rights plan since the purpose of this plan is to
motivate employees to meet “certain financial objectives.”
In addition, the board’s ability to issue blank check preferred stock without shareholder approval
is often cited as giving the board the ability to adopt a poison pill at will. Many firms explicitly mention
that they have a blank check preferred stock that could be used to create a poison pill. For instance, Phase
Forward in their 2010 annual report states that the board has the ability “…to designate the terms of an
issue new series of preferred stock without stockholder approval, which could be used to institute a
stockholders rights plan, or a poison pill.” Consistent with the prior literature, blank check preferred stock
provisions were not considered as poison pills.
To determine the actual dates of poison pill adoption, we first looked for a date that the board
voted to adopt the pill. Approximately 80% of poison pill adoptions report a date of board approval.
Absent this date, we utilized the date when shareholder rights were distributed to shareholders of record.
If the annual report did not disclose this date, we utilized the date of the public announcement of the
poison pill as reported in the 8-K. In cases where the adoption year was known, but not the month and
day, (<5% of cases), we utilized the expiration month and day (ten years later) as the adoption date. In
cases where only the year and month are known, we are still able to utilize the month to determine the
fiscal year in which the poison pill was adopted. In cases where only the year of the poison pill is reported
by the annual report, we assume that the poison pill was adopted in the fiscal year of the annual report
when it is first reported. At some times, the annual report makes no mention of a poison pill in one year,
80
but explicitly mentions a poison pill in the subsequent year, without giving information about when the
poison pill was adopted. In this case, we assume the poison pill was adopted in the fiscal year where it is
first mentioned in the annual report. These cases account for only approximately 3% of our poison pill
adoptions.
This yields a total of 334 poison pill adoptions and 75 poison pill removals over the 18 year time
period of our sample.
Restrictions on amending the bylaws, amending the charter, or supermajority requirements for
mergers
To gather this information, we follow Cremers and Ferrell (2013) who utilize charters from the
10-K and 10-Q filings of the firms. We are able to get the charters as well as the bylaws of our sample of
firms since they are all filed with the prospectus. We look for changes in takeover defenses in the
subsequently filed annual report, quarterly report, and press release (8-K filings). We search each 8-K
looking specifically at 8-Ks with an item 5.03 which is a change to the firm bylaws or charter.
We find a total of 34 firms which remove their bylaw restrictions and 27 firms which remove
their charter restriction over the 18 years in our sample. We find three firms which remove bylaw
restrictions and 2 firms which remove charter restrictions in our sample.
Golden Parachutes
To determine if a firm has a golden parachute or a severance package conditional on a change of
control for the firm, we examine the annual reports of the firms and the addenda related to CEO contracts.
Many detailed CEO contracts are reported as addenda of the 10-K filing. We search through the CEO
contract for the following terms: golden parachute, severance package, CEO severance, severance, change
of control, and change in control. If the CEO receives any compensation, lump sum, or vesting of options
in response to a change in control, then the firm is classified as having a golden parachute. If it is unclear
if the firm has a golden parachute, we refer to the firm’s proxy statement to determine if the firm has a
golden parachute.
81
An example of a golden parachute change to a firm is as follows. On January 9, 1997, Sun
Hydraulics went public without a golden parachute. However, the firm decided in 2009 to adopt a golden
parachute to ensure the continuity of the management of the firm in case of a takeover. The firm writes in
its proxy statement, “In the event of executive termination following a change in control, the executive is
entitled to a lump sum payment equal to twice the amount of his or her annual salary.” The explanation
for the golden parachute adoption was listed as follows in the firm’s proxy statement.
“In December 2009, the Board of Directors approved, and the Company entered into an
Executive Continuity Agreement (the “Agreement”) with each of Allen J. Carlson, President and
CEO, and Tricia Fulton, CFO Officer, respectively. The intent of the Agreement is to assure the
Company and the executive of continuity of management in the event of any actual or threatened
change in control of the Company, by providing for specific benefits to such executives in the
event of the termination of their employment with the Company following a change in control.”
https://www.sec.gov/Archives/edgar/data/1024795/000102479514000012/definitiveproxy.htm
Data Reliability and Error Rates
To supplement our data, we consult the ISS (formerly IRRC) data utilized by Bebchuk et al
(2009) and Gompers Ishii and Metrick (2003). We find that 1,887 of our IPO firm year observations are
also in the samples of either Bebchuk et al (2009) or Gompers, Ishii, and Metrick (2003). We therefore
cross-check our classifications of the firms as having classified boards, poison pills, golden parachutes,
supermajority requirements for mergers, changing bylaws, or changing the corporate charter with the data
from ISS. This allows us to create a robust dataset of clean data and to compare our coding error rates
with those in the ISS dataset.
Check against ISS Data
One difficulty of our dataset which has similar collection procedures as those used by Cremers
and Ferrell (2014) and Cremers, Litov, and Sepe (2015) is that we must collect our data using a
combination of electronic and hand collection. To ensure the accuracy of our data, we rely on the fact that
82
our data has 1,887 overlapping observations (11.96% of our total of 15,775 observations) with the ISS
data more commonly used in the literature. We find that for each of our six takeover provisions the
correlations between our hand-collected data is always above .90. For instance, we find that the classified
board measure we hand collect has a correlation of 0.95 with the ISS data.
Below is a table comparing the coding error rates in the ISS data with our data. When we find a
discrepancy between the ISS data and our data, we go to the original source documents on the SEC’s
EDGAR web page to determine whether our data or the ISS data is correct.
Takeover Defense
ISS
Johnson, Karpoff, and Yi
data error rate
data error rate
Classified board
0.37%
0.53%
Poison Pill
1.32%
1.80%
Bylaw Restriction
22.89%
7.37%
Charter Restriction
27.34%
9.49%
Supermajority to Approve Mergers
36.40%
6.95%
Golden Parachute
20.46%
9.86%
For classified boards, where we can easily find an annual observation for each firm using the
proxy statement, we then go back to the original data sources (proxies and 10-Ks) to try to determine why
these discrepancies exist.21 We find that of the cases where there are discrepancies between the hand
collected data and the ISS data, the vast majority involve a difference in the years when a declassification
occurs. This is largely due to the fact that the ISS data is not collected every year. In addition, we find that
Unlike the classified board, the takeover defenses explicitly located in the firm’s charter are not observable on an
annual basis, but rather only updates to the charter are observable. As such, we do not confirm the accuracy of all of
the takeover defenses on an annual basis.
21
83
there are N=7 clear errors in the coding in the ISS data when compared to the original documents filed by
the firm at the SEC. In addition, we find N=10 cases where there are clear coding errors in our dataset.
This yields a likely error rate of 0.5% (10/1887) for our dataset and 0.3% for the ISS data for the
classified board characterization. After finding these coding errors in our dataset and calculating our rate
of error, we correct the errors in our coding to ensure an improved dataset.
For poison pills, we repeat the procedure merging the N=1,887 firms in our sample with the ISS
takeover provision data. We find that there are N=25 unique firm coding errors in the ISS data amounting
to a 1.32% (25/1887) error rate. In addition, we find that there are N=34 unique firm coding errors in our
data amounting to a 1.80% error rate (34/1887). In addition, there were three cases in ISS and three cases
in our dataset where the adoption or removal of the poison pill was off by one year, due to a fiscal year
end problem. After calculating these error rates, we corrected the problems in our data.
In addition, for the N=1,887 firm years in our sample that also have ISS data, we compare our
coding results for bylaw amendment restrictions and charter amendment restrictions to those of ISS. We
find that they are positively correlated, but only 60.10% of the Bylaw restrictions are coded identically
and 61.92% of the charter restrictions are coded identically. This implies that the data quality is much
lower for bylaw and charter restrictions compared to the data quality for classified boards and poison
pills. When there is a discrepancy between our coding and the ISS coding, we go back to the original
documents filed with the SEC using the EDGAR database. For instance, in the year of discrepancy we go
to the 10-K in that year and examine the filing to determine if the filing explicitly states that there either is
or is not a restriction of bylaw and charter amendments. For instance, our data codes Niku Corp, which
went through its IPO on February 28, 2000 as having restrictions on both amending its bylaw and charter
for all five years the firm is in our sample (2000-2004). The ISS data also has this firm in its data in 2002,
coding the firm as having a bylaw amendment restriction in 2002, but no charter amendment restriction.
We therefore go to the 10-K for 2002
(http://www.sec.gov/Archives/edgar/data/1076641/000089161803001850/f88965e10vk.htm) and find the
following statement in the annual report (emphasis added).
“Provisions of Delaware law, our certificate of incorporation and bylaws could have the effect of
delaying or preventing a third party from acquiring us, even if a change in control would be
beneficial to our stockholders. These provisions include: …requiring two-thirds of the
84
outstanding shares to approve amendments to some provisions of our certificate of
incorporation and bylaws;”
Because the firm requires a two-thirds majority to amend both its certificate of incorporation and
its bylaws, it is clear that the firm should be coded as having a bylaw restriction and a charter restriction.
Therefore, we note that our data collection is correct and the ISS data is incorrect. In cases where the 10K does not give an explicit clarification about the charter and bylaws, the bylaws and charter are
referenced from a prior filing with the SEC, usually the firm’s prospectus at the IPO (S-1) or a prior press
release (8-K) or quarterly filing (10-Q) or annual report (10-K). We then go back to this prior filing to
determine if our data is correct or if the ISS data is correct.
Overall, we find that our results are incorrectly coded 7.37% of the time for bylaw restrictions
and 9.49% for charter restrictions.22 In contrast, the ISS data is incorrectly coded 22.89% of the time for
bylaw restrictions and 27.34% of the time for charter restrictions.23 We also find some substantial
apparent shifts in ISS coding around 2006 and 2010 which may be a strong driver of the ISS errors
documented.
In addition, we compare our data coding to that in the ISS data for supermajority requirements to
approve mergers. Many of these requirements are explicitly mandated by the state, however, some firms
opt out of their state requirements for supermajorities to approve mergers. We find that the ISS data is
incorrectly coded for supermajority merger requirements 36.40% of the time where our data error rate is
6.95%.24
Finally, we compare our data coding to the ISS data for golden parachutes and find the following
results. For the 1,887 firms with observations in our dataset and ISS, we find that the ISS data is coded
incorrectly 20.46% of the time and our data is incorrectly coded 9.86% of the time.25
22
We find that the bylaw amendment restriction categorizations in our dataset have a Type I error rate of 3.23% and
a Type II error rate of 4.13%. The charter amendment restriction categorizations in our dataset have a Type I error of
3.6% and a Type II error rate of 5.88%.
23
We find that the bylaw amendment restriction categorizations in the ISS data has a Type I error rate of 15.90%
and a Type II error rate of 7.00%. The charter amendment restriction categorizations in the ISS data has a Type I
error of 19.61% and a Type II error rate of 7.74%.
24
The supermajority requirement for mergers data error rates can be coded as follows. The ISS data has a Type I
error rate of 6.19% and a Type II error rate of 30.21%. The Type I and Type II error rates for supermajority
requirements for mergers in our dataset are 1.05% and 6.20%, respectively.
25
The golden parachute data has the following error rates. The ISS data has a Type I error rate of 0.95% and a Type
II error rate of 19.50%. Our golden parachute data has a Type I error rate of 3.55% and a Type II error rate of
6.31%.
85
Based on the above results, we feel comfortable stating that our overall error rates are well below
the 10% level which would be considered adequate for examining most empirical questions in the data
and our error rates are at worst half those from the prior literature.
86